Journal of Baltic Science Education, Vol. 18, No. 3, 2019

Vol.18, No.
3, 2019
ISSN 1648-3898 /Print/
ISSN 2538–7138 /Online/
The International Journal of the Scientia Socialis Ltd., & SMC “Scientia Educologica”
Journal of Baltic Science Education, Vol. 18, No. 3, 2019
Editorial Board
ISSN 1648–3898 /Print/
Editor-in-Chief ISSN 2538–7138 /Online/
Prof., Dr. Vincentas Lamanauskas Siauliai University, Lithuania
Deputy Editor-in-Chief
Prof., Dr. Andris Broks University of Latvia, Latvia
Deputy Editor-in-Chief
Prof., Dr. Miia Rannikmäe University of Tartu, Estonia
Executive Secretary
Dr. Laima Railienė SMC “Scientia Educologica”, Lithuania
Editors:
Prof., Dr. Boris Aberšek University of Maribor, Slovenia
Prof., Dr. Agnaldo Arroio University of Sao Paulo, Brazil
Prof., Dr. Martin Bilek Charles University, Czech Republic
Dr. Paolo Bussotti University of Udine, Italy
Prof., Dr. Hana Čtrnáctová Charles University, Czech Republic
Dr. Peter Demkanin Comenius University, Slovakia
Dr. André du Plessis Nelson Mandela University, South Africa
Prof., Dr. Jack Holbrook University of Tartu, Estonia
Prof., Dr. Ryszard M. Janiuk Maria Curie Sklodowska University, Poland
Dr. Milan Kubiatko University of Zilina, Slovakia
Dr. Todar Lakhvich Belarusian State Medical University, Republic of Belarus
Prof., Dr. Jari Lavonen University of Helsinki, Finland
Dr. Rita Makarskaitė-Petkevičienė Vilnius University, Lithuania
Prof., Dr. Aadu Ott Göteborg University, Sweden
Prof., Dr. Paul Pace Malta University, Malta
Prof., Dr. Valfrids Paškevičs Daugavpils University, Latvia
Prof., Dr. Jongwon Park Chonnam National University, Korea
Prof., Dr. Raffaele Pisano Lille 3 University, France
Prof., Dr. Pavol Prokop Trnava University, Slovakia
Prof., Dr. Konstantinos Ravanis University of Patras, Greece
Dr. Alona Rauckienė-Michaelsson Klaipėda University, Lithuania
Dr. Dušica Rodić University of Novi Sad, Serbia
Prof., Dr. Heimo Saarikko Helsinki University, Finland
Dr. Hae-Ae Seo Pusan National University, Korea
Dr. Uladzimir K. Slabin University of Oregon, USA
Prof., Dr. Borislav V. Toshev Sofia University, Bulgaria
Dr. Georgios Tsaparlis University of Ioannina, Greece
Dr. Muhammet Usak Science and Medical Education Research Center, Turkey
A scientific journal JBSE issued by the Scientia Socialis Ltd., in cooperation with SMC Scientia Educologica, Lithuania,
emphasizes theoretical, experimental and methodical studies in the field of science education. JBSE is an international
academic journal. In order to maintain the high standards appropriate to such a journal, all contributions received are
submitted for anonymous review by two experts, additionally to review by the Editor. The decision of the Editor on
the acceptance of articles is final and no correspondence can be entered into on reasons for rejection of a submitted
contribution.
Published since 2002 Address:

The journal is published bimonthly. Scientia Socialis, Ltd.
IF: 0.638 (2017) Donelaičio Street 29, LT-78115 Siauliai, Lithuania
H Index: 10 (2016) E-mail: mail.jbse@gmail.com
SJR: 0.326 (2017) Phone: +370 687 95668
ICDS: 10.7 (2017) Home page: http://www.scientiasocialis.lt/jbse/
ICV: 143.26 (2015)
ISSN 1648–3898 (Print) © Scientia Socialis Ltd., Lithuania, 2019
ISSN 2538–7138 (Online)

The articles appearing in this journal are indexed/abstracted in British Education Index
(http://www.leeds.ac.uk/bei/bei.htm), Copernicus Index (http://journals.indexcopernicus.com), The Asian
Education Index (http://www.asian-education-index.com/sciences_index.php), EBSCO: Academic Search Premier
318 (http://search.ebscohost.com), Social Scisearch (Thomson Reuters) (http://science.thomsonreuters.com/index.html),
Journal Citation Reports / Social Sciences Edition (Thomson Reuters)
(http://thomsonreuters.com/products_services/scientific/Journal_Citation_Reports),
and SCOPUS (http://www.scopus.com)
ISSN 1648–3898 /Print/ CONTENTS
CONTENTS
Editorial
MENTORING ON SCIENCE TEACHER EDUCATION IN RUSSIA AND INTERNATIONAL PERSPECTIVES

Muhammet Usak, Alfiya R. Masalimova. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
Articles
THE EFFECT OF EARTHCOMM LEARNING MODEL AND SPATIAL THINKING ABILITY ON

GEOGRAPHY LEARNING OUTCOMES
Muhammad Aliman, Budijanto, Sumarmi, I Komang Astina, Rozana Eka Putri, Muhammad Arif . . . . . 323
THE EFFECT OF COMPUTER-AIDED 3D MODELING ACTIVITIES ON PRE-SERVICE TEACHERS’

SPATIAL ABILITIES AND ATTITUDES TOWARDS 3D MODELING
Ali Ihsan Benzer, Bunyamin Yildiz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
FACTORS INFLUENCING THE FIRST YEAR STUDENTS’ ADAPTATION TO NATURAL

SCIENCE STUDIES IN HIGHER EDUCATION
Rita Birzina, Dagnija Cedere, Liva Petersone. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
THE EFFECT OF THE USE OF AUGMENTED REALITY APPLICATIONS ON THE ACADEMIC SUCCESS
AND MOTIVATION OF 7TH GRADE STUDENTS
Esma Buluş Kirikkaya, Melek Şentürk Başgül . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
RESEARCH TRENDS AND ISSUES OF EDUCATION FOR SUSTAINABLE DEVELOPMENT-RELATED

RESEARCH IN SOUTH KOREA
Hyuksoo Kwon, Eunsang Lee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
LEARNING ENVIRONMENT, STUDENTS’ BELIEFS, AND SELF-REGULATION IN LEARNING PHYSICS:

STRUCTURAL EQUATION MODELING
Maison, Syahrial, Syamsurizal, Tanti . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389
A MODEL OF INTEREST IN STEM CAREERS AMONG SECONDARY SCHOOL STUDENTS

Lilia Ellany Mohtar, Lilia Halim, Norshariani Abd Rahman, Siti Mistima Maat,
Zanaton H. Iksan, Kamisah Osman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
THE RELATION OF GENDER AND TRACK ON HIGH SCHOOL STUDENTS’ ATTITUDE TOWARD
CONVERGENCE
Yustika Sya’bandari, Minsu Ha, Jun-Ki Lee, Sein Shin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
MONODISCIPLINARITY IN SCIENCE VERSUS TRANSDISCIPLINARITY IN STEM EDUCATION

Mitja Slavinec, Boris Aberšek, Dino Gačević, Andrej Flogie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
THE EFFECT OF USERS’ BEHAVIORAL INTENTION ON GAMIFICATION AUGMENTED REALITY IN

STEM (GAR-STEM) EDUCATION
Chung-Ho Su . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
EXPLORING CHANGES IN PRIMARY STUDENTS’ ATTITUDES TOWARDS SCIENCE, TECHNOLOGY,

ENGINEERING AND MATHEMATICS (STEM) ACROSS GENDERS AND GRADE LEVELS
Shao-Na Zhou, Hui Zeng, Shao-Rui Xu, Lu-Chang Chen, Hua Xiao . . . . . . . . . . . . . . . . . . . . . . . . 466
Information
INFORMATION FOR CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481
PROBLEMS OF PSYCHOLOGY IN THE 21ST CENTURY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482
NATURAL SCIENCE EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
319
MENTORING ON SCIENCE TEACHER EDUCATION IN RUSSIA AND INTERNATIONAL

PERSPECTIVES
Muhammet Usak, Alfiya R. Masalimova

Kazan Federal University, Russia
Nowadays, mentoring as an educational form of Russia’s occupational activity is undergoing fundamental

changes due to the variety of interrelated factors affecting the mentoring system. The requirements for the mentor-
ing system itself are affected by these changes as well. Nature of the mentoring activity influences socio-economic,
political, socio-cultural, pedagogical, methodical and individual factors. Socio-economic factors that determine not
only the enterprises’ state order, but also modern educational institutions’ order to revive and improve the traditions
of mentoring receive the highest priority among these factors (Antipin, 2011).
Mentoring is one of the forms of training in the process of occupational activity of a novice teacher, aimed at the
development of his/her applied occupational competencies and the disclosure of their potential to determine the
vectors of teachers’ individual professional development (Charina, 2004; Chernikova, 2013; Masalimova, 2013). The
effectiveness of the mentorship stems from the fact that the external training of teachers (refresher courses, intern-
ships) does not take into account the characteristics of a specific educational institution. Therefore, any educational
institution should prepare mentors within their teaching staff, who will train young novice teachers taking into account
the established traditions and values of this institution.
The mentoring process affects the interests of at least three interactive entities: a new teacher, a mentor/mentor
teacher and an educational institution (du Plessis, 2016; Hudson, Usak, & Savran-Gencer, 2010; Jung & Nam, 2018).
The novice teacher acquires professional knowledge, skills and abilities, improves his/her professional skills and abili-
ties; builds a professional career; learns to construct relationships with mentors, which will then help to do it with the
environment of the institution in which he/she will adapt; receives data about the traditions, values and norms of
the educational institution, in which he/she carries his/her occupational activities. As far as the mentor is concerned,
he also improves his professional and commercial qualifications and pedagogical skills in the mentoring process.
Likewise, the educational institution increases the level of cultural and vocational education of its staff; strengthens
relationships between employees; ensures continuity of the mentoring corps.
Modern mentorship in the Russian concept is based on the following regulatory norms or principles:
- the principle of occupational and mentoring activities’ integration not only allows mentors to optimize
the process of acquisition of occupational competencies, but also the interprofessional competencies
that contribute to the transfer of occupational experience to young staff and the accumulation of the
organization’s collective knowledge.
- the principle of tradition-based nature and innovation, which implies, on the one hand, reliance on already
established traditional ideas, concepts and mentoring experience, and on the other – constant updating
of its content and technologies in accordance with the latest developments in science and technology.
The choice of this principle stems from the fact that mentors’ adherence to traditional teaching meth-
320
https://doi.org/10.33225/jbse/19.18.320
ods only, which are usually reproductive in nature, do not direct young teachers to the formulation and
resolution of creative tasks under the updated production conditions. Such a lack of mentors’ training to
innovative changes in the content and technologies of young personnel’s training hinders the process of
their training and adaptation to occupational activities, which adversely affects the strategic objectives
of modern educational institution.
- the principle of outstripping training, involving advanced entry into mentoring activities, namely the for-
mation of positive motivation, value attitude to mentoring, psychological, pedagogical and methodical
competencies of the modern mentor.
The main mechanism for the implementation of the proposed principles of mentoring is an integrative ap-
proach, involving the integration and interaction of 1) two processes – the process of occupational development
of teachers and their readiness to implement mentoring activities; 2) two types of activities – occupational and
mentoring; 3) all entities of the educational institution in the process of mentoring.
When it comes to international practice in the field of mentoring, it should be noted that there are different
perspectives on the concept of mentoring. This is due to two different theoretical concepts: North American and
European one.
The interpretation of the North American concept sees the mentor as a senior and more authoritative person
expecting dedication in response to his/her helpful advice, wise guidance and help (Roberts, 2000; Zachary, 2005,
2009). Within this personification, a mentor can be attributed as a direct supervisor or as a person’s department
manager, and these are relationships in which the term protégé can be applied to identify them. In this case, a
mentor is more interested in promoting his/her mentee than providing effective leadership in learning. Thus,
mentoring seems to have essential attributes: a process; a supportive relationship; assistance process; a teaching-
learning process; a reflective process; a career development process; a formalized process; and a role constructed
by mentor or for a mentor. Coaching, sponsorship, role modeling, evaluation and an informal process appear to
be conditional features of mentoring as a phenomenon (Roberts, 2000).
On the other hand, the interpretation of the European mentoring suggests that a mentor has a lot of practical
experience rather than ability to promote the mentee. One of the characteristics of an effective mentoring rela-
tionship is an open collar dialogue, i.e. avoiding significant differences in the relationship so that both parties are
on an equal footing (Cain, 2009; Colvin & Ashman, 2010; Peterson et al., 2010). As a result, European mentors are
usually independent in their decisions, mainly because it is very difficult for anyone to open up to someone who
can influence their salary, status and overall well-being. The main purpose of business relations is education and
development, although the outcomes may be closely associated with the development of the mentee’s ability to
manage his/her career.
Compared to national experience, the mentoring activity in international corporate training practice is
closely related to the use of technologies intended to transfer professional experience to novice teachers, to
develop their professional qualifications, as well as to identify and develop their potential. In the international
mentoring practice, there may be categories such as mentoring (targeted experience transfer), coaching (ex-
plaining the potential of the student’s personality), e-coaching (online coaching), tutoring (accompanying the
learning process, discussion of the experience of knowledge transfer into real practice), shadowing (monitoring
of the work process), supervision (monitoring of the occupational activities of the novice worker and evaluation
of the level of formation of competences) (Hudson, Usak, & Savran-Gencer, 2010; Schmidt, 2008; Tillema, Smith,
& Leshem, 2011; Yavuz, 2011).
Among the most common forms and methods used in mentoring activities abroad, we can distinguish:
budding (involving the student in another’s activity), business simulations (corporate management simulation),
storytelling (telling of stories), etc. (Masalimova, Usak, & Shaidullina, 2016; Masalimova & Shaidullina, 2016).
Under the conditions of the modern Russian mentoring system, this form of vocational training is strictly
integrated with the international technologies offered, taking into account the existing Russian mentoring tradi-
tions, the realities and requirements of modern educational institutions.
Acknowledgements
The work is conducted according to the Russian Government Program of Competitive Growth of Kazan
Federal University.
321
References
Antipin, S. G. (2011). Traditions of mentoring in the history of national education (PhD Thesis). Nizhny Novgorod.
Cain, T. (2009). Mentoring trainee teachers: How can mentors use research? Mentoring & Tutoring Partnership in Learning, 17(1),
53-66. https://doi.org/10.1080/13611260802233498.
Charina, Ye. V. (2004). Relations in the system “mentor-young specialist” in the process of professionalization (PhD Thesis). Moscow.
Chernikova, Ye. A. (2013). Mentoring as a means of accompanying the professional-personal development of a social service specialist
(PhD Thesis). Volgograd.
Colvin, J. W., & Ashman, M. (2010). Roles, risks, and benefits of peer mentoring relationships in higher education. Mentoring &
Tutoring Partnership in Learning, 18(2), 121-134. https://doi.org/10.1080/13611261003678879.
du Plessis, A. (2016). Student-teachers’ pedagogical beliefs: Learner-centred or teacher-centred when using ICT in the science
classroom? Journal of Baltic Science Education, 15(2), 140-158.
Hudson, P., Usak, M., & Savran-Gencer, A. (2009). Employing the five-factor mentoring instrument: Analysing mentoring practices
for teaching primary science. European Journal of Teacher Education, 32(1), 63-74.
Hudson, P., Usak, M., & Savran-Gencer, A. (2010). Benchmarking Mentoring Practices: A Case Study in Turkey. Eurasia Journal of
Mathematics, Science & Technology Education, 6(4), 245-252. https://doi.org/10.12973/ejmste/75245
Jung, D., & Nam, J. (2018). Analysis of mentor teachers’ mentoring type and type changes in collaborative mentoring programs for
professional development of beginning science teachers’ teaching practice. Journal of the Korean Chemical Society-Daehan
Hwahak Hoe Jee, 62(6), 441-452. HTTPS://DOI.ORG/10.5012/JKCS.2018.62.6.441.
Masalimova, A. R. (2013). Corporate training for mentors: Priority approaches. World of Science, Culture, Education, 4(41), 96-99.
Masalimova, A. R., & Shaidullina, A. R. (2016). Study of international mentoring and coaching practices and their constructive
application in the Russian system of corporate education and training. International Journal of Environmental and Science
Education, 11(8), 1797-1806.
Masalimova, A. R., Usak, М., & Shaidullina, A. R. (2016). Advantages and disadvantages of national and international corporate
training techniques in adult education. Current Science, 111(9), 1480-1485. https://doi.org/10.18520/CS/V111/I9/1480-1485.
Peterson, S. M., Valk, C., Baker, A. C., Brugger L., & Hightower, A. D. (2010). “We’re not just interested in the work”: Social and
emotional aspects of early educator mentoring relationships. Mentoring & Tutoring Partnership in Learning, 18(2), 155-175.
https://doi.org/10.1080/13611261003678895
Roberts, A. (2000). Mentoring revisited a phenomenological reading of the literature. Mentoring & Tutoring Partnership in Learn-
ing, 8(2), 145-170. https://doi.org/10.1080/713685524.
Schmidt, M. (2008). Mentoring and being mentored: The story of a novice music teacher’s success. Teaching and Teacher Educa-
tion, 24(3), 635-648. https://doi.org/10.1016/j.tate.2006.11.015.
Tillema, H. H., Smith, K., & Leshem, S. (2011). Dual roles - conflicting purposes: A comparative study on perceptions on assess-
ment in mentoring relations during practicum. European Journal of Teacher Education, 34(2), 139-159. https://doi.org/10.
1080/02619768.2010.543672.
Yavuz, A. (2011). The problematic context of mentoring: Evidence from an English language-teaching department at a Turkish
University. European Journal of Teacher Education, 34(1), 43-59. https://doi.org/10.1080/02619768.2010.516431.
Zachary, L. J. (2005). Creating a mentoring culture: The organization’s guide. San Francisco, CA: Jossey-Bass.
Zachary, L. J. (2009). The mentee’s guide: Making mentoring work for you. San Francisco, CA: Jossey-Bass.
Received: March 20, 2019 Accepted: May 22, 2019
Muhammet Usak PhD, Associate Professor, Institute of Pedagogy and Psychology,

Kazan Federal University, 1 Mezhlauk St., Kazan, 420021, Russia.
E-mail: musaktr@gmail.com
Website: https://orcid.org/0000-0002-6537-9993
Alfiya R. Masalimova PhD, Associate Professor, Institute of Pedagogy and Psychology,
Kazan Federal University, 1 Mezhlauk St., Kazan, 420021, Russia.
Website: https://www.researchgate.net/profile/Alfiya_
Masalimova2
322
THE EFFECT OF EARTHCOMM
LEARNING MODEL AND
SPATIAL THINKING ABILITY ISSN 1648-3898 /Print/
ON GEOGRAPHY LEARNING ISSN 2538-7138 /Online/
OUTCOMES
Abstract. Geography learning is

meaningless if it is not followed by
Muhammad Aliman, the development of students’ spatial
Budijanto, thinking ability. Research aimed to 1)
Sumarmi, determine the effect of earthcomm
learning and conventional learning on
I Komang Astina, geography learning outcomes, 2) find
Rozana Eka Putri, out the effect of spatial thinking ability
Muhammad Arif on geography learning outcomes, 3)
find out the interaction between the
application of earthcomm learning
and spatial thinking ability on geogra-
phy learning outcomes. Research used
Introduction a quasi-experimental non-equivalent
control group design. The research
Geography is a school subject that can develop character and skills in sample was taken from three high
the face of global competition and the industrial revolution 4.0. Skills in ge- schools in Malang City, Indonesia. Ge-
ography learning teach students to have geography literacy and map literacy ography learning outcomes data were
skills (Sugiyanto, Maryani, & Ruhimat, 2018; Wiwik Sri Utami, Sumarmi, Ruja, obtained from paper-based test and
& Utaya, 2016a; Wiwik Sri Utami, Zain, & Sumarmi, 2018) and spatial intelli- spatial thinking ability data obtained
gence (Kim & Bednarz, 2013; Metoyer & Bednarz, 2017; Mulyadi, Yani, Ismail, from spatial thinking test instruments.
Data were analyzed using Two Way
& Rosita, 2018; Yani, Mulyadi, & Ruhimat, 2018). However, in the reality, it was
ANOVA test using SPSS devices. The
found that learning outcomes, skills and geography literacy of high school
results of the research, 1) there is an
students are low (Homoki & Sütő, 2013; Utami et al., 2018), including low effect of the application of learning
spatial ability (Metoyer & Bednarz, 2017; Sudatha, Degeng, & Kamdi, 2018), to students’ geography learning out-
spatial intelligence (Mulyadi et al., 2018), and low spatial thinking (Fleming comes, 2) there is an influence of the
& Mitchell, 2017; Yusup, Santoso, & Istifarida, 2018). Efforts are needed to im- ability of spatial thinking on geogra-
prove students’ spatial thinking ability through effective geography learning. phy learning outcomes, 3) there is no
Geography learning can be done flexibly. Geography learning outside interaction between the application
the classroom is one of effective learning and has a positive influence on of learning and spatial thinking on
geography learning outcomes (Railienė, 2003; Yani et al., 2018). This learning geography learning outcomes. Efforts
involves students with the environment that requires teachers to know and to improve learning outcomes and
spatial thinking ability are needed
understand environmental problems (Jeronen, 2004). Geography lessons can
by students to face the challenges of
develop the ability in the field compared to other subjects such as physics
global competition.
and chemistry (Soobard & Rannikmäe, 2013). A learning is considered good Keywords: Earthcomm learning
if it is able to motivate students in improving the final grade of students model, geography learning, geogra-
(Bahri & Corebima, 2015). One of the learning that develops students’ skills phy learning outcome, quasi experi-
and creativity in scientific literacy is earthcomm learning. This learning was mental, spatial thinking ability.
originally developed for chemistry subject in 1988 in America, for biology
subject in 1996 and physics subject in 1998. It had just been developed Muhammad Aliman,
Budijanto, Sumarmi, I Komang Astina
for geoscience subject including geography in 2001. Earthcomm learning State University of Malang, Indonesia
(Earth science in the community) is learning that involves students inside Rozana Eka Putri
and outside the classroom by emphasizing the process of inquiry (Park, STKIP PGRI Sumatera Barat, Indonesia
Muhammad Arif
Yager, & Smith, 2005; Park, Park, & Lee, 2009) Some studies investigating STKIP Pesisir Selatan, Indonesia
earthcomm learning include: earthcomm learning towards creative thinking
323
(P. 323-334) ISSN 2538–7138 /Online/
(Suharto, 2016), earthcomm towards critical thinking (Dewi, 2014), earthcomm towards 21st century capability
(Mari, Sumarmi, & Handoyo, 2017), earthcomm based electronic modules (Mauliddia, Muryani, & Rintayati, 2018),
earthcomm-based multimedia learning on motivation to learn (Hidayat, Sarwono, & Yusup, 2017), earthcomm in
developing teacher professionalism (DY Park et al., 2005). However, earthcomm learning research in improving the
quality of geography learning and student learning outcomes has not been widely studied.
Earthcomm learning is learning that develops students’ literacy skills and scientific investigations at all
levels of education. Scientific investigations can help students understand the earth’s system contextually (Park
& Park, 2013; Sumarmi, 2015). The stage of earthcomm that forms scientific inquiry is Challenge: giving stimulus
to students in order to motivate them in conducting investigations, Think about it: helping students to think in
planning investigations, Investigate: students do an investigation followed by Digging deeper: analyzing the
results of investigations in depth by collaborating other sources, Inquiry further: the teacher can invite students
to conduct further investigations. Investigations conducted require students to be creative and able to solve daily
problems. Investigations must be contextual with their lives in order to form students’ scientific reasoning, (Park
et al., 2005). Contextual inquiry requires the ability to think in knowing the real conditions. This cognitive ability
consists of recognizing space, applying space recognition technology, and using several reasons in determining
space (Metoyer & Bednarz, 2017).
Cognitive ability to recognize space and its phenomenon is the ability to think spatially. This ability can help
students demonstrate real space into dimensional space or vice versa (Metoyer & Bednarz, 2017). In developing
the knowledge, skills and practices of geography, students need spatial thinking ability (Jo & Bednarz, 2014). The
spatial thinking ability is needed by students in the future for a variety of things, including determining the strategic
location of public facilities, planning regional spatial planning, analyzing impacts and benefits of land changes,
projecting population numbers, distribution and its interaction with the environment. Geography learning can
provide opportunities to develop these spatial thinking abilities because it studies the material and systems of the
earth in terms of space, environment and region.
Some previous research showed that spatial thinking ability can be improved by various media and methods
including the application of PBL-GIS (Liu, Bui, Chang, & Lossman, 2010), geospatial technology (Metoyer & Bednarz,
2017), GIS Learning (Kim & Bednarz, 2013), development of an e-book (Yusup et al., 2018) and test instruments
(Aliman, Mutia, & Yustesia, 2018; Huynh & Sharpe, 2013; Jo, Bednarz, & Metoyer, 2010; Mulyadi et al., 2018). However,
there are not many studies that examine the spatial thinking ability of high school students that influence the
learning outcomes of geography by applying earthcomm learning although many studies have applied various
learning models to improve learning outcomes. However, research to obtain an effective and efficient learning
model in improving geography learning outcomes by measuring spatial thinking ability still needs to be done.
Problem of Research
The research problem was the low geography learning outcomes of high school students. In addition, the
selection of inappropriate geography learning models is one of the causes of the low learning outcomes of ge-
ography. The effort that can be done to improve geography learning outcomes is to develop students’ abilities in
spatial thinking. The spatial thinking ability should be a point of view in studying geography. For this reason, it is
necessary to know in advance the students’ spatial thinking skills so that the application of learning models can
be optimal in improving geography learning outcomes.
Research Focus
Research focuses on 1) knowing the application of the earthcomm learning model to the geography learning
outcomes, 2) knowing the role and ability of spatial thinking on geography learning outcomes.
Research Methodology
General Background
This research used 2x2 factorial quasi-experimental with non-equivalent control group design. The details
can be seen in table 1.
324
ISSN 1648–3898 /Print/ THE EFFECT OF EARTHCOMM LEARNING MODEL AND SPATIAL THINKING ABILITY ON
ISSN 2538–7138 /Online/ (P. 323-334)
Table 1. Pattern of 2x2 factorial experimental.
Learning Model
Type of Spatial Thinking
Earthcomm Learning Conventional Learning
High spatial Group 1 Group 2

Spatial Thinking Ability
Low spatial Group 3 Group 4
Based on table 1, there were two applications of learning models, namely earthcomm learning models and
conventional learning models. The ability of spatial thinking also has two types, namely high spatial thinking and
low spatial thinking. Therefore, 1) research examined the effect of applying the earthcomm learning model and
conventional learning models regardless of the influence of students’ spatial thinking types, 2) examined the
influence of spatial thinking types without assessing their relationship with the application of learning models.
Research of interactions between variables is 3) whether there is an interaction between the application of earth-
comm learning models and spatial thinking ability towards geography learning outcomes of high school students.
The scope of the research was to examine the application of the earthcomm learning model, the spatial thinking
ability of high school students and the geography learning outcomes of high school students. The research was
conducted from March 27 to April 5, 2018.
Sample
The Senior High Schools used as the research sample came from the school population in Malang City, In-
donesia. The samples were selected from 47 high schools in Malang City, Indonesia by using a stratified random
sampling technique based on accreditation of superior schools, national standard schools and private schools.
The high school that became the sample of the research was Public High School 5 as a superior school (the best
school in the city based on the students’ score average in national examination), Public High School 6 as a national
standard school and Private High School PANJURA as a private school. The two classes were taken from each school
to be used as experimental and control classes. The selection of the two research classes was based on the equal-
ity test from the results of the final examination in the previous semester. The selection of the experimental class
and the control class was based on the average score of geography learning outcomes between classes that is
almost similar. The two classes that have similar geographic learning outcomes, were assumed to have the same
cognitive abilities (Table 2).
Table 2. Selection of experimental classes and control classes.
Experimental Control
Learning
Criterion Schools Classes Total
Outcomes
Class Male Female Class Male Female
X Social 1 91
X
Public High X Social
Superior X Social 2 85 Social 7 23 13 14 57
School 5 3
1
X Social 3 89
X Social 1 76
X Social 2 86
X
Public High X Social
National X Social 3 78 Social 5 23 11 17 56
School 6 5
2
X Social 4 80
X Social 5 85
X Social 1 70
Private X
X Social
Private High School X Social 2 73 Social 10 15 8 21 54
3
Panjura 2
X Social 3 74
Total 22 61 32 52 167
325
(P. 323-334) ISSN 2538–7138 /Online/
The experimental class consisted of 84 students and the control class consisted of 83 students. The 167
students in the research sample (113 female students and 54 male students), were given a test to find out their
spatial thinking skills. Based on the data obtained, the 121 students have high spatial ability and 46 students have
low spatial ability.
Instrument and Procedures
The instruments used in this research consisted of three instruments. First, instruments that were related to
the application of earthcomm learning models and conventional learning models which included learning tools.
Learning tools consisted of learning implementation plans (RPP), teaching materials, problem training, and stu-
dent worksheets for investigations in the field. In the national curriculum of geography subject, there are several
demands for competency for students, namely basic knowledge competencies and basic skills competencies. One
of the basic competency demands for class X is the knowledge based competency in the form of analyzing the
dynamics of the hydrosphere and its impact on human life. Meanwhile, basic skills competencies are in the form
of presenting the hydrosphere dynamics process using maps, tables and graphics. The instrument of this learning
device was validated by learning experts from Malang state of university and discussed with geography teachers
in research schools to suit the needs of students.
Second, the final evaluation instrument for geography learning was used to determine the learning outcomes
of geography. This instrument was made to measure student competency in accordance with national curriculum
objectives. The instrument was also validated by learning evaluation experts and geography teachers at research
site. The learning outcome assessment instrument consisted of 25 multiple choice questions with five answer
choices and this data becomes a unit of analysis in data processing.
Third, the instrument used to measure spatial thinking ability was adapted from (Huynh & Sharpe, 2013)
and adjusted to the thinking level of high school students in Indonesia. Spatial thinking instruments consist of 26
questions containing indicators: analysis, comprehension, representation, application, scale, spatial interaction. The
questions were in the form of multiple choices and each item is composed of four answer choices. Each question
has one point for a correct answer and zero for an incorrect answer. The spatial thinking instrument was validated
according to the learning evaluation expert. Expert validation was done to adjust the contents of the instrument
with Indonesian students’ thinking ability and adjust to the national curriculum. Instrument trials were conducted
on 92 students at National Senior High School 3 in Malang. Based on Instrument test results analyzed empirically
by reliability tests, it was obtained Cronbach Alpha value of (.729) which means that this instrument has a high
consistency. Validity test showed r results greater than r table (.207) which means that 26 instrument questions are
valid empirically and can be tested in the research class. The results of spatial thinking ability obtained by students
were grouped into high spatial thinking and low spatial thinking. The group divided based on total number of
correct answers in which 0-13 answers defined students has low spatial thinking and 14-26 answers defined has high
spatial thinking. The data gained during classroom first meeting which requires 100 minutes to complete the test.
Based on the test of spatial thinking ability that was tested on students, it was obtained high level and low
level spatial thinking abilities data. The details of the distribution of the research subject can be seen in table 3.
Table 3. Distribution of research subject.
Learning Model
Type of Spatial
Total
Thinking
Earthcomm Learning Conventional Learning
High spatial 64 57 121

Spatial Thinking Ability Low spatial 19 27 46
Total 83 84 167
Distribution of the research samples in table 3 shows that each data analysis unit column is filled. The num-
ber of research samples for high spatial thinking columns is more in number than low spatial thinking columns.
However, analysis using Two Way Anova can still find out the differences that occur when there are differences in
the number of research samples.
326
ISSN 2538–7138 /Online/ (P. 323-334)
Data Analysis
Quantitative data analysis used prerequisite and statistical tests using SPSS version 23 software for Windows.
There were two types of prerequisite tests used, namely normality test and homogeneity test. The normality test used
Kolmogorov-Smirnov with a significance value (p > .05) and homogeneity test used Leven’s test with a significance
value (p > .05). Meanwhile, the statistical test used Two Way ANOVA. The details can be seen in the following table.
Table 4. Normality tests.
Kolmogorov-Smirnova
Statistic df p
Standardized Residual for Post .044 167 .200*
Based on the normality test in table 4, it was obtained a significance value (p > .05), this indicates that the
spatial thinking ability data is classified as normal. Furthermore, the homogeneity value was tested with the Lev-
ene’s test as in table 5.
Table 5. Homogeneity test.
F df1 df2 p
.881 3 163 .452

Table 5 shows that the significance value obtained was .45 that means greater than the significant level of
p > .05. It can be concluded that the spatial thinking ability data is classified as homogeneous.
Research Results
The first step in processing data was to find the average geography learning outcomes of students based
on spatial thinking skills (high and low) in the experimental class and the control class. Furthermore, the average
learning outcomes were distinguished statistically using the independent sample t test to see whether there were
differences in the average geography learning outcomes between students who have high level spatial thinking
skills in the experimental and the control class and low level spatial thinking skills in the experimental and control
class. The next stage, data on geography learning outcomes and spatial thinking abilities of students were processed
using parametric statistics using the two way ANOVA test to see whether there were any influences and interactions
between the two research variables. The details can be seen in the table 6, table 7 and table 8.
Table 6. Descriptive statistics.
Class Spatial M SD N
High 20.14 3.01 64

Experimental Low 18.63 3.77 19
Total 19.80 3.24 83
High 15.68 3.48 57
Control Low 14.81 3.74 27
Total 15.40 3.57 84
High 18.04 3.93 121
Total Low 16.39 4.17 46
Total 17.59 4.05 167
327
(P. 323-334) ISSN 2538–7138 /Online/
The results in table 6 explain that there are differences in the M (Mean) total geography learning outcomes
between the experimental class and the control class namely (M experimental = 19.80; SD = 3.241) and M control =
15.40; SD = 3.567) there is a range of M values = 4.40. From these results, the average experimental class geography
learning outcomes were higher than the control class geography learning outcomes. The results of independent
test sample t test table 7 below support statistical differences from the average value of spatial thinking ability.
Table 7. Comparison of Spatial Thinking Ability Using the Independent Sample t-test.
Levene’s Test for

t-test for Equality of Means
Equality of Variances
Equal
variances
95% CID
assumed p
F p t df MD SED
(2-tailed)
Lower Upper
Spatial High 13.123 .000 3.195 119 .002 1.228 .384 .467 1.989
Thinking
Skills Low 1.056 .310 2.393 44 .021 2.096 .876 .331 3.861
Table 7 shows the significance value of high-level spatial thinking skills (F = 13.12, p (2-tailed) < .05). It means
that there were significant differences between students who have high spatial thinking ability in the experimen-
tal class and students who have high spatial thinking ability in control class. Students who had low level spatial
thinking ability in the experimental and control class have significant differences according to the values (F = 1.05,
p (2-tailed) < .05) in table 7.
In addition, the geography learning outcomes of students who had high spatial thinking ability in the
experimental class are greater than the geography learning outcomes of students who have low spatial ability.
This is proven in table 5 which shows that the average geography learning outcomes of students who have high
spatial thinking ability obtain scores (M = 20.14; SD = 3.01). Geography learning outcomes of students who had
low spatial thinking ability obtain scores (M = 18.63; SD = 3.77). Of the two average learning outcomes, there was
a difference M = 1.51. The details can be seen in Figure 1.
Figure 1. Geography learning outcomes based on spatial thinking ability.
In the control class, there were differences in geography learning outcomes between students who have
high spatial ability and students who have low spatial ability. It is proven that the geography learning outcomes
328
ISSN 2538–7138 /Online/ (P. 323-334)
of students with high spatial thinking ability were greater in value (M = 15.68; SD = 3.48) than geography learning
outcomes of students with low spatial ability (M = 14.81; SD = 3.74), there is a range of M values = .87. If we con-
sider the difference in the average range of each experimental class and the control class between the geography
learning outcomes of students who have high and low spatial ability, it is not too much difference. However, when
it is compared to the differences in the average total geography learning outcomes of the experimental class and
control class, there were quite large differences. The table 6 below shows how much influence of learning and
spatial thinking ability on geography learning outcomes.
Table 8. Results of Two Way ANOVA of Earthcomm learning and spatial thinking ability.
Source Type III SS df MS F p
Corrected Model 851.946a 3 283.982 24.746 .0001

Intercept 39063.838 1 39063.838 3404.037 .0001
Class 557.204 1 557.204 48.555 .0001
Spatial 46.053 1 46.053 4.013 .047
Class * Spatial 3.331 1 3.331 .290 .591
Error 1870.545 163 11.476
Total 54375.000 167
Corrected Total 2722.491 166
a. R Squared = .313 (Adjusted R Squared = .300)
Based on table 6, it shows that there are differences in geography learning outcomes between the two
research classes, as evidenced by the significance value (F = 48.55; p < .05). In spatial thinking ability data, there
were significant differences between students who have high and low spatial ability in influencing their geography
learning outcomes, as evidenced by the significance value (F = 4.01; p < .05). Moreover, there was no interaction
between earthcomm learning and spatial thinking ability in influencing geography learning outcomes, as evidenced
by the value (F = .29; p > .05).
Discussion
Effects of Earthcomm Learning on Geography Learning Outcomes
Earthcomm learning emphasizes that students directly experienced the process of learning, especially
experience in conducting scientific investigations. Earthcomm learning was not only done in the classroom but
also done outside the classroom by inviting students to investigate river water quality and the social conditions
of the people that affect river water quality. Investigations were carried out in groups at three locations in the
upper Brantas watershed area. In-depth investigations conducted by students are able to stimulate students’
awareness to find out the basic problems about the river. This learning not only emphasizes the students’ memory
to understand the subject matter, but also teaches students to be able to provide ideas and concepts in dealing
with problems in their environment (Carpenter & Hoover, 2018; Park et al., 2005). In accordance with the inves-
tigation phase, students are directly involved in investigating the causes of flooding in the Brantas river basin,
Malang City. Furthermore, at the digging deeper stage, students conduct in-depth discussions to strengthen
the results of investigations from other supporting sources such as reference books and the internet. Both of
these stages function to improve students’ cognitive abilities in geography learning (Park & Park, 2013). This is
evidenced by data analysis which shows that earthcomm learning is also able to improve geography learning
outcomes and provide students direct experience when conducting investigations in the field.
The results of this research are supported by several previous studies, among others (Dewi, 2014; Hidayat
et al., 2017; Mari et al., 2017; Mauliddia et al., 2018; Park et al., 2005; Suharto, 2016). The study also proved that
learning involving students in investigations in the field is also able to improve students’ ability cognitively
(Eysenck, 2018; Fatchan, Soekamto, Sumarmi, & Utaya, 2016; Mayer & Alexander, 2011). Even field studies involv-
ing students in certain areas and trips can sustain learning outcomes consistently (Jolley et al., 2018). Besides
329
(P. 323-334) ISSN 2538–7138 /Online/
being able to improve geography learning outcomes, learning directly in the field can also improve knowledge
memory, quality of knowledge and interest in knowledge (Korže, 2005). It is the same as earthcomm learning
which requires students to conduct a direct investigation into the problem discussed previously. In the thought
process, students’ brains are able to connect spatial reasoning functions because there are various spatial basic
structures which are then compiled and stimulated by learning (Anthamatten, 2010).
The application of earthcomm learning can answer the challenges of low geography literacy and geography
skills. This is evidenced by the average geography literacy value of high school students in Surabaya at 64.11
and map literacy at 36.35 (Utami et al., 2018). Therefore, the application of earthcomm learning is expected to
be able to improve geography literacy, as well as the application of REACT strategies that are able to develop
geography skills especially in the knowledge component of geography (Wiwik Sri Utami et al., 2016a). In addi-
tion, the reality in the field found that only 8% of social studies teachers in Surakarta City understood geography
literacy (Sugiyanto et al., 2018). It can be interpreted that through the application of earthcomm learning it is
alleged that it can develop the geography literacy skills of social studies teachers.
The implementation of earthcomm learning involves all students’ ability cognitive and affective. In earth-
comm learning, there is a stage of investigation, which involves students that can stimulate their cognitive and
affective responses. The response raised by students is then represented in the form of geography knowledge
and attitudes in dealing with environmental problems. In addition to the investigation phase, the in-depth
study phase and the understanding stage of the application play a role in training students to think critically as
outlined in the form of portfolios so their knowledge can be assessed and evaluated (Chetcuti & Pace, 2012). This
stage trains students to study the earth system and problems that occur from various other secondary sources.
The process experienced by students can directly construct their own knowledge, this is in accordance with the
concept of constructivism in education (Atkin, 2018; Park et al., 2005).
The concept of educational constructivism is very much in line with the needs of students in facing the
challenges of global competition. This concept provides space for active students to be able to compile their
own knowledge directly from the experience that it passes (Jančič & Hus, 2019). In earthcomm learning, inves-
tigations in the field by students are conducted in groups. In this group, students are able to train their ability
to communicate with friends in getting answers of the problems found. This stage fosters the habits of students
to communicate well with their peers and train students to always find the right information in accordance
with the references they are looking for. The communication system between students is also useful in shaping
new knowledge and enhance their knowledge memory. The process of investigation in the field by students is
guided by earthcomm learning field worksheets so that it is very helpful for students in improving the learn-
ing outcomes of geography (Wiwik Sri Utami, Sumarmi, Ruja, & Utaya, 2016b). This is in accordance with social
constructivism theory which states that knowledge can be built from the relationship and closeness among
students (Nassaji & Tian, 2018).

Effect of Spatial Thinking Ability on Geography Learning Outcomes
Based on the results of the research in tables 6 and 8, it has been proven that there is a significant influ-
ence between students’ spatial thinking ability (high and low) on geography learning outcomes. The results of
this study reinforce research that states that some students in Athens have good geography learning outcomes
because they are influenced by high spatial thinking ability (Klonari & Likouri, 2015). In Japan, learning outcomes
related to topographic maps are influenced by spatial thinking ability (Wakabayashi, 2013). The use of giant
travel map media can improve geography learning outcomes in South California (Fleming & Mitchell, 2017). A
similar study states that the use of geospatial technology can improve students’ geography thinking (Metoyer
& Bednarz, 2017), PBL-GIS learning is also able to improve student learning outcomes in Singapore (Liu et al.,
2010), this learning can improve learning outcomes because in GIS has integrated the component of spatial
thinking. Some of these studies prove that the components of spatial thinking such as analysis, representation,
scale, comprehensiveness and spatial interaction (Huynh & Sharpe, 2013) in geography learning have an influ-
ence to improve geography learning outcomes. Spatial thinking skills in practice can be realized in various forms
such as: using colors on grammar tests that can improve test results in students who have visual-spatial learning
styles in the UAE (Moradkhan, Karimi, & Aryan, 2014), the use of GIS-based learning (Kim & Bednarz, 2013) and
PBL-GIS-based e-books (Yusup et al., 2018).
The ability of spatial thinking as a whole has been in the students. The students’ knowledge and skills related
330
ISSN 2538–7138 /Online/ (P. 323-334)
to spatial thinking skills that were not trained caused the difference of ability level among them. The high or low
level of spatial thinking skills of students can be functioned optimally through geography learning (Yani et al.,
2018). Good geography learning is learning that brings basic concepts of geography and geographic approaches
such as spatial, environmental and regional. The concept and approach of geography that has been integrated
in each learning material can train students’ spatial thinking skills indirectly. Moreover, good geography learning
is learning that involves students contextually in their closest lives so that students recognize and understand
the geographical environment including the spatial conditions of the region (Aliman et al., 2018).
This research is different from previous studies which examined the relationship between spatial ability and
geography skills (learning outcomes). The study stated that there was a decline in geography skills (geography
learning outcomes) in high school students due to differences in the question of spatial ability with the mate-
rial being studied (Yani et al., 2018). This research also continues the research on the development of spatial
intelligence instruments that are integrated with geography material (Mulyadi et al., 2018). Measurement of
intelligence and spatial thinking of students can be integrated with geography material so that this instrument
can also measure students’ geography learning outcomes. Other studies in Vietnam state that questions and
spatial concepts in geography textbooks contain little spatial components compared to non-spatial components
(Nguyen, Muniz-Solari, Tien Dang, & Nguyen, 2018).
Earthcomm Learning Interaction and Spatial Thinking Ability to

Geography Learning Outcomes
This research proves that there is no interaction between earthcomm learning and students’ spatial thinking
ability in influencing geography learning outcomes. The results of research are assumed to have no interaction
because the instrument of spatial thinking ability does not integrate the material being studied by students
when the tests are conducted. This is understandable because lessons at the high school level are specific and
not integrated with other lessons like in elementary and junior high school lessons that apply integrated lessons
such as social studies and natural sciences. Meanwhile, the spatial thinking ability test developed is a test that
can be used to measure spatial thinking ability of all high school levels. The questions contained in the instru-
ment are detailed components such as answering questions on topographic maps that have a large scale that
requires spatial thinking ability. Therefore, the spatial ability of students to answer questions on a small scale
matters such as answering questions about thematic maps on a small scale depends only on general knowledge
and geographical knowledge, not on spatial thinking (Wakabayashi, 2013).
Although there is no interaction between the application of earthcomm learning and spatial thinking on
the results of geography learning, each of the independent variables has a significant influence on the increase
in geography learning outcomes. The absence of interaction between the two variables is assumed as the effect
of the stages in earthcomm learning that were oriented towards engaging students directly in investigating
environmental problems while students did not see the problem from a spatial perspective. In fact, one of the
fundamental philosophies of earthcomm learning is community (Park et al., 2005). The existence of the com-
munity must not be involved in investigations, but students must examine that there is community involvement
in these environmental problems. This viewpoint is a larger viewpoint of thinking and is a spatial thinking pat-
tern. Not all students are able to have a mindset on a larger scope of space, but this spatial mindset must still
be trained in every geography learning.
This is different from the results of research on elementary school students in Mexico. The results of research
explain that there are differences in spatial ability based on the spatial scale between search tasks in the field
with search tasks on the computer (Rosetti, Valdez, & Hudson, 2017). The results of this study are suspected be-
cause students are involved in the real process in the field so that students use all their motor sensors (Downs
& Stea, 2011), in contrast to students who use computers because they only use smaller real spaces. It can be
interpreted that learning involving students (especially elementary and junior high school students) directly
in the field can make it easier for students to understand spatial conditions compared to learning done in class
or through a computer.
Conclusions
Based on the discussion of research results, it can be concluded that 1) the results of student geography
331
(P. 323-334) ISSN 2538–7138 /Online/
learning taught earthcomm learning and students taught conventional learning differ significantly. This shows
that earthcomm learning has a significant effect on improving geography learning outcomes compared to
conventional learning; 2) geography learning outcomes for students who have low spatial thinking ability and
students who have high spatial thinking ability differ significantly. This proves that the high spatial ability of
students is better at learning the results of their geography compared to students’ low spatial ability; 3) the
application of earthcomm learning and conventional with spatial thinking ability students have no interaction
with the learning outcomes of geography.
Based on the conclusions, there are a number of suggestions relating to future research, among others 1)
For researchers, especially teachers in schools, they can map students ‘spatial thinking ability at the beginning
of learning so that they can plan appropriate learning to improve geography learning outcomes, 2) researchers
and teachers can modify earthcomm learning that can improve students’ low spatial thinking ability, 3)Teach-
ers need to create a meaningful lesson about spatial thinking ability through either digital application (google
map, google earth, GIS) or apply manually using analog map, 4) the further research is needed to develop test
instruments in measuring integrated spatial thinking ability to learning materials that can be analyzed based
on gender, learning style, learning motivation and other variables.
Acknowledgements
The authors would like to thank the heads of Senior High School 5 Malang, the Head of Senior High School 6 Malang and the
Head of the PANJURA High School Malang who have provided the opportunity to conduct research.
References
Aliman, M., Mutia, T., & Yustesia, A. (2018). Integritas kebangsaan dalam tes berpikir spasial [Nationality integrity in spatial think-
ing tests]. In Prosiding Seminar Nasional Pendidikan Geografi FKIP UMP 2018 (pp. 82–89). Purwokerto: UM Purwokerto Press.
Anthamatten, P. (2010). Spatial thinking concepts in early grade-level geography standards. Journal of Geography, 109(5), 169–180.
https://doi.org/10.1080/00221341.2010.498898.
Atkin, A. (2018). Charles Sanders Peirce: Pragmatism. In Internet encyclopedia of philosophy. A Peer Reviewed Academic Resource.
Retrieved from http://www.iep.utm.edu/peircepr/.
Bahri, A., & Corebima, A. D. (2015). The contribution of learning motivation and metacognitive skill on cognitive learning outcome
of students within different learning strategies. Journal of Baltic Science Education, 14(4), 487–500.
Carpenter, M., & Hoover, M. (2018). What is earthcomm, second edition? [html]. Retrieved January 15, 2018, from http://www.
agiweb.org/education/earthcomm2/about.html.
Chetcuti, D., & Pace, P. (2012). Portofolios in environmental education: The TEPEE Project. Journal of Baltic Science Education,
11(1), 16–28.
Dewi, A. I. (2014). Pengaruh pendekatan earth science community (Earthcomm) dan pembelajaran berbasis masalah terhadap
kemampuan berpikir kritis peserta didik [The influence of the earth science community (Earthcomm) and problem-based
learning approaches on students’ critical thinking skills] (Master’s thesis). Universitas Pendidikan Indonesia, Bandung.
Retrieved from www.repository.upi.edu.
Downs, R. M., & Stea, D. (2011). Cognitive maps and spatial behaviour: Process and products. In The Map Reader (pp. 312–317).
John Wiley & Sons. https://doi.org/10.1002/9780470979587.ch41.
Eysenck, H. (2018). Intelligence: A new look. Retrieved December 11, 2018, from https://www.taylorfrancis.com/
books/9781351310031.
Fatchan, A., Soekamto, H., Sumarmi, & Utaya, S. (2016). Effect of learning “outdoor study” ability to communicate in writing and
social-geography student learning outcomes at “Mataraman” East Java-The Republic of Indonesia*. Mediterranean Journal
of Social Sciences, 7(3), 429-435. https://doi.org/10.5901/mjss.2016.v7n3p429.
Fleming, J., & Mitchell, J. (2017). Effects of giant traveling map use on student spatial thinking. The Geographical Bulletin, 58, 67–77.
Hidayat, A., Sarwono, & Yusup, Y. (2017). Earthcomm-based multimedia learning of geography in improving learning motiva-
tion and spatial ability of the high school students. In Proceedings of The International Conference on Teachinng Training
and Education 2017 (ICTTE 2017). Universitas Sebelas Maret, Surakarta, Central Java, Indonesia: Atlantis Press. https://doi.
org/10.2991/ictte-17.2017.29.
Homoki, E., & Sütő, L. (2013). Studying the public opinion of geography as a subject and its knowledge elements: A case of
Hungary. Journal of Baltic Science Education, 13(4), 508–522.
Huynh, N. T., & Sharpe, B. (2013). An assessment instrument to measure geospatial thinking expertise. Journal of Geography,
112(1), 3–17. https://doi.org/10.1080/00221341.2012.682227.
Jančič, P., & Hus, V. (2019). Representation of teaching strategies based on constructivism in social studies. International Journal
of Innovation and Learning, 25(1), 64–77.
Jeronen, E. (2004). Development of conceptions of student teachers in initial teacher education in biology and geography
education. Journal of Baltic Science Education, 3(1), 24–33.
Jo, I., Bednarz, S., & Metoyer, S. (2010). Selecting and designing questions to facilitate spatial thinking. The Geography Teacher,
7(2), 49–55. https://doi.org/10.1080/19338341.2010.510779.
332
ISSN 2538–7138 /Online/ (P. 323-334)
Jo, I., & Bednarz, S. W. (2014). Dispositions toward teaching spatial thinking through geography: Conceptualization and an ex-
emplar assessment. Journal of Geography, 113(5), 198–207. https://doi.org/10.1080/00221341.2014.881409.
Jolley, A., Hampton, S. J., Brogt, E., Kennedy, B. M., Fraser, L., & Knox, A. (2019). Student field experiences: designing for different
instructors and variable weather. Journal of Geography in Higher Education, 43(1), 71–95. https://doi.org/10.1080/030982
65.2018.1554632.
Kim, M., & Bednarz, R. (2013). Development of critical spatial thinking through gis learning. Journal of Geography in Higher Educa-
tion, 37(3), 350–366. https://doi.org/10.1080/03098265.2013.769091.
Klonari, A., & Likouri, A. A. (2015). The Relation of multiple intelligences and spatial perception with performance in geography
education. GI_Forum, 1, 359–362. https://doi.org/10.1553/giscience2015s359.
Korže, A. V. (2005). Transition and transformation: The influence of a learning package on student performance in soils geography
studies. Journal of Geography in Higher Education, 29(3), 369–384. https://doi.org/10.1080/03098260500290926.
Liu, Y., Bui, E. N., Chang, C.-H., & Lossman, H. G. (2010). PBL-GIS in secondary geography education: Does it result in higher-order
learning outcomes? Journal of Geography, 109(4), 150–158. https://doi.org/10.1080/00221341.2010.497541.
Mari, N. A. H., Sumarmi, & Handoyo, B. (2017). Kontribusi Model Earthcomm dalam Pembelajaran Geografi untuk Membangun
Keterampilan Hidup (Life Skills) Abad 21 [Contribution of the earthcomm model in learning geography to building 21st
century life skills]. In Seminar Nasional (pp. 231–237). Aula Ki Hajar Dewantara FIS UM. Malang: Fakultas Ilmu Sosial Uni-
versitas Negeri Malang.
Mauliddia, H. P., Muryani, C., & Rintayati, P. (2018). Earthcomm-based electronic module: The learning material of natural resource
management wisdom. IOP Conference Series: Earth and Environmental Science, 145, 012003. https://doi.org/10.1088/1755-
1315/145/1/012003.
Mayer, R. E., & Alexander, P. A. (Eds.). (2011). Handbook of research on learning and instruction. New York: Routledge.
Metoyer, S., & Bednarz, R. (2017). Spatial thinking assists geographic thinking: Evidence from a study exploring the effects of
geospatial technology. Journal of Geography, 116(1), 20–33. https://doi.org/10.1080/00221341.2016.1175495.
Moradkhan, D., Karimi, K., & Aryan, A. (2014). Colorful ESL test papers and spatial intelligence: should one simply throw paint on
exam papers to ameliorate them in the eyes of spatially intelligent candidates? International Journal of Applied Linguistics
and English Literature, 3(6), 176–204.
Mulyadi, A., Yani, A., Ismail, A., & Rosita, R. (2018). Students’ spatial intelligence measurement on social science and geography sub-
jects. IOP Conference Series: Earth and Environmental Science, 145, 012043. https://doi.org/10.1088/1755-1315/145/1/012043.
Nassaji, H., & Tian, J. (2018). Constructivism. Leiden, The Netherlands: Interactive Factory. https://doi.org/10.1163/9789004380882_003.
Nguyen, N.-A., Muniz-Solari, O., Tien Dang, D., & Nguyen, T. (2018). Reviewing spatial thinking in Vietnam geography textbooks
questions. Presented at the Geography for Global Understanding-Sustainable Changes in Environment, Society and People
(SEAGA International Conference 2017), Jakarta.
Park, D. Y., Yager, R. W., & Smith, M. (2005). Implementing EarthComm: Teacher professional development and its impact on
student achievement scores in a standards-based earth science curriculum. Electronic Journal of Science Education, 9(3).
Retrieved from http://ejse.southwestern.edu/article/view/7733.
Park, D. Y., & Park, M. (2013). Examining the features of earth science logical reasoning and authentic scientific inquiry demon-
strated in a high school earth science curriculum: A case study. Journal of Geoscience Education, 61(4), 364–377. https://
doi.org/10.5408/12- 360.1.
Park, M., Park, D.-Y., & Lee, R. E. (2009). A comparative analysis of earth science curriculum using inquiry methodology between
korean and the US textbooks. Eurasia Journal of Mathematics, Science & Technology Education, 5(4), 395-411. https://doi.
org/10.12973/ejmste/75289.
Railienė, L. (2003). How geography and natural science subjects come into contact. Journal of Baltic Science Education, 2(2), 62–69.
Rosetti, M. F., Valdez, B., & Hudson, R. (2017). Effect of spatial scale on children’s performance in a searching task. Journal of
Environmental Psychology, 49, 86–95. https://doi.org/10.1016/j.jenvp.2016.12.004.
Soobard, R., & Rannikmäe, M. (2013). Upper secondary students’ self-perceptions of both their competence in problem solving,
decision making and reasoning within science subjects and their future careers. Journal of Baltic Science Education, 13(4),
544–558.
Sudatha, G. W., Degeng, N. S., & Kamdi, W. (2018). The effect of visualization type and student spatial ability on learning achieve-
ment. Journal of Baltic Science Education, 17(4), 551–563.
Sugiyanto, S., Maryani, E., & Ruhimat, M. (2018). A preliminary study on developing geography literacy based on social studies
learning model to improve teachers 21st century skills. IOP Conference Series: Earth and Environmental Science, 145(1),
012062. https://doi.org/10.1088/1755-1315/145/1/012062.
Suharto, Y. (2016). The effect of the earthcomm zooming model toward creative thinking skill senior high school student. In
Strengthening Synergy of Geography for Our Future World (p. 184). Atria Hotel, Malang, East Java, Indonesia: Department of
Geography, Faculty of Social Science, Universitas Negeri Malang.
Sumarmi. (2015). Model-model pembelajaran geografi [Geography learning models]. Yogyakarta: Aditya Media Publishing.
Utami, Wiwik Sri, Zain, I. M., & Sumarmi. (2018). Geography literacy can develop geography skills for high school students: Is it true?
IOP Conference Series: Materials Science and Engineering, 296(1), 012032. https://doi.org/10.1088/1757-899X/296/1/012032.
Utami, Wiwik Sri, Sumarmi, Ruja, I. N., & Utaya, S. (2016a). React (Relating, experiencing, applying, cooperative, transferring)
strategy to develop geography skills. Journal of Education and Practice, 7(17), 100–104.
Utami, Wiwik Sri, Sumarmi, Ruja, I. N., & Utaya, S. (2016b). The effectiveness of geography student worksheet to develop learning
experiences for high school students. Journal of Education and Learning, 5(3), 315. https://doi.org/10.5539/jel.v5n3p315.
333
(P. 323-334) ISSN 2538–7138 /Online/
Wakabayashi, Y. (2013). Role of geographic knowledge and spatial ability in map reading process: Implications for geospatial
thinking. Geographical Reports of Tokyo Metropolitan University, 48, 37–46.
Yani, A., Mulyadi, A., & Ruhimat, M. (2018). Contextualization of spatial intelligence: Correlation between spatial intelligence,
spatial ability, and geography skills. Journal of Baltic Science Education, 17(4), 564–575.
Yusup, Y., Santoso, S., & Istifarida, B. (2018). Pengembangan e-book berbasis problem-based learning-gis untuk meningkatkan
kecakapan berpikir keruangan pada siswa kelas X SMAN 1 [E-book development based on problem-based learning-gis
to improve spatial thinking skills in class X SMAN 1]. Paedagogia, 20(2), 134–149. https://doi.org/10.20961/paedagogia.
v20i2.8596.
Received: February 04, 2019 Accepted: May 05, 2019
Muhammad Aliman PhD Student, Geography Education Department, State University

of Malang, Jl. Semarang No.5, Malang 65145, Indonesia.
E-mail: muhammad.aliman.1607219@students.um.ac.id
Website: http://www.um.ac.id
Budijanto PhD, Professor, Geography Education Department, State University
E-mail: budijanto.fis@um.ac.id
Sumarmi PhD, Professor, Geography Education Department, State University
E-mail: sumarmi.fis@um.ac.id
I Komang Astina PhD, Geography Education Department, State University of
Malang, Jl. Semarang No.5, Malang 65145, Indonesia.
E-mail: komang.astina.fis@um.ac.id
Rozana Eka Putri M.Sc, Geography Education Department, STKIP PGRI Sumatera
Barat,
E-mail: rozawirman@gmail.com
Website: http://www.stkip-pgri-sumbar.ac.id/
Muhammad Arif M.Ed, Geography Education Department, STKIP Pesisir Selatan,
E-mail: m_arif@stkip-pessel.ac.id
Website: http://stkip-pessel.ac.id
334
THE EFFECT OF COMPUTER-
AIDED 3D MODELING
ACTIVITIES ON PRE-SERVICE ISSN 1648-3898 /Print/
TEACHERS’ SPATIAL ABILITIES ISSN 2538-7138 /Online/
AND ATTITUDES TOWARDS 3D

MODELING
Ali Ihsan Benzer, Abstract. The computer-aided 3D

modelling which is one of the innovative
Bunyamin Yildiz technologies can offer great opportuni-
ties to improve students’ skills. The aim of
this research was to examine the effects
of computer-aided 3D modeling activities
on pre-service teachers’ spatial abilities
Introduction and attitudes towards 3D modeling and
the relevant course. The study group of the
Spatial ability is one of the leading cognitive competencies that can research was composed of 55 pre-service
affect the success of individuals in their daily and professional lives. There- IT teachers at a state university in Turkey.
fore, developing spatial ability has become one of the main objectives of The research was carried out in quasi-
education. Spatial ability has been the subject of debate for nearly a hundred experimental design based on pre-post test
years (Bishop, 1980; Sorby, 1999). Galton’s work in 1918 has been referred model. The experimental research was car-
to as the main source of research on spatial ability (Friedman, 1992). When ried out with the experimental group for 14
the literature on spatial ability is examined, it was seen that terms such as weeks. During the 3D modeling learning/
“spatial skill”, “visuo-spatial ability”, “spatial thinking”, “spatial reasoning” teaching process, a five-stage education
and “spatial sense” which have a similar meaning were used instead of the framework based on problem-based and
term “spatial ability”. So far, many definitions have been made about spatial project-based learning approaches was
ability. Still, there is no consensus on the definition of spatial ability (Kösa & used. “Purdue Visualization of Rotations
Karakuş, 2018). Linn and Petersen (1985) described spatial ability as the abil- Test” and “Attitude Scale towards 3D Mod-
ity to describe, transform, create and remember symbolic and non-linguistic eling and 3D Modeling Course” were used
information. Tartre (1990) defined spatial ability as the entire of mental skills as data collection tools in the research. As
related to understanding, changing, reorganizing, or interpreting relations a result, it was found out that computer-
in visual terms. Lohman (1996) defined spatial ability as the ability to cre- aided 3D modeling activities improved the
ate and keep in mind, and retrieve, and transform visual images. When the spatial abilities of pre-service teachers also
definitions are considered together, spatial ability can be defined as the increased their attitudes towards 3D mod-
capability to perform manipulative operations on visual images in mind and eling and the course. It was concluded that
to remember, understand and analyze relations between visual images. In the education of computer-aided three-
the literature, there are studies in which spatial ability is subdivided into two dimensional modeling offers important
components as spatial visualization and spatial orientation (Clements, 1998; opportunities to improve spatial abilities.
McGee, 1979; Tartre, 1990) or spatial relations and spatial visualization (Olkun,
Keywords: 3D modeling education, 3D
2003; Olkun & Altun, 2003; Pellegrino, Alderton, & Shute, 1984) or into three
modeling, 3D modeling attitude, pre-
components as spatial perception, mental rotation and spatial visualization
service teachers, spatial ability.
(Linn & Petersen, 1985).
Spatial ability that is one of the important components of multiple in-
telligences (Gardner, 2011) and one of the important individual differences
Ali Ihsan Benzer, Bunyamin Yildiz
(Lubinski, 2010), can vary depending on gender (Battista, 1990; Erkek, Işıksal, Hatay Mustafa Kemal University, Turkey
& Çakıroğlu, 2017; Ganley & Vasilyeva, 2011; Hacıömeroğlu & Hacıömeroğlu,
2017; Yenilmez & Kakmacı, 2015; Yıldırım Gül & Karataş, 2015), age and spatial
experience (Salthouse, Babcock, Skovronek, Mitchell, & Palmon, 1990).
335
(P. 335-348) ISSN 2538–7138 /Online/
Spatial ability is an important factor that can affect success in the fields of mathematics and geometry
(Battista, 1990; Bulut & Köroğlu, 2000; Guay & McDaniel, 1977; Tartre, 1990; Ünlü & Ertekin, 2017), engineering
(Sorby, 1999; Sorby & Baartmans, 2000; Yue, 2008), chemistry (Bodner & Guay, 1997; Carter, Larussa, & Bodner,
1987; Pribyl & Bodner, 1987; Wu & Shah, 2004), physics (Delialioğlu & Aşkar, 1999; Kozhevnikov, Motes, & Hegarty,
2007), geology (Kali & Orion, 1996; Uttal & Cohen, 2012), architecture (Arslan & Dazkir, 2017), health, medicine
and dentistry (Hegarty, Keehner, Khooshabeh, & Montello, 2009; Nguyen, Mulla, Nelson, & Wilson, 2014). Spatial
ability which can affect success in various fields and can lead to significant individual differences in learning
environments is often neglected in educational and instructional settings (Lubinski, 2010).
Science, mathematics, and engineering are among the fields where spatial ability receives the biggest
emphasis for success. Similarly, spatial ability plays a critical role for student achievement in Science, Technol-
ogy, Engineering, and Mathematics (STEM) education, which is an innovative teaching approach (Stieff & Uttal,
2015; Uttal & Cohen, 2012; Wai, Lubinski, & Benbow, 2009). Developing spatial abilities of students studying in
STEM education, one of the popular educational approaches of today, can increase their academic achievement
(Stieff & Uttal, 2015).
Even if spatial ability that is critical to success in many fields is considered as an inborn characteristic,
whether the spatial ability can be changed or to what extent it can be changed by the environment is one of the
important research topics of interest (Ben-Chaim, Lappan, & Houang, 1988). This situation attracts the interest
of many researchers and leads to many studies to be done on the development of spatial ability.
According to many studies in the literature, it is possible to develop spatial abilities through appropriate
training activities (e.g. Alkan & Erdem, 2011; Baki, Kösa, & Güven, 2011; Demirkaya & Masal, 2017; Güven & Kösa,
2008; Kösa & Kalay, 2018; Kösa & Karakuş, 2018; Kurtuluş & Uygan, 2010; Olkun, 2003; Rafi, Anuar, Samad, Hayati,
& Mahadzir, 2005; Sorby, 1999; Sorby & Baartmans, 2000; Uygan & Kurtuluş, 2016).
Computer-aided 3D design and modeling programs, which have become widespread in recent years, offer
significant opportunities for the development of spatial ability. It has been reported that the 3D modeling activi-
ties using SketchUp software have improved the spatial abilities of engineering students (Martín-Dorta, Saorín,
& Contero, 2008), pre-service mathematics teachers (Kurtuluş & Uygan, 2010; Uygan & Kurtuluş, 2016), 8th grade
students (Toptaş, Çelik, & Karaca, 2012), the 3D modeling and design activities using Tinkercad software have
improved the spatial abilities of 6th grade students (Dere, 2017) and design activities using AutoCAD software
have improved the spatial abilities of engineering students (Kösa & Karakuş, 2018) in researches conducted to
improve spatial abilities.
There is a close interaction between the computer-aided 3D design and modeling process and the spatial
ability (Huang, Chen, & Lin, 2019). It is because various perspectives are employed in the modeling process car-
ried out through computer-aided 3D modeling software and 3D objects may require formative operations such
as scaling, rotation, cutting, and bending.
3D modeling, which is the process of developing the mathematical representation of a three-dimensional
object through special software (Spallone, 2015), is widely used in the fields of cinema, advertising, medicine,
industry, engineering, forensics, architecture, games, culture and education (O’Malley, 2015). The widespread
use of 3D printer technology in industry and education settings (Huang et al., 2019; Kuzu Demir, Çaka, Tuğtekin,
Demir, İslamoğlu, & Kuzu, 2016) and decreased costs of 3D printing technologies in recent years push up the
interest in 3D design and modeling technologies (Huang & Lin, 2017; Huang et al., 2019). Also, as a result of the
developments in the field of informatics since the 1980s, computer-aided 3D modeling and design courses have
been given place in higher education curricula (Huang & Lin, 2017; Varinlioğlu, Alaçam, Başarır, Genca, & Üçok,
2016; Yue, 2008). Similarly, there is an increasing interest in 3D modeling in Turkey and 3D modeling courses
are introduced to curricula for associate degree and undergraduate degree programs related to technical fields
and design. Finally, Turkish Council of Higher Education (CoHE) as the body regulating Turkish higher education
institutions, updated the curriculum of teacher training programs in 2018. In this scope, “Modeling and Design
in Education” course related to 3D modeling has been added to the curriculum of the Computer Education and
Instructional Technologies program (CoHE, 2018).
Teaching innovative thinking skills among the 21st century skills (Ananiadou & Claro, 2009; Binkley, Erstad,
Herman, Raizen, Ripley, Miller-Ricci, & Rumble, 2012; Voogt & Roblin, 2012) is one of the goals of contemporary
education systems. Students can design original and innovative products through 3D modeling and design
software and convert their designs into tangible concrete objects via 3D printers. For this reason, 3D modeling
336
ISSN 1648–3898 /Print/ THE EFFECT OF COMPUTER-AIDED 3D MODELING ACTIVITIES ON PRE-SERVICE TEACHERS’
ISSN 2538–7138 /Online/ (P. 335-348)
and design technologies can also offer great opportunities to gain innovative thinking skills to the students
(Huang et al., 2019; Kostakis, Niaros & Giotitsas, 2015). In addition, 3D modeling and design education has
begun to occupy an important place in STEM education and Maker movement, which are innovative teaching
approaches (Bicer, Nite, Capraro, Barroso, Capraro, & Lee, 2017; Bull, Chiu, Berry, Lipson, & Xie, 2014; Taylor, 2016).
Problem of Research
Developing the spatial abilities of pre-service teachers and organizing activities to improve their spatial abilities
can help pre-service teachers to contribute more to their students in their future professional lives (Özcan, Akbay,
& Karakuş, 2016). Spatial ability is regarded as an indispensable factor for success in STEM, Science, Technology,
Engineering, Art and Mathematics (STEAM), and Maker movement trainings. One main goal of the STEM, STEAM,
and Maker movement trainings is to teach innovative thinking skills to students. Bearing this in mind, it seems to
be a promising topic of interest to address development of the spatial abilities and attitudes towards 3D modeling
of pre-service teachers who will be able to assume big responsibilities in those trainings.
The review of the related literature yielded only few studies examining effects of 3D modeling activities using
3D modeling software on participants’ spatial abilities and attitudes towards 3D modeling.
The aim of this research was to examine the effects of computer-aided 3D modeling activities on the spatial
abilities of pre-service information technologies (IT) teachers and their attitudes towards 3D modeling and 3D
modeling course.
The research was carried out in search of answers to the following questions:
a) Do computer-aided 3D modeling activities lead to a meaningful improvement in the spatial abilities
of the pre-service teachers?
b) Do computer-aided 3D modeling activities lead to a meaningful improvement in the attitudes of the
pre-service teachers towards 3D modeling and the relevant course?
General Background
This research tested the hypothesis that computer-aided 3D modeling activities positively affect the spatial
abilities of participants and their attitudes towards 3D modeling. For this reason, the post-positivist paradigm was
adopted to carry out the research which employed experimental design and quantitative data were collected using
reliable measurement tools. This was an experimental research based on quantitative methodology. Experimental
researches are performed in order to explain and predict the possible relationship between dependent and in-
dependent variables by manipulating independent variables in the supervision of the researcher (Creswell, 2012;
Karasar, 2016). A control group was used for comparisons with the experimental group. Since the participants
could not be assigned to groups randomly, quasi-experimental design was used in the research (Creswell, 2012).
This research was conducted in the spring semester of 2016-2017 academic year. The research was carried out for
one semester (14 weeks).
Participants
The minimum sample size required for research was calculated using the G*power program: the input param-
eters were the tail=one, the effect size d= .70, p = .05, power = .80. As a result, the minimum number of participants
for each group was found as 26. Participants of the research were 55 pre-service IT teachers studying at Computer
Education and Instructional Technology (CEIT) program of a state university. The experimental group (EG) consisted
of 27 fourth-year pre-service IT teachers taking the elective course in which 3D modeling activities were carried
out. The control group (CG) consisted of 28 third-year pre-service IT teachers who were not enrolled in the elective
course. The participants included 30 male and 25 female pre-service IT teachers. Prior to the implementation, the
participants were informed about the purpose of the research and they all participated on voluntary basis. The
volunteer participants were then assured about the confidentiality of their answers and use of the data for scientific
purposes only. The ROT test was applied to all pre-service IT teachers studying at CEIT program for the selection
337
(P. 335-348) ISSN 2538–7138 /Online/
and determination of the CG. When the ROT test scores obtained were analyzed, it was seen that the mean score
of the third-year pre-service teachers’ group was the closest one to the mean score of the EG. Since courses taken
by these two groups were a lot similar than others, the third-year and the fourth-year pre-service teachers were
decided to match by taking into account ROT test results.
Instruments
Two data collection tools were used in the research. These tools were “Purdue Visualization of Rotations
Test (ROT)” and “Attitude Scale towards 3D Modeling and 3D Modeling Course (AS3DM)”.
The ROT was a shorter version of the Purdue Spatial-Visualization Test: Visualization of Rotations (PSVT: R)
which was developed by Guay in 1976. The original PSVT: R had 30 multiple-choice items (Maeda & Yoon, 2013).
The ROT test used in this research was constructed by removing 10 questions from the PSVT: R by Bodner and
Guay (1997).
The shorter version ROT, a paper-and-pencil test, consisted of 20 multiple-choice items. The purpose of the
test was to measure participants’ abilities to comprehend and visualize the rotation of 3D objects mentally. In this
research, the ROT test which was constructed by Bodner and Guay (1997) and translated into Turkish by Yılmaz
(2012) was used to measure the spatial ability of the participants after obtaining the necessary permissions.
Each item in the ROT test consisted of three parts. In part one, the respondent was expected to understand
the relation between the given object and rotated version of the object in space. Part two required the respon-
dent to visualize rotation of the given object in the same way as the object given in the previous part. In the
last part, it was required to choose the correct rotation among five options.
In responding to the ROT, being quick was equally important to giving the correct answer. During the
implementation of the test, the participants were given 10 minutes to respond to 20 questions, as recommended
by Bodner and Guay (1997). The responses were evaluated against the answer key sent by Bodner. Each cor-
rect answer was scored as 1 and each wrong and blank answer as 0. The scores of the test ranged from 0 to 20.
The validity of ROT test has been validated in most studies so far. Bodner and Guay (1997) reported the reli-
ability coefficients in the range of 0.78 and 0.85. The ROT test has been used in many studies to measure spatial
ability (e.g. Akıllı & Seven, 2014; Alkan & Erdem, 2011; Anvari, Tran, & Kavakli, 2013; Battista, 1990; Brownlow,
McPheron, & Acks, 2003; Brudigam & Crawford, 2012; Carter et al., 1987; Harris, Peck, Colton, Morris, Neto, & Kal-
lio, 2009; Karaçöp & Doymuş, 2013; Morgil, Yavuz, Oskay, & Arda, 2005; Pribyl & Bodner, 1987; Poulin, O’Connell,
& Freeman, 2004; Unal, Jakubowski, & Corey, 2009; Wang, Chang, & Li, 2007).
“Attitude Scale towards 3D Modeling and 3D Modeling Course (AS3DM)” was used to examine pre-service
teachers’ attitudes towards 3D modeling and 3D modeling course. The scale was developed by Benzer (2018)
with both validity and reliability verified. The scale had three factors as “Importance”, “Interest”, and “Anxiety”.
It was a 5point Likert type scale with responses as “Strongly Agree”, “Agree”, “Neutral”, “Disagree” and “Strongly
Disagree”. Of the items, 16 were positive statements, while the rest of 14 were reversely stated. The overall scale
yielded the reliability coefficient of 0.94. Specifically, the sub-scales were found to be reliable at 0.90, 0.91, and
0.82, respectively (Benzer, 2018).
Procedure
Before carrying out the main research, a pilot research was conducted on 3D modeling activities and data col-
lection tools at the vocational college of a state university. In the planning of the activities to be carried out during
the implementation process, it was taken advantage of two domain experts` views, reference books regarding 3D
modeling, 3D modeling videos shared on online video sharing websites.
The main implementation was carried out in the computer laboratory for 14 weeks, three hours per week as
an elective course. The aim of this course was to teach students how to work with 3D objects, develop 3D designs,
models and animations, and how to perform tasks related to materials, mapping, camera and lighting on models
they constructed. According to Huang et al. (2019), the aim of the 3D modeling course in technical and vocational
education is to integrate theories and practices so that students can use the knowledge they learned to produce
work. The course content is presented in Table 1.
338
ISSN 2538–7138 /Online/ (P. 335-348)
Table 1. Course content.
Week Topics
1 Introduction to 3D modelling
2 User Interface
3 3D objects and their properties
7 3D animation
8 Modeling techniques
9 Modeling techniques
10 Cameras
11 Rendering
12 Materials and Mapping
13 Materials and Mapping
14 Lighting

During the implementation, architectural modeling was practiced by means of table, chair, vase, house, sofa
set, and building modeling activities. As a part of the implementation, 3D modeling activities were realized with
the aid of 3ds Max software, commercial software by Autodesk. The company offers students a three-year license of
3ds Max free of charge for educational purposes. The software enables 3D modeling through imaging of an object
from the top, right, left, front and other viewpoints on the planes x, y, and z in space.
Kozma (1994) pointed out that using the media and appropriate teaching methods together can positively af-
fect learning. Therefore, it is important that computer-aided 3D modeling and design teaching should be performed
with useful teaching methods and techniques effectively and efficiently to meet the needs of 21st-century learn-
ers. Teaching and activities during the implementation period were built upon problem-based and project-based
learning approaches, which are contemporary learning approaches. Problem-based and project-based learning
are student-centered and teacher-supported innovative approaches in which students are responsible for their
own learning and can manage their own learning process in collaboration, learning takes place around real-life
problems, and necessary learning opportunities are extended for the 21st century (Bell, 2010; Blumenfeld, Soloway,
Marx, Krajcik, Guzdial, & Palincsar, 1991; Kocaman-Karoğlu, 2016; Ravitz, Hixson, English, & Mergendoller, 2012). Such
models of learning help students combine theoretical knowledge with practical skills (Savery, 2006). Problem-based
and project-based learning hold the potential to increase students’ achievement and motivation for learning issues
and environments, and also foster their interests and attitudes (Bell, 2010; Demirel & Dagyar, 2016; Hmelo-Silver,
2004; Hung, Hwang, & Huang, 2012; Musa, Mufti, Latiff, & Amin, 2011; Kaufman & Mann, 1997; Sezgin-Selcuk, 2010).
When students experience success, they may become more likely to develop positive attitudes and self-con-
fidence towards teaching (Usta, 2016). To assure a sense of accomplishment for all of the pre-service teachers and
the planning of the teaching and activities for an effective and efficient learning-teaching process, a method was
adopted, which was stepwise (Chien, 2006), simple to complex (Olkun, 2003), got more difficult gradually (Martín-
Dorta et al., 2008), was inductive (Dere, 2017), allowed cooperation (Martín-Dorta et al., 2008), and held learning in the
master-apprentice framework (cognitive apprenticeship strategy) (Huang et al., 2019). Apart from that, the teaching
process was divided into stages. The stages followed for the teaching process were classified and determined by
benefiting from the research of Justi and Gilbert (2002). Justi and Gilbert (2002) divided the competence levels of
the modeling process related to the field of science into five stages. The education framework used in this research
consisted of five stages, as shown in Figure 1. The stages of the teaching/learning process are described below.
1. Learning prior knowledge: This stage covered the process of teaching to learners the basic knowledge
about the topic and the ability to use the interface of the software. Knowledge was usually provided
by the instructor.
2. Learning to use 3D models: This stage was dedicated to demonstrating the construction steps of the
339
(P. 335-348) ISSN 2538–7138 /Online/
pre-planned 3D models by the instructor, creating of the demonstrated 3D models by pre-service teach-
ers (Demonstration-practice teaching) (Huang et al., 2019) and discussing differentiated methods for
building 3D models. The primary objective of this stage was to make sure that each pre-service teacher
successfully creates the 3D models previously demonstrated. The 3D model (problem) and how to create
it (answer) were provided by the instructor. At this stage, the instructor was responsible for showing
the construction phases of the 3D models, supporting the pre-service teachers in designing the 3D
models within master-apprentice relationship (cognitive apprenticeship), guiding them, encouraging
the pre-service teachers to have conversations, supporting cooperation among them, and increasing
the motivation of the pre-service teachers.
3. Learning how to revise 3D models: This stage involved the process in which students made some revi-
sions (formal changes) on the 3D model that was constructed in the previous stage. The main objective
of this stage was to enable each pre-service teacher to successfully make the desired revisions on the
3D models. Knowledge of how to make revisions (answer) was partially provided by the instructor. At
this stage, the instructor undertook a similar role to the one at the previous stage.
4. Learning to reconstruct 3D models: At this stage, problem-based learning method was used. The
stage involved the process of combining (Huang et al., 2019) and reconstructing the 3D modeling tasks
performed in the previous lessons and transforming them into complex, advanced and ill-structured
modeling problems. Ill-structured problems were explained but their answers were not provided by
the instructor. At this stage, pre-service teachers applied their previous 3D modeling knowledge and
skills for solving the newly given 3D modeling problems so that their problem-solving skills could be
improved, and they could learn effectively. At this stage, the role of the instructor was to give feedback
to the pre-service teacher on their performances and to assist them to solve the problems they face in
cooperation with their peers. However, the instructor avoided directly showing the solution leading
to the answer.
5. Learning to construct new 3D model: This stage was closely related to project-based learning method.
Towards the end of the semester, the participant pre-service teachers, in addition to the in-class ac-
tivities, were assigned a project in which they were supposed to model a three-storey architectural
structure by using all of the modeling skills and knowledge acquired in class. It helped to make sure
that the pre-service teachers applied all the knowledge and skills they learned about 3D modeling to
the 3D modeling project, the learning became more permanent, and their innovative thinking skills
flourished. At this stage, the role of the instructor was to give feedback to the pre-service teachers on
their project performances and to help them solve the problems faced in cooperation with their peers.
Figure 1. The education framework.
340
ISSN 2538–7138 /Online/ (P. 335-348)
Most of the time, the first three stages were combined in one week and stage four in the following week or
vice versa. As for the last stage, the participants were then instructed to carry out works related to 3D architectural
modeling by reverting to all of their modeling skills and knowledge towards the end of the semester, in addition
to in-class activities.
Before and after the treatment, the EG was administered the ROT test and AS3DM. However, only the ROT
test was administered to the CG. The pre-service teachers were given 10 minutes to solve 20 multiple-choice items
in the ROT test in the paper-pencil format. The data collection tools were administered by the same researcher.
Data Analysis
In the ROT test, each correct answer was given 1 point, but wrong and blank answers were given 0 point. The
highest and the lowest test scores were 20 and 0, respectively. The AS3DM was a Likert-type scale which ranked
positive items from 5 to 1, while negative items were scored from 1 to 5. In this research, collected data were
analyzed by using SPSS. In order to determine the statistical methods to be used, pre-test and post-test data were
tested for normal distribution. In this scope, the Shapiro-Wilk normality test results were examined. It was found
out that all the test scores except the AS3DM pre-test scores (p <.05) showed normal distribution (p >.05). The ROT
test scores with normal distribution were analyzed with the paired samples t-test and the independent samples
t-test among parametric tests; whereas non-normally distributed the AS3DM pre-test scores were analyzed with
non-parametric Wilcoxon signed rank test.
Research Results
In this section, the ROT and AS3DM scores obtained by the pre-service teachers are presented to put forward
the change in their spatial ability and attitudes before and after the treatment. Table 2 presents the mean and
standard deviation values of the EG and CG scores obtained from the ROT pre-test and the independent samples
t-test results.
Table 2. Results of descriptive statistics and paired samples t-test for ROT pre-test scores of EG and CG.
Group N M SD df t p
EG 27 9.33 3.56
53 .199 .843
CG 28 9.14 3.52
Table 2 shows that the EG had a mean score of M=9.33 and the CG had M=9.14 in the spatial ability test
administered at the beginning of the research. As also seen in Table 2, the independent samples t-test yielded no
significant difference between the EG and CG based on the ROT pre-test scores [t (53) = .199, p > .05]. It can be said
that both groups were equivalent in the spatial ability test before the treatment.
Table 3 shows the results of the dependent samples t-test, which was administered to calculate the mean
scores of the EG in the ROT before and after the treatment, the standard deviation values, and the statistically
significant difference, if any, between the mean scores.
Table 3. Results of descriptive statistics and paired samples t-test for ROT pre-post test scores of the EG.
Test N M SD df t p
ROT Pre-test 27 9.33 3.56

26 -4.256 .0001
ROT Post-test 27 12.07 3.63
It is seen in Table 3 that the ROT pre-test mean score of the EG was M=9.33. After the treatment, the ROT was
re-administered as post-test where the pre-service teachers’ mean score increased to M=12.07. A statistically sig-
nificant difference was noted between the ROT pre-test and ROT post-test scores of the EG in the paired samples
t-test [t(26) = -4.256, p <.01]. This result implies that computer-aided 3D modeling activities proved influential on
pre-service teachers’ spatial abilities in a positive way.
341
(P. 335-348) ISSN 2538–7138 /Online/
Table 4 shows the results of the paired samples t-test, which was administered to calculate the mean scores
of the CG in the ROT before and after the treatment, the standard deviation values, and the statistically significant
difference, if any, between the scores.
Table 4. Results of descriptive statistics and paired samples t-test for ROT pre-post test scores of the CG.
Test N M SD df t p
ROT Pre-test 28 9.14 3.52

27 -.316 .754
ROT Post-test 28 9.36 3.87
It is seen in Table 4 that the ROT pre-test mean score of the CG was M=9.14. At the end of the treatment, the ROT
was re-administered as post-test. In this test, the pre-service teachers recorded a slightly higher mean score, being
M=9.36. A statistically significant difference was not found between the ROT pre-test and ROT post-test scores of
the CG in the paired samples t-test [t (27) = - .316, p >.05]. It can be inferred that the pre-service teachers who had
not been taught through 3D modeling activities did not experience a significant increase in their spatial abilities.
Table 5 shows the mean and standard deviation values of the EG and CG scores obtained from the ROT post-
test, and the independent samples t-test results.
Table 5. Results of descriptive statistics and independent samples t-test for ROT post-test scores of the EG
and CG.
Group N M SD df t p d
EG 27 12.07 3.63
53 2.684 .010 .72
CG 28 9.36 3.87
In the spatial ability test as post-test, the mean score of the pre-service teachers in the EG was M=12.07, while
the same value was M=9.36 in the CG. When Table 3, Table 4 and Table 5 are considered together, an increase was
observed in the ROT post-test scores of the EG. As a result of the computer-aided 3D modeling activities, the EG
achieved an increase by 29.4 % in their spatial ability. Conversely, there was no significant improvement in the
spatial ability scores of the CG. As seen in Table 5, the independent samples t-test delivered a statistically significant
difference in favor of the EG in the ROT post-test scores [t (53) = 2.684, p < .05]. It seems that computer-aided 3D
modeling activities were influential on improving the pre-service teachers’ spatial abilities. The Cohen`s d value was
calculated to measure the effect size. According to the benchmarks proposed by Cohen (1988) for interpreting the
Cohen`s d value, .2 indicates a small effect, .5 a medium effect, and .8 a large effect. Given the Cohen`s d value for
t-test (.72), it can be said that the intervention had a medium effect on the spatial ability scores.
Table 6 exhibits the mean scores and standard deviation values from the AS3DM scale administered to the EG
before and after the treatment for examining their attitudes towards 3D modeling and the course.
Table 6. Results of descriptive statistics for the AS3DM scores obtained by the EG.
Test N M SD
AS3DM Pre-test 27 3.77 .48

AS3DM Post-test 27 4.09 .54
According to the results shown in Table 6, there was an increase in the AS3DM post-test mean score of the EG.
The participants in this group noted the mean score of M=3.77 in the AS3DM pre-test. At the end of the research,
the scale was re-administered. In the post-test, the pre-service teachers in the EG achieved higher average scores,
resulting in M = 4.09. Due to the non-normal distribution of AS3DM pre-test scores, the scores obtained from this
test were analyzed by using non-parametric Wilcoxon signed-rank test.
Table 7 presents results of the Wilcoxon signed-rank test, which was applied to find out significant differences,
if any, between the participants’ scores in AS3DM pre-post tests.
342
ISSN 2538–7138 /Online/ (P. 335-348)
Table 7. Results of Wilcoxon signed-rank test for the AS3DM pre-post test scores of the EG.
Post Test-Pre Test N Mean Rank Sum of Ranks z p
Negative ranks 5 15.80 79.00

Positive ranks 22 13.59 299.00 2.64* .008
Ties 0
*Based on negative ranks
As seen in Table 7, Wilcoxon signed-rank test gave away a significant increase in AS3DM post-test taken by
the EG (z= 2.64, p < .05). So, it can be said that computer-aided 3D modeling activities carried out in the five-stage
education framework based on problem-based learning and project-based learning approaches, have a positive
impact on pre-service teachers’ attitudes towards 3D modeling and 3D modeling course.
Discussion
In this research, the effect of computer-aided 3D modeling activities on the spatial abilities of pre-service IT
teachers and their attitudes towards 3D modeling was examined. The research was planned as an experimental
research involving an experimental and a control group. The research data were collected by means of “Purdue
Visualization of Rotations Test (ROT)” consisting of 20 questions and “Attitude Scale towards 3D Modeling and 3D
Modeling Course” with 30 items. While the EG used both of the instruments at the beginning and at the end of the
treatment, the CG was administered only by the ROT test. The intervention lasted for 14 weeks.
The results obtained at the end of the research were summarized under the two main headings below with
reference to comparable studies.

The Effect of 3D Modeling Activities on the Spatial Ability
In this research, it was found out that 3D modeling activities significantly improved the spatial abilities of
the pre-service teachers in the EG. On the other hand, there was no such increase in the spatial abilities of the CG,
who had not gone through the same process of learning. This suggests that 3D modeling activities are effective
in developing spatial abilities. However, it was observed that the curriculum courses taken by the CG students
throughout the semester were not effective in developing their spatial abilities. This indicates the need that train-
ings targeting the spatial ability should include goal-oriented, well-planned, and specific activities.
It is possible to divide 3ds Max software interface used for 3D modeling activities divided into several windows
that allow simultaneous viewing of 3D objects from different viewpoints such as top, right, left, and perspective.
In this way, 3D objects can be viewed from different perspectives and changes on the object can be monitored
simultaneously in different windows. Perspectives of objects can be two-dimensional such as from top, front, and
left; or they can give three-dimensional views like a perspective view. The boosting effect of 3D modeling activities
on the students’ spatial abilities could be explained with the fact that the students had to perform the manipulat-
ing operations such as rotating, moving, and changing dimensions of 3D objects on the axes x, y, and z through
various points of views (viewing windows) for several times before completing the 3D modeling assignments. Most
researchers have studied the concept of spatial ability in two parts as spatial visualization and spatial orientation.
Spatial visualization is the ability to manipulate, rotate, bend, or invert visual objects in the mind, whereas spatial
orientation refers to the ability not to confuse different orientations of visual object(s) or to properly understand the
order of such objects given in a specific pattern (McGee, 1979). Spatial orientation tasks require the understanding
of the change and representation between two objects, rather than mentally rotating the objects (Tartre, 1990). In
view of this classification and definitions, 3D modeling activities performed on 3ds Max seem to present marked
opportunities to improve both of the sub- abilities referred above.
The literature lends abundance of studies in compliance with findings of the research in that computer-aided
3D drawing, design, and modeling activities develop students’ spatial abilities (e.g. Dere, 2017; Güven & Kösa, 2008;
Kösa, 2016; Kösa & Karakuş, 2018; Kösa & Kalay, 2018; Kurtuluş & Uygan, 2010; Martín-Dorta et al., 2008; Šafhalter,
Vukman, & Glodež, 2016; Toptaş et al., 2012; Uygan & Kurtuluş, 2016). However, there is also a research (Shavalier,
2004) which could not prove the effect of computer-aided design and modeling activities on the spatial ability.
343
(P. 335-348) ISSN 2538–7138 /Online/
The Effect of 3D Modeling Activities on Attitude toward 3D Modeling
In the practical part of this research, it was found out that the five-stage computer-aided 3D modeling
education using problem-based and project-based learning approaches led to a meaningful improvement in the
attitudes of the pre-service teachers towards 3D modeling and the course. The education was started with the
modeling activities which excluded high-level skills and were simple enough for all participants to taste the sense
of achievement. Thanks to this choice, the pre-service teachers could not only arouse self-confidence towards
3D modeling and 3D modeling course but also enhance their motivation and upgrade their attitudes towards
3D modeling. Furthermore, the fact that a substantial part of the teaching was fulfilled within the framework of
master-apprentice relationship, peer collaboration was promoted, and the participant pre-service teachers were
given feedback increasing their motivation and morale might have played a role in improving their attitudes. Last
but not least, the desired outcome might be partially accounted for by the five-stage education framework utiliz-
ing problem-based and project-based learning approaches as it bestowed an active part to the participants in the
modeling activities, engaged them in the process of learning by doing, living, and designing, and motivated them
to develop new products with their personal knowledge and experience.
All in all, the findings of this research in the matter of the effects of 3D modeling and design activities on the
attitudes of the participants are in conformity with Dere (2017), Halıcı, Turhan, Aksu, and Varinlioğlu (2017), Huang
et al. (2019), Martín-Dorta et al. (2008) and Shavalier (2004).
Conclusions
It is thought that 3D design and modeling teaching is likely to pose important opportunities for acquisition of
“innovative thinking” skill among the 21st century skills. Also, it might pave the way importantly for development
of spatial abilities, which are of critical importance for success in a number of fields such as science, technology,
engineering, arts and mathematics.
The International Society for Technology in Education (ISTE) has set standards to support educators to be ef-
fective in learning settings using technology. According to ISTE standards for teachers, it is necessary and important
for 21st century teachers to have necessary competencies to facilitate learning, improve innovative thinking skills
of their students and develop the technology-based activities by using their knowledge of subject matter, peda-
gogical, and technology (ISTE, 2008). It is considered that spatial ability is an important factor for success in STEM,
STEAM and Maker movement trainings and that one main goal of these trainings is to acquire innovative thinking
skills to students. IT teachers might undertake major duties on STEM, STEAM, and Maker trainings. Therefore, it
seems useful to give pre-service IT teachers trainings to develop the spatial abilities and give them proficiency and
positive attitudes regarding 3D modeling. The 3D modeling education can help pre-service teachers to educate
students equipped with 21st century skills in their future careers.
Within the scope of the research, it was found that computer-aided 3D modeling activities positively affect the
spatial abilities of pre-service IT teachers. Contrarily, there was no significant increase in the spatial abilities of the
students who did not attend the 3D modeling classes. Another result reached in this research is that the five-stage
computer-aided 3D modeling education using problem-based learning and project-based learning approaches was
proven to have a positive effect on pre-service teachers’ attitudes towards 3D modeling and 3D modeling course.
References
Akıllı, M., & Seven, S. (2014). 3D bilgisayar modellerinin akademik başarıya ve uzamsal canlandırmaya etkisi: Atom modelleri [The
effects of 3D computer models to academic achievement and spatial ability: Atomic models]. Turkish Journal of Education,
3(1), 11-23.
Alkan, F., & Erdem, E. (2011). A study on developing candidate teachers’ spatial visualization and graphing abilities. Procedia-
Social and Behavioral Sciences, 15, 3446-3450.
Ananiadou, K., & Claro, M. (2009). 21st century skills and competences for new millennium learners in OECD Countries, OECD Educa-
tion Working Papers, No. 41, OECD Publishing. http://dx.doi.org/10.1787/218525261154.
Anvari, F., Tran, H. M. T., & Kavakli, M. (2013). Using cognitive load measurement and spatial ability test to identify talented
students in three-dimensional computer graphics programming. International Journal of Information and Education Tech-
nology, 3(1), 94-99.
Arslan, A. R., & Dazkir, S. S. (2017). Technical drafting and mental visualization in interior architecture education. International
Journal for the Scholarship of Teaching and Learning, 11(2), Article 15.
344
ISSN 2538–7138 /Online/ (P. 335-348)
Baki, A., Kösa, T., & Güven, B. (2011). A comparative study of the effects of using dynamic geometry software and physical ma-
nipulatives on the spatial visualisation skills of pre-service mathematics teachers. British Journal of Educational Technology,
42(2), 291-310.
Battista, M. T. (1990). Spatial visualization and gender differences in high school geometry. Journal for Research in Mathematics
Education, 21(1), 47-60.
Bell, S. (2010). Project-based learning for the 21st century: Skills for the future. The Clearing House, 83(2), 39-43.
Ben-Chaim, D., Lappan, G., & Houang, R. T. (1988). The effect of instruction on spatial visualization skills of middle school boys
and girls. American Educational Research Journal, 25(1), 51-71.
Benzer, A.İ. (2018). Bilgisayar destekli üç boyutlu modelleme dersi etkinliklerinin öğrencilerin uzamsal yetenekleri ve derse yönelik
tutumları üzerine etkisi [The effect of computer aided three dimensional modeling course activities on students` spatial
abilities and attitudes towards the course]. Unpublished doctoral dissertation, Mustafa Kemal University, Hatay, Turkey.
Bicer, A., Nite, S. B., Capraro, R. M., Barroso, L. R., Capraro, M. M., & Lee, Y. (2017, October). Moving from STEM to STEAM: The effects
of informal STEM learning on students’ creativity and problem solving skills with 3D printing. Paper presented at the IEEE
Symposium on Frontiers in Education Conference (FIE), Indianapolis, IN, USA.
Binkley, M., Erstad, O., Herman, J., Raizen, S., Ripley, M., Miller-Ricci, M., & Rumble, M. (2012). Defining twenty-first century skills.
In P. Griffin, B. McGaw, & E. Care (Eds.), Assessment and teaching of 21st century skills (pp. 17–66). Heidelberg: Springer.
Bishop, A. J. (1980). Spatial abilities and mathematics education - A review. Educational Studies in Mathematics, 11(3), 257-269.
Bodner, G. M., & Guay, R. B. (1997). The Purdue visualization of rotations test. The Chemical Educator, 2(4), 1-17.
Brownlow, S., McPheron, T. K., & Acks, C. N. (2003). Science background and spatial abilities in men and women. Journal of Science
Education and Technology, 12(4), 371-380.
Brudigam, K. L., & Crawford, R. H. (2012, June). Spatial ability in high school students. Paper presented at 2012 ASEE Annual Confer-
ence & Exposition, San Antonio, Texas. https://peer.asee.org/21928.
Bulut, S., & Köroğlu, S. (2000). Onbirinci sınıf öğrencilerinin ve matematik öğretmen adaylarının uzaysal yeteneklerinin incelen-
mesi [Investigation of spatial abilities of 11th grade students and mathematics prospective teachers]. Hacettepe University
Journal of Education, 18, 56-61.
Blumenfeld, P. C., Soloway, E., Marx, R. W., Krajcik, J. S., Guzdial, M., & Palincsar, A. (1991). Motivating project-based learning:
Sustaining the doing, supporting the learning. Educational Psychologist, 26(3-4), 369-398.
Bull, G., Chiu, J., Berry, R., Lipson, H., & Xie, C. (2014). Advancing children’s engineering through desktop manufacturing. In J.M.
Spector, M.D. Merrill, J. Elen, & M.J. Bishop (Eds.) Handbook of research on educational communications and technology (pp.
675-688). New York: Springer.
Carter, C. S., Larussa, M. A., & Bodner, G. M. (1987). A study of two measures of spatial ability as predictors of success in different
levels of general chemistry. Journal of Research in Science Teaching, 24(7), 645-657.
Chien, J. J. (2006). A study on basic level 3D MAYA animation instruction in college: An action research approach. The Inter-
national Journal of Arts Education, 4(1), 158-181. Retrieved 12/12/2018, from http://ed.arte.gov.tw/uploadfile/periodi-
cal/1328_arts_education41_158181.pdf.
Clements, D. H. (1998). Geometric and spatial thinking in young children, (ERIC Servis No. ED436232). Retrieved 12/10/2018,
from http://files.eric.ed.gov/fulltext/ED436232.pdf.
CoHE, (2018). Bilgisayar ve Öğretim Teknolojileri Öğretmenliği Lisans Programı [Computer Education and Instructional Tech-
nology Undergraduate Program]. YÖK, Retrieved 12/10/2018, from http://www.yok.gov.tr/documents/10279/41805112/
Bilgisayar_ve_Ogretim_Teknolojileri_Ogretmenligi_Lisans_Programi.pdf.
Cohen, J. (1988). Statistical power analysis for the behavioural sciences. Hillsdale, NJ: Lawrence Erlbaum.
Creswell, J. W. (2012). Educational research: Planning, conducting, and evaluating quantitative and qualitative research. 4th Ed.
Boston, MA: Pearson.
Delialioğlu, Ö., & Aşkar, P. (1999). Contribution of students’ mathematical skills and spatial ability to achievement in secondary
school physics. Hacettepe University Journal of Education, 16, 34-39.
Demirel, M., & Dağyar, M. (2016). Effects of problem-based learning on attitude: A Meta-analysis study. Eurasia Journal of Math-
ematics, Science & Technology Education, 12(8), 2115-2137.
Demirkaya, C., & Masal, M. (2017). Geometrik-mekanik oyunlar temelli etkinliklerin ortaokul öğrencilerinin uzamsal düşünebilme
becerilerine etkisi [The effect of geometric-mechanic games based activities on the spatial skills of secondary school
students]. Sakarya University Journal of Education, 7(3), 600-610.
Dere, H. E. (2017). Web tabanlı 3B tasarım uygulamalarının ortaokul öğrencilerinin uzamsal görselleştirme ve zihinsel döndürme
becerilerine etkisi [The effects of using web based 3D design applications on spatial visualisation and mental rotation abili-
ties of middle school students]. Unpublished master’s thesis, Başkent University, Ankara, Turkey.
Erkek, Ö., Işıksal, M., & Çakıroğlu, E. (2017). A study on pre-service teachers’ spatial visualization ability and spatial anxiety.
Kastamonu Education Journal, 25(1), 33-50.
Friedman, L. (1992). A Meta-analysis of correlations of spatial and mathematical tasks. ERIC document reproduction service no.
ED 353270 Retrieved 11/10/2018, from https://files.eric.ed.gov/fulltext/ED353270.pdf.
Ganley, C. M., & Vasilyeva, M. (2011). Sex differences in the relation between math performance, spatial skills, and attitudes.
Journal of Applied Developmental Psychology, 32(4), 235-242.
Gardner, H., (2011). Frames of mind: The theory of multiple intelligences. New York: Basic Books.
Guay, R. B., & McDaniel, E. D. (1977). The relationship between mathematics achievement and spatial abilities among elementary
school children. Journal for Research in Mathematics Education, 8(3), 211-215.
345
(P. 335-348) ISSN 2538–7138 /Online/
Güven, B., & Kösa, T. (2008). The effect of dynamic geometry software on student mathematics teachers’ spatial visualization
skills. The Turkish Online Journal of Educational Technology (TOJET), 7(4), 100-107.
Hacıömeroğlu, G., & Hacıömeroğlu, E. S. (2017). Cinsiyet, uzamsal beceri, mantıksal düşünme becerisi ve çözüm tercihleri arasındaki
ilişkinin incelenmesi [Examining the relationship between gender, spatial ability, logical reasoning ability, and preferred
mode of processing]. Adıyaman University Journal of Educational Sciences, 7(1), 116-131.
Halıcı, S. M., Turhan, G. D., Aksu, M. S., & Varinlioğlu, G. (2017, June). Uzay mimarlığında sayısal tasarım ve üretim araçlarının
değerlendirilmesi üzerine Mars özelinde bir çalışma [Digital design and fabrication tools in space architecture workshop].
Paper presented at 11th Computational Design in Architecture National Symposium, Ankara, Turkey.
Harris, M. A., Peck, R. F., Colton, S., Morris, J., Neto, E. C., & Kallio, J. (2009). A combination of hand-held models and computer
imaging programs helps students answer oral questions about molecular structure and function: A controlled investiga-
tion of student learning. CBE-Life Sciences Education, 8(1), 29-43.
Hegarty, M., Keehner, M., Khooshabeh, P., & Montello, D. R. (2009). How spatial abilities enhance, and are enhanced by, dental
education. Learning and Individual Differences, 19(1), 61-70.
Hmelo-Silver, C. E. (2004). Problem-based learning: What and how do students learn? Educational Psychology Review, 16(3), 235-266.
Huang, T. C., & Lin, C. Y. (2017). From 3D modeling to 3D printing: Development of a differentiated spatial ability teaching model.
Telematics and Informatics, 34(2), 604-613.
Huang, T. C., Chen, M. Y., & Lin, C. Y. (2019). Exploring the behavioral patterns transformation of learners in different 3D modeling
teaching strategies. Computers in Human Behavior, 92, 670-678.
Hung, C. M., Hwang, G. J., & Huang, I. (2012). A Project-based digital storytelling approach for improving students’ learning
motivation, problem-solving competence and learning achievement. Educational Technology & Society, 15 (4), 368–379.
ISTE, (2008). ISTE standards: Teachers. ISTE, Retrieved 12/12/2018, from http://www.iste.org/docs/pdfs/20-14_ISTE_Standards-
T_PDF.pdf.
Justi, R. S., & Gilbert, J. K. (2002). Modelling, teachers’ views on the nature of modelling, and implications for the education of
modellers. International Journal of Science Education, 24(4), 369-387.
Kali, Y., & Orion, N., (1996). Spatial abilities of high-school students in the perception of geologic structures. Journal of Research
in Science Teaching, 33(4), 369-391.
Karaçöp, A., & Doymuş, K. (2013). Effects of jigsaw cooperative learning and animation techniques on students’ understanding
of chemical bonding and their conceptions of the particulate nature of matter. Journal of Science Education and Technol-
ogy, 22(2), 186-203.
Karasar, N. (2016). Bilimsel araştırma yöntemi: Kavramlar, ilkeler, teknikler (31. Basım) [Scientific research method: Concepts, prin-
ciples, techniques (31. Edition)]. Ankara: Nobel Akademik Yayıncılık.
Kaufman, D. M., & Mann, K. V. (1997). Basic sciences in problem‐based learning and conventional curricula: Students’ attitudes.
Medical Education, 31(3), 177-180.
Kocaman-Karoğlu, A. (2016). Proje tabanlı öğrenme [Project-based learning]. In K. Çağıltay & Y. Göktaş (Eds.), Öğretim teknolo-
jilerinin temelleri: Teoriler, araştırmalar, eğilimler [The basics of instructional technologies: Theories, researches, trends] (pp.
243-258). Ankara: Pegem Akademi.
Kostakis, V., Niaros, V., & Giotitsas, C. (2015). Open source 3D printing as a means of learning: An educational experiment in two
high schools in Greece. Telematics and Informatics, 32(1), 118-128.
Kozhevnikov, M., Motes, M. A., & Hegarty, M. (2007). Spatial visualization in physics problem solving. Cognitive Science, 31(4),
549-579.
Kozma, R. B. (1994). A reply: Media and methods. Educational Technology Research and Development, 42(3), 11-14.
Kösa, T. (2016). Effects of using dynamic mathematics software on preservice mathematics teachers’ spatial visualization skills:
The case of spatial analytic geometry. Educational Research and Reviews, 11(7), 449-458.
Kösa, T., & Kalay, H. (2018). 7. sınıf öğrencilerinin uzamsal yönelim becerilerini geliştirmeye yönelik tasarlanan öğrenme ortamının
değerlendirilmesi [Evaluation of the learning environment designed for developing 7th grade students’ spatial orientation
skills]. Kastamonu Education Journal, 26(1), 83-92.
Kösa, T., & Karakuş, F. (2018). The effects of computer-aided design software on engineering students’ spatial visualisation skills.
European Journal of Engineering Education, 43(2), 296-308.
Kurtuluş, A., & Uygan, C. (2010). The effects of Google Sketchup based geometry activities and projects on spatial visualization
ability of student mathematics teachers. Procedia Social and Behavioral Sciences, 9, 384-389.
Kuzu Demir, E. B., Çaka, C, Tuğtekin, U., Demir, K., İslamoğlu, H., & Kuzu, A. (2016). Üç boyutlu yazdırma teknolojilerinin eğitim
alanında kullanımı: Türkiye’deki uygulamalar [The use of three dimensional printing technologies in education: Practices
in Turkey]. Ege Journal of Education, 17(2), 481-503.
Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in spatial ability: A meta-analysis. Child
Development, 56(6), 1479-1498.
Lohman, D. F. (1996). Spatial ability and G. In I. Dennis and P. Tapsfield (Eds.), Human abilities: Their nature and measurement (pp.
97-116), Mahwah, NJ: Erlbaum.
Lubinski, D. (2010). Spatial ability and STEM: A sleeping giant for talent identification and development. Personality and Individual
Differences, 49(4), 344-351.
Maeda, Y., & Yoon, S. Y. (2013). A meta-analysis on gender differences in mental rotation ability measured by the Purdue spatial
visualization tests: Visualization of rotations (PSVT: R). Educational Psychology Review, 25(1), 69-94.
346
ISSN 2538–7138 /Online/ (P. 335-348)
Martín-Dorta, N., Saorín, J. L., & Contero, M. (2008). Development of a fast remedial course to improve the spatial abilities of
engineering students. Journal of Engineering Education, 97(4), 505-513.
McGee, M. G. (1979). Human spatial abilities: Psychometric studies and environmental, genetic, hormonal, and neurological
influences. Psychological Bulletin, 86(5), 889–918.
Morgil, I., Yavuz, S., Oskay, Ö. Ö., & Arda, S. (2005). Traditional and computer-assisted learning in teaching acids and bases. Chem-
istry Education Research and Practice, 6(1), 52-63.
Musa, F., Mufti, N., Latiff, R. A., & Amin, M. M. (2011). Project-based learning: Promoting meaningful language learning for work-
place skills. Procedia-Social and Behavioral Sciences, 18, 187-195.
Nguyen, N., Mulla, A., Nelson, A. J., & Wilson, T. D. (2014). Visuospatial anatomy comprehension: The role of spatial visualization
ability and problem-solving strategies. Anatomical Sciences Education, 7(4), 280-288.
O’Malley, S. (2015). 3D modeling and animation. Retrieved 11/12/2018, from http://um3D.dc.umich.edu/wp-content/
uploads/2015/10/3D-Modeling-and-Animation-Novice.pdf
Olkun, S. (2003). Making connections: Improving spatial abilities with engineering drawing activities. International Journal of
Mathematics Teaching and Learning, 3(1), 1-10.
Olkun, S., & Altun, A. (2003). İlköğretim öğrencilerinin bilgisayar deneyimleri ile uzamsal düşünme ve geometri başarıları arasındaki
ilişki [The relationship among elementary school students’ computer experiences, spatial thinking and geometry achieve-
ments]. The Turkish Online Journal of Educational Technology (TOJET), 2(4), 86-91.
Özcan, K. V., Akbay, M., & Karakuş, T. (2016). Üniversite öğrencilerinin oyun oynama alışkanlıklarının uzamsal becerilerine etkisi
[Effects of gaming habits of university students on their spatial skills]. Kastamonu Education Journal, 24(1), 37-52.
Pellegrino, J. W., Alderton, D. L., & Shute, V. J. (1984). Understanding spatial ability. Educational Psychologist, 19(3), 239-253.
Poulin, M., O’Connell, R. L., & Freeman, L. M. (2004). Picture recall skills correlate with 2D: 4D ratio in women but not men. Evolu-
tion and Human Behavior, 25(3), 174-181.
Pribyl, J. R., & Bodner, G. M. (1987). Spatial ability and its role in organic chemistry: A study of four organic courses. Journal of
Research in Science Teaching, 24(3), 229-240.
Ravitz, J., Hixson, N., English, M., & Mergendoller, J. (2012, April). Using project based learning to teach 21st century skills: Find-
ings from a statewide initiative. Paper presented at Annual Meetings of the American Educational Research Association,
Vancouver, Canada.
Rafi, A., Anuar, K., Samad, A., Hayati, M., & Mahadzir, M. (2005). Improving spatial ability using a Web-based Virtual Environment
(WbVE). Automation in Construction, 14(6), 707-715.
Šafhalter, A., Vukman, K. B., & Glodež, S. (2016). The effect of 3D-modeling training on students’ spatial reasoning relative to
gender and grade. Journal of Educational Computing Research, 54(3), 395-406.
Salthouse, T. A., Babcock, R. L., Skovronek, E., Mitchell, D. R., & Palmon, R. (1990). Age and experience effects in spatial visualiza-
tion. Developmental Psychology, 26(1), 128-136.
Savery, J. R. (2006). Overview of problem-based learning: Deﬁnitions and distinctions. Interdisciplinary Journal of Problem-based
Learning, 1(1), 9-20.
Sezgin-Selcuk, G. (2010). The effects of problem-based learning on pre-service teachers’ achievement, approaches and attitudes
towards learning physics. International Journal of Physical Sciences, 5(6), 711-723.
Shavalier, M. (2004). The effects of CAD-like software on the spatial ability of middle school students. Journal of Educational
Computing Research, 31(1), 37-49.
Sorby, S. (1999). Developing 3-D spatial visualization skills. Engineering Design Graphics Journal, 63(2), 21-32.
Sorby, S. A., & Baartmans, B. J. (2000). The development and assessment of a course for enhancing the 3D spatial visualization
skills of first year engineering students. Journal of Engineering Education, 89(3), 301-307.
Spallone, R. (2015). Digital reconstruction of demolished architectural masterpieces, 3D modeling, and animation: The case
study of Turin Horse Racing by Mollino. In S. Brusaporci (Ed.), Handbook of research on emerging digital tools for architectural
surveying, modeling, and representation (pp. 476-509). Hershey: IGI Global.
Stieff, M., & Uttal, D. (2015). How much can spatial training improve STEM achievement? Educational Psychology Review, 27(4),
607-615.
Tartre, L. A. (1990). Spatial orientation skill and mathematical problem solving. Journal for Research in Mathematics Education,
21(3), 216-229.
Taylor, B. (2016). Evaluating the benefit of the maker movement in K-12 STEM education. Electronic International Journal of
Education, Arts, and Science (EIJEAS), 2, 1-22.
Toptaş, V., Çelik, S., & Karaca, E. T. (2012). Improving 8th grades spatial thinking abilities through a 3D modeling program. The
Turkish Online Journal of Educational Technology (TOJET), 11(2), 128-134.
Unal, H., Jakubowski, E., & Corey, D. (2009). Differences in learning geometry among high and low spatial ability pre-service
mathematics teachers. International Journal of Mathematical Education in Science and Technology, 40(8), 997-1012.
Uttal, D. H., & Cohen, C. A. (2012). Spatial thinking and STEM education: When, why, and how? Psychology of Learning and Mo-
tivation, 57, 147-181.
Uygan, C., & Kurtuluş, A. (2016). Effects of teaching activities via Google Sketchup and concrete models on spatial skills of pre-
service mathematics teachers. Turkish Journal of Computer and Mathematics Education, 7(3), 510-535.
Usta, E. (2016). Öğretim teknolojileri ve davranışçılık [Instructional technologies and behaviorism]. In K. Çağıltay & Y. Göktaş
(Eds.), Öğretim teknolojilerinin temelleri: Teoriler, araştırmalar, eğilimler [The basics of instructional technologies: Theories,
researches, trends] (pp. 159-176). Ankara: Pegem Akademi.
347
(P. 335-348) ISSN 2538–7138 /Online/
Ünlü, M., & Ertekin, E. (2017). A structural equation model for factors affecting eighth graders’ geometry achievement. Educational
Sciences: Theory & Practice, 17(5), 1815–1846.
Varinlioğlu, G., Alaçam, S., Başarır, L., Genca, Ö., & Üçok, I. (2016). Bilgisayar destekli teknik çizimde yeni yaklaşımlar: Temsil araçları
arası dönüşüm [New approaches to computer-aided technical drawing: Inter-representational transfer]. Yapı, 419, Ekim,
137-141.
Voogt, J., & Roblin, N. P. (2012). A comparative analysis of international frameworks for 21st century competences: Implications
for national curriculum policies. Journal of Curriculum Studies, 44(3), 299-321.
Wai, J., Lubinski, D., & Benbow, C. P. (2009). Spatial ability for STEM domains: Aligning over 50 years of cumulative psychological
knowledge solidifies its importance. Journal of Educational Psychology, 101(4), 817-835.
Wang, H. C., Chang, C. Y., & Li, T. Y. (2007). The comparative efficacy of 2D-versus 3D-based media design for influencing spatial
visualization skills. Computers in Human Behavior, 23(4), 1943-1957.
Wu, H. K., & Shah, P. (2004). Exploring visuospatial thinking in chemistry learning. Science Education, 88(3), 465-492.
Yenilmez, K., & Kakmacı, Ö. (2015). Investigation of the relationship between the spatial visualization success and visual/spatial
intelligence capabilities of sixth grade students. International Journal of Instruction, 8(1), 189-204.
Yıldırım Gül, Ç., & Karataş, İ. (2015). 8. sınıf öğrencilerinin dönüşüm geometrisi başarılarının uzamsal becerileri, geometri anlama
düzeyleri ve matematiğe yönelik tutumları arasındaki ilişkinin incelenmesi [Investigation of correlation among the 8th
grade students’ achievement on transformation geometry, spatial ability, levels of geometry understanding and attitudes
towards mathematics]. Karaelmas Journal of Educational Sciences, 3(1), 36-48.
Yılmaz, T. (2012). Comparison of the effects of model - based and computer - based instruction on 9th grade students’ spatial abilities
and conceptual understanding of ionic lattice. Unpublished master’s thesis, Boğaziçi University, Istanbul, Turkey.
Yue, J. (2008). Spatial visualization by realistic 3D views. Engineering Design Graphics Journal, 72(1), 28-38.
Ali Ihsan Benzer PhD, Assistant Professor, Hatay Mustafa Kemal University,
Department of Management Information Systems, Hatay, Turkey.
E-mail: aibenzer@gmail.com
Bunyamin Yildiz PhD, Professor, Hatay Mustafa Kemal University, Department of
Mathematics, Hatay, Turkey.
E-mail: byildiz@mku.edu.tr
348
FACTORS INFLUENCING
THE FIRST YEAR STUDENTS’
ADAPTATION TO NATURAL ISSN 1648-3898 /Print/
SCIENCE STUDIES IN HIGHER ISSN 2538-7138 /Online/
EDUCATION
Abstract. The mass-approach in higher

Rita Birzina, education promotes the broadening of
Dagnija Cedere, students’ number and diversity therefore
Liva Petersone the question about the young people’s
readiness to adapt to studies becomes more
and more topical. Students who enrol in the
natural science faculties of the University
of Latvia (UL) are motivated and have
Introduction high assessment in biology and chemistry
as school subjects. However, the studies
The study of student retention is one of the most widely studied areas
already during the 1st term prove that stu-
in higher education. Already some 40 years ago, student attrition was typi-
dents encounter problems to get adjusted
cally viewed through the lenses of psychology as the reflection of individual
attributes, skills, and motivation (Tinto, 2006). During the last 20 years, there to the study process. In order to find out
has been significant interest in factors leading to student drop out from the the key factors that determine the adapta-
first-year higher education studies (TEQSA, 2017). tion of the first-year students to studies,
The transition from secondary school to university is a challenging an e-questionnaire was developed and 79
phase in a student’s life (Sotardi, & Brogt, 2016, p. 3). These changes usually students from the Faculty of Chemistry and
are connected with the study environment, the fulfilment of academic tasks Faculty of Biology were surveyed. Institu-
and how students are able to achieve the formulated learning outcomes. The tional and personal factors that are focused
expectations of first-year students are becoming more widely researched and on the application of transferable skills
understood. Yet there are fewer studies about how well they feel prepared are mainly those that affect the students’
to the transition from secondary school to the tertiary learning environment
adaptation in the first term of studies. At
(McPhail, 2015) and what is their first-year study experience of improving
the same time such issues of the academic
academic success and engagement (Kift, Nelson, & Clarke, 2010).
factor as students’ prior knowledge in biol-
By its nature “adaptation is the process and outcome whereby thinking
and feeling persons as individuals or in groups use conscious awareness and ogy, chemistry, physics and mathematics
choice to create human and environmental integration” (Vera, 2014). Accord- and their skill to learn independently are
ing to the experience of Malaysian researchers (Abdullah, Elias, Mahyuddin, intrinsic for students’ successful studies.
& Uli, 2009) the first six weeks are a critical period of adapting to the new Key words: adaptation to studies, first
environment and the solution of conflicts and problems. year natural science students, institutional,
personal and academic factors, transfer-
Problem of Research able skills.
Approximately one third of students in higher education in the OECD

countries do not complete their studies (OECD, 2016). There is a lack of
Rita Birzina, Dagnija Cedere,
systematic knowledge, data and indicators on study success in Europe: only Liva Petersone
12 out of 35 European countries regularly report a national indicator of University of Latvia, Latvia
completion (Vossensteyn et al., 2015), and graduation rates still vary widely
among European countries (from 10 %–60 %). Low completion rates point to
349
FACTORS INFLUENCING THE FIRST YEAR STUDENTS’ ADAPTATION TO NATURAL SCIENCE
STUDIES IN HIGHER EDUCATION
(P. 349-361) ISSN 2538–7138 /Online/
problems with adaptation to bachelor-level studies, and many students interrupted their studies in the first year.
Students’ drop-out in many EU countries has attracted the attention of many researchers, policy makers and edu-
cators. Although the situation in different countries is different and the reasons for students’ early drop-out differ
from student to student, there are still observable common trends including learning difficulties, socio-economic
problems and the lack of motivation or insufficient support (European Comission/EACEA/ Eurydice, 2015).
According to the data of internal statistics of the University of Latvia (LUIS, 2017) – in 2017, the Faculty of
Biology enrolled 70 students, after the 2nd term there were 53; the drop-out was 30%. Sixty students started their
studies in the Faculty of Chemistry, after the 2nd term there remained 46 students; the drop-out – 25%.
The regular study performed by the Study Department of the UL shows that on the whole students are satis-
fied with the study environment; however, the students of natural sciences faculties in comparison with students
of other faculties consider problematic the high difficulty level of studies, the skills to learn independently, the lack
of prior knowledge and the inability to plan their time for studies.
A topical issue is how students actually adapt to university studies. Is this the level of students’ prior knowledge
or their ability to adapt to university learning, or teaching methods used by the university lecturer? Or are poor
study habits, time management, and study skills among the obstacles that stand in the way of learning for many
students (Tennen & Hagar, 2011)? Sometimes, graduates “discover that they do not have the skills and competences”
for studying (European Commission/EACEA/Eurydice, 2015, p. 3). There is no better way to find out than to ask the
students directly if something has helped them to learn and to succeed or has held them back and discouraged
them from learning (Bain, 2004).
Research Focus
In order to understand the main reasons that hinder successful studies and lead to great students’ drop-out in
the first year, it is important to clarify the key factors that influence students to leave the studies. However, there is
no complete agreement amongst researchers as to which factors are the most significant, and it is very difficult to
create an absolute hierarchy of impact amongst them (Quinn, 2013). She considers that these could be the above
mentioned six key (socio-cultural, structural, policy, institutional, personal and learning) factors leading students to
drop-out. At the same time, it is not clear whether there is one key factor, or it is the combination of these factors
that leads to the drop-out and there is no one essential reason.
According to ideas of several authors (Cherif, Movahedzadeh, Adams, & Dunning, 2013; Sotardi & Brogt, 2016;
Sotardi & Friesen, 2017; TEQSA, 2017) the body of many factors define the student’s readiness to studies. For instance,
Sevinc and Gizir (2014) and McCulloch (2014) regard the academic, social, personal-emotional, and institutional
adjustment of first-year university students as essential factors that affect negatively the studies. Cherif et al. (2013,
p. 7), in their turn, consider that student failure has multiple contributing factors. They group student failure into
seven main categories (motivation, study habits and related issues, instruction and related issues, student’s academic
preparedness, external factors, student’s attitudes, and other relevancy issues) and fifteen sub-categories. Also,
Krause and Armitage (2014, p. 4) admit that “student engagement, belonging, retention and success as integrated
parts of a complex structure rather than a linear process” play an important role. They identify such categories as
academic challenge, active learning, student/staff interactions, enriching educational experiences, supportive
learning environment, work integrated learning. Aljohani (2016) singles out such categories of influencing fac-
tors as family background, student-related factors, social factors, economic factors, students’ goals, institutional
experience and institutional factors. Students’ age, whether students were local or international, time since high
school and the learning strategy adopted (Bone & Reid, 2013) are mentioned as other factors. Trujillo and Tanner
(2014) maintain that students’ prior knowledge and sense of belonging are important. They find interesting the
opinion that “science identity is hypothesized to be key in student persistence and retention in the sciences” (p. 12).
Learning during the study process to a greater or lesser degree is connected with cognitive stress (Romanenko &
Nikitina, 2016); this then could also be one of the factors affecting the first-year students. The ideas of the above-
mentioned researches (Figure 1) were used as the theoretical background for researching the factors influencing
exploring studies of the first-year students while focusing on exploring institutional, personal and academic factors.
350
ISSN 1648–3898 /Print/ FACTORS INFLUENCING THE FIRST YEAR STUDENTS’ ADAPTATION TO NATURAL SCIENCE
ISSN 2538–7138 /Online/ (P. 349-361)
Figure 1. Description of institutional, personal and academic factors influencing the studies.
The research has been performed in the frame of two natural sciences study programs of one university that
in principle corresponds to the fact that “the majority of the reported research on the first-year experience is based
on single institutions studies, often with small samples of students, not uncommonly from a single programme of
study” (Harvey, Drew, & Smith, 2006, p. I) and “science has been a popular area for evaluations of first-year knowl-
edge” (ibid., p. 28 ). The aim of the present study is to explore how the first-year students are able to adapt to the
university process. The following research question was put forward: which are the main reasons that hinder the
first-year students’ adaptation to studies?
General Background
General education in Latvia in total lasts 12 years consisting of compulsory 9-years basic education and
3-years secondary education. General secondary education programmes are implemented in secondary schools,
gymnasiums, evening (shift) and extramural secondary schools (General Education). The research was conducted
at the University of Latvia, faculties of Biology and Chemistry in 2016/2017 study year after finishing the first study
semester. The research design: a case study. The choice of the case study was defined by the assumption that the
current study could give an insight into situations that describe the particular target group. This is one of the ways
how to study a concrete problem in order to reach a common understanding of the situation (Kumar, 2014). To
specify the views of the first-year students about their adaptation to the university studies, the researchers used
open-ended questions for supplementing the students’ views obtained by closed questions, which gave an op-
portunity to gain insight into the predetermined set of opinions and for the respondents to give replies in their
own words (Williams, 2007).
Sample
Eighty percent of the total number (N = 99) of the first-year students in the faculties of Biology and Chemistry
participated in the survey. The sample of the research consisted of 79 first year students, of whom there were 56
female (71 %) and 23 male (29 %) students studying at the Faculty of Biology (N = 38; 48%) and the Faculty of Chem-
istry (N = 41; 52%) of the University of Latvia. Respondents have received their previous education in secondary
351
(P. 349-361) ISSN 2538–7138 /Online/
schools in different regions of Latvia: 35 (44%) of them have graduated from a gymnasium, 40 (51%) – secondary
school, 4 (5%) – another educational institution. The students’ average age is 20.2 years (SD = 2.5).
In order to find out the main reasons that affect the studies of the first-year students, they were surveyed at
the end of the first term. Students’ participation in the survey was voluntary and the respondents’ anonymity was
guaranteed. The online survey with close-ended 5-point Likert scale and open-ended questions was used as the
research instrument on the QuestionPro platform (https://www.questionpro.com/). The developed questionnaire
consisted of two parts: general and conceptual. The whole structure of the questionnaire is presented in Figure 2.
Figure 2. The structure of the developed questionnaire.
The general part of the questionnaire comprised open-ended questions that included questions about students’
previous learning experience at school evaluating positive and negative impacts on studying the chosen subject.
The conceptual part of the questionnaire consisted of three sections (Figure 2; Appendix 1). Conceptual part
focused on finding out the student’s opinion about studies at the university assessing the factors that influence
studies on the 5-point Likert scale (1 – strongly disagree, 5 – strongly agree). Items include issues aimed for assess-
ing institutional, personal and academic factors. Institutional factor is related to such issues as an organisation of
students’ study activities, communication and provision of study resources. Personal factor characterizes issues
featured by the student as a personality, including his personal qualities as learning motivation, learning skills and
attitudes to study activities, and contact with other course-mates, friends and family. Academic factor is charac-
terized by issues that are related to the student’s learning activities. The choice of the issues of these factors was
based on the outcomes of the previous studies (Birzina, & Cedere, 2017).
To determine the mutual influence of the issues characterizing the mentioned factors on the student’s adapta-
tion, the students were classified into three groups (A, B, and C) according to their performance in the first study
year. Group A (“high performing” students) consisted of students with assessment 8–10 in 10-point scale. Group B
(‘medium performing” students) – 6–7 and Group C (“low performing” students) – 4–5. The breakdown in groups
was based on students’ scores ranging from 4–10. When reaching a score (4–satisfactory and 5–almost satisfactory),
the student demonstrates his competence at a lower level. In the assessment (6–almost good and 7–good) the
student reaches the average (medium) level and only (8–very good, 9–excellent, 10–with distinction) has reached
the highest level. This ranking system is in the entire university (Grading system).
Subsequently, a comparison was made between these groups in search of common and different issues.
Data Analysis
The descriptive statistics was used in order to analyse the mean of respondents’ answer (1 ≤ M ≤ 5), standard
deviation, mode and median. Cohen’s d was calculated for estimating the effect size for the difference between
two means. Comparison of groups with different sample size (Cohen’s d, Hedges’ g) was used. As the sample size
is small, non parametric test was performed, and Spearman’s correlation coefficient was used in order to iden-
tify the strength between variables and relationships among the institutional, personal and academic factors.
Cronbach alfa reliability test which estimates the internal consistency of items in the survey, the relevancy of the
consistency among items was found out. The Mann-Whitney test was used to compare two independent groups
352
ISSN 2538–7138 /Online/ (P. 349-361)
and the Kruskal-Wallis Test to compare three independent groups. The obtained qualitative data were processed
with the SPSS program.
Qualitative data (Survey General Part) were used to supplement those obtained in the quantitative research,
focusing on the unique features of the personality. By means of the qualitative analysis problems of personal,
academic and institutional character were clarified.
Data about students’ average academic assessment in studies in the first term and at the end of the first year
were obtained from the UL authorised internal LUIS electronic study system (LUIS, 2017). Confidentiality of the
obtained information was guaranteed.
Research Results
The university entrance score of these students is very high within the range of 8–10 in 10-point system)
which indicates a high level of prior knowledge (Table 1). Using LUIS (2017) data were obtained about students’
total mean assessment, the mark in biology/chemistry in the secondary school certificate and assessment in the
first term. As it is seen in Table 1, there are differences between students’ marks in biology and chemistry as school
subjects and in University. Students predominantly enter the university with high academic achievements – the
average mark in the secondary school certificate is almost 8 points and in the particular school subjects it is even
higher – above 8 points. Cohen’s d < .2 show that there is no statistically significant difference in average marks
between Biology and Chemistry students. It means that in subsequent measurements students of both faculties
can be seen as one group.
Table 1. Students’ performance before studies and during studies in university.
Item Study program N M SD Cohen’s d
Biology 38 7.83 0.88

Average mark in the certificate 0.05
Chemistry 41 7.88 1.07
School
Biology 38 8.41 1.52

Mark in the speciality subject 0.02
Chemistry 41 8.44 1.25
Biology 53 6.40 1.14

University
Average mark after finishing the first term

0.10
(LUIS data) Chemistry 46 6.56 1.90
Analysis of the Factors Influencing Student Adaptation
High mean values (mainly M = 3–4.45), median (from Mdn = 4–5) and mode (predominantly 5) dominated
in the students’ answers on the 5-point Likert scale (Appendix 1). This proves that students on the whole assess
positively their adaptation to studies in the first year. However, a more detailed analysis of the data showed that
there were differences in the understanding of different issues. Identifying the internal consistency among the
items included in the survey, it was concluded that Cronbah alfa in all factors is in the limits of the mean reliability
(α = .67; .69; .81 – respectively Personal (PF), Institutional (IF) and Academic (AF) factors) which indicates that the
obtained data correspond to the consistency criteria and are applicable for describing these factors.
In order to analyse students’ answers in more detail, the mutual relations of the obtained quantitative data
were identified among the factors and within the boundaries of one factor defining the Spearman’s rank correlation
coefficient. Corresponding to the number of respondents the correlation between personal factor and institutional
for students was weak, r(78) = .24, p < .05. None of these factors correlate with the academic factor. It means that
during the first term students connect themselves more with settling in the social environment than the acquisition
of the organizational procedures. They assess their studies from the aspect of personal well-feeling participating
in the study process and think much less about the academic aspect of studies.
353
(P. 349-361) ISSN 2538–7138 /Online/
Figure 3. Characteristic features influencing students’ adaptation to study environment.
Performing a more detailed exploration of the relations of different issues, the most characteristic features
of each factor were studied among which there is a medium correlation (Figure 3; Appendix 1, 2). The moderate
Spearman correlation coefficient value of .41–.43 confirm positive correlation among PF4 and IF9, IF10 and IF11.
This correlation coefficient is significant at the 0.01 level of significance. Students have no problems with the avail-
ability of resources in the university library (IF10, M = 3.71) and the possibilities to use the electronic system (IF11,
M = 3.95), although polarity was observed also in these answers (SD = 1.26–1.32, respectively).
One of the issues, which serves as evidence for student’s adaptation to the study environment (PF4) is the
ability to plan the time of studying (PF3, M = 3.25; SD = 1.36). An important role to the solution of the study-related
problems is given to the curator (IF9, M = 4.20; SD = 0.99). whom the student can meet any time needed to receive
more detailed information about uncertainties in the study process. It means, social skills are important for the
students, i.e., they are well adapted to the new environment because they could communicate freely with the
course curator. Work of curators (a selected person of the academic staff ) was one of student support measures
to promote adaptation of students in academic environment and study process. The curators assisted students in
grasping the opportunities for applying the skills and knowledge offered by study courses (Latvijas Universitātes
2017. gada pārskats, 2018).
There was a weak positive correlation between PF4 and PF3. It shows that they did not experience any
problems with planning the study time (PF3). That students’ experience in turn, has a weak correlation with his
personal learning abilities (PF2). Within the academic factor (AF) there is a relatively low correlation between the
student’s learning skills and the sufficiency of the amount of information to be acquired (AF2, M = 3.20; SD = 1.23)
and moderate correlation with independent learning ability (M = 3.08), which is simultaneously characterized by
SD value 1.43, which indicates the difference between these skills among the students.
Since the values of these issues are dominated by relatively high SD values, it is essential to find out how these
differences appear in the A, B and C groups related to the level of students’ academic preparedness.
354
ISSN 2538–7138 /Online/ (P. 349-361)
The Impact of Academic Performance on Student Adaptation
In order to clarify whether the adaptation to studies is related to students’ academic achievements at the
university, it was important to find out the proportion of students with high, medium and low assessment. Sixty-
five students responded to this question in the survey. The Kruskal-Wallis test (Table 2; Appendix 1) helped to
explore a statistically significant difference issues among A group, B group and C group students revealing the
most significant issues in personal and academic factors that affect student adaptation.
Table 2. Issues affecting the studies of A, B and C group students in Test Statisticsa.b.
Total (N=65) Group A (N=18) Group B (N=37) Group C (N=10)

Issues Chi-Square df p
M SD M SD M SD M SD
PF6 4.26 1.15 4.50 0.92 4.46 0.99 3.10 1.45 11.10 2 .004
PF7 4.12 1.22 4.22 1.00 4.38 1.06 3.00 1.56 8.73 2 .013
PF2 3.40 1.38 4.33 0.97 3.20 1.41 3.00 1.14 12.16 2 .002
PF3 3.25 1.40 3.83 1.20 3.11 1.39 2.70 1.57 5.05 2 .080
PF11 2.97 1.46 3.28 1.18 3.11 1.60 1.90 0.88 6.48 2 .039
AF12 3.65 1.32 4.22 1.17 3.49 1.39 3.20 1.03 6.03 2 .049
AF14 3.55 1.40 4.22 1.11 3.38 1.38 3.00 1.63 7.11 2 .029
AF15 3.55 1.42 4.39 0.98 3.27 1.47 3.10 1.37 10.13 2 .006
AF13 3.45 1.52 4.56 0.98 3.14 1.51 2.60 1.35 16.23 2 <.001
AF4 3.03 1.46 3.89 1.28 2.89 1.43 2.00 1.05 11.53 2 .003
a. Kruskal Wallis H Test
b. Grouping Variable: Group
Results of the independent sample indicated that there is a statistically significant difference between Group
A, Group B and Group C students. The obtained data (Table 2) allow concluding that the difficulties between the
“high performing”, “medium performing” and “low performing” students in the transition from learning at school
to studying at the university are connected with personal and academic factors. Besides, the specific weight of the
academic factor is the decisive. This is proved by the statistically significant difference in such issues: the skill to learn
independently (AF4), the prior knowledge in biology (AF12), chemistry (AF13), physics (AF14), and mathematics
(AF15). The mentioned issues of the academic factor show that there has been a consistent decline in the average
value among high, medium and low performing students.
Concerning the personal factor, as the main issues are mentioned learning skills (PF2); studies planning skills
(PF3); support from family (PF6) and friends (PF7), as well as stress (PF11). A consistent average decline is observed
in student responses for learning skills (PF2) and study planning skills (PF3).
Effect size was calculated to compare groups in pairs: A–B, A–C, B–C. For this purpose, the comparison of
groups with different sample size (Cohen’s d, Hedges’ g) was used. As shown in Table 3, there is a small difference
between learning skills (PF2) for groups B and C, but they are much larger for group A. Whereas support for friends
(PF7) and family (PF6) has received similar and stronger over A and B students Group C students.
Interestingly, there is finding on the stress level (PF11), which shows that “high performing” students are more
concerned about their studies as a “low performing” students. It can be concluded that successful students take
more responsibility for the success of the study process.
Table 3. Effect sizes (d) among students’ groups A, B and C.
PF6 PF7 PF2 PF3 PF11 AF12 AF14 AF15 AF13 AF4
dA-B -0.04 0.15 -0.88 -0.54 -0.12 -0.55 -0.65 -0.84 -0.88 -0.72
dA-C -1.24 -1.00 -1.29 -0.88 -1.27 -0.91 -0.93 -1.14 -1.29 -1.57
dB-C -1.24 -1.17 -0.15 -0.29 -0.82 -0.22 -0.26 -0.12 -0.15 -0.65
355
(P. 349-361) ISSN 2538–7138 /Online/
The analysis of students’ answers to open-ended questions, too, (Table 4) show that there are differences
in students’ opinions how learning at school has influenced them for the study process at the University. Their
academic performance at the university to some extent is also related to their learning experience in secondary
school. Students assess their learning at school very differently starting from the acquisition of the knowledge in
the given subject to the skill of learning and master new information as well as plan their time and the importance
of personal qualities.
Table 4. Comparison of students’ learning experience acquired at school.
Positive ideas Negative ideas

Issue
(quotes from students’ answers) (quotes from students’ answers)
Prior knowledge Basic knowledge in almost all fields. maybe not so good Insufficient knowledge from the secondary school.
prior knowledge in molecular biology and microbiology.
Time management To plan one’s work and to analyse the outcomes. Out-of-university activities (sports, theatre) and new
friends from Riga. Chemistry “moved to the second
place”.
Amount of information I can systematize information, work with vast amount of The unusual amount of information and the skill to
information work with it made studying difficult.
Skills to learn independently I can work independently and continuously. Understand My short span of concentration and my attention can
systematics, learn how to understand biology better, to be easily distracted, I cannot “force” myself to learn.
make my own system of learning. To think logically and
to analyse.
Personal qualities My persistence and conscientiousness. The willingness My laziness and not so high motivation to try. The lack
to study and explore things. Self-invested work and of self-initiative.
determination.
Discussion
The given research emphasizes the interaction of the institutional, personal and academic factors in the
student’s adaptation to the University. The obtained results (Figure 3) showed that students in the first study year
paid greater attention to the personal and institutional factor than the academic. This means that they are more
oriented to the use of their “soft” study skills – a set of individual transferable skills that are typically considered
as not specifically related to academic discipline or area of knowledge but required for learning and assessment
at university and that can be used in a wide variety of situations (Cattanach & Watkins, 2017; UNESCO, 2013).
The student feels good if there is social communication with the study support staff. Also, the curator plays a
significant role in the establishment of a positive study environment (Figure 3). This is proved also by the ideas
expressed by Sotardi and Friesen (2017) that students’ perceptions of the learning environment as clear and
organised, interesting and personally relevant, and supportive of building a peer-based learning network were
linked to positive changes in their learning and study practices. Formal and informal interaction with the faculty
contributes to students’ university adjustment, therefore it is good if an effective advisory system and mentor-
ing programs for new students are developed at the university. This means that the student’s belonging to his/
her faculty is perceived as a prominent factor affecting students’ adjustment. If the student develops negative
relations with the academic staff and he/she is unable to form friendly relations with the course-mates then the
teaching quality suffers as a result of the social adjustment (Sevinc & Gizir, 2014).
The personal factor is the one that largely determines the formation of student’s feeling of comfort. The
majority of students are enrolled in the chosen study program with a high motivation to acquire this branch
of science (Birzina, & Cedere, 2017; Cēdere, Jurgena, & Targamadze, 2018) however, their personal qualities can
affect their studies. The research outcomes (Table 2, 3, 4) show that students often lack skills to plan their time
and they are unable to define their priorities in the course of the studies as well as laziness is one of the reasons
that hinders successful studying. Cherif and his co-authors (2013), too, as well as researches by Goldfinch and
Hughes (2007) confirm that the rate of academic failure among students affects the connection between poor
study habits: poor study skills, poor time management, and an inability to identify and establish priorities, as
well as their laziness. They prove that students with poorly developed skills in organizing self-directed learning
356
ISSN 2538–7138 /Online/ (P. 349-361)
have more chances to become dropouts. It means that the interaction of social and academic factors is especially
important because “poor first-year academic performance alone will not result in withdrawal” (Harvey, Drew, &
Smith, 2006, p. 16). The student’s well-feeling is also decisive in the continuation of the studies.
Student knowledge is a crucial, complex resource when regarding university-level assessment (Rohaeti, &
Prodjosantoso, 2018; Sotardi & Brogt, 2016, p. 7). The research findings show (Table 1; 2; 3) that academic fac-
tors are also those that influence significantly successful studies. Although students consider that their school
knowledge in the subjects of their specialization is high, there are still observable differences not only in the
assessment of the school and university performance (Table 1) but also between the Group A, Group B un Group
C students (Table 2; 3). Students of both faculties admit that there should be good prior knowledge not only
in the chosen science in the branch (biology or chemistry) but also in the field of other natural sciences – in
physics and mathematics. It means that insufficient level of prior knowledge could be one of the main and this
is proved by the contradictions in the answers given by “high performing” and “low performing” students to
the open-ended questions (Table 2; 3). Harvey, Drew, and Smith (2006) consider that student’s prior knowledge
could be one of the predictive factors, which determine the student’s successful studying. Also, other researchers
(McCulloch, 2014) maintain that students with higher prior knowledge are less subjected to the dropout risk.
Students responses show that there are differences between “high performing” “medium performing” and “low
performing” students, especially in the assessment of prior knowledge in chemistry and mathematics.
However, not always regardless the depth of knowledge in the subject, students are able to fulfil the re-
quirements defined at the University, they are unable to fulfil the expected requirements of the task and fail to
receive full marks. This means that “they consistently lack discipline-specific skills that are needed for achievement
on 100-level course tasks” (Sotardi & Brogt, 2016, p. 15). It is possible to explain it that although the majority
of students enrolling at the university expect that studies will differ from learning at school (Crisp et al., 2009),
they still are not fully ready to face the reality and that significantly influences their ability to function well in
the unfamiliar university environment (Bone & Reid, 2013).
Students have diverse educational experiences, life experiences and personal circumstances. This diversity
has significant implications for students’ expectations, motivations, and the academic and personal resources
that they bring to their university experience (James, Krause, & Jennings, 2010). This is one of the factors that has
to be taken into account and the present research also shows the differences between the “high performing”,
“medium performing” and “low performing” students in such aspects as their time planning, the learning skills
and prior knowledge (Table 2; 3). It is possible that these differences are mainly connected with the student’s
learning skill and his/her prior knowledge because “learning in a particular area of knowledge such as life sci-
ences can be viewed as a continuum from novice to expert status” (Wood, 2009, p. 5) and if there is no prior
knowledge then “new information unrelated to prior knowledge is difficult to learn and remember” (ibid. p. 4).
Thus, it is important to take into consideration “what students can do with knowledge, rather than what units
of knowledge they have” (Silva, 2009, p. 631). Most likely “low performing” students lack higher-order thinking
skills and they are more oriented to mechanic memorizing of unrelated facts and are unable to form meaningful
acquisition of mutually related information therefore the amount of information that needs to be memorized
seems too big. Marouchou (2012) considers that students with a higher level of thinking concentrate more on
meaningful understanding of the acquirable topics and are more responsible for their learning and personal
development. This means that if the students lack skills to learn independently, then the amount of information
to be acquired seems too large (Table 2). Thus, if the students are unable to comprehend to their mind the large
amount of information, then it is possible that they are more oriented to the surface learning – the memoriza-
tion and recalling of facts based on “cramming” not on the deep learning – the learning with understanding
and the formation of interconnections.
Implications for Future
As the present research is a small-scale study with the sample of 79 participants, then the generalization
of the obtained results is limited; however, as a case study research it marks the tendencies in students’ adapta-
tion to the natural sciences studies that should be taken into consideration to reduce the student dropout. The
authors’ previous studies (Birzina & Cedere, 2017), too, showed a similar situation with the students’ readiness
to become a successful student. The research continues with broadening the target audience and exploring the
opinion and experience of the academic staff, how to reduce the student dropout at the university. It would be
357
(P. 349-361) ISSN 2538–7138 /Online/
interesting to continue that research in an international environment, and to prepare an international compara-
tive study, clarify if the same factors appear to be important in cultural context of the other universities.
Conclusions
Students’ adaptation to studies in the first year is a complex process which is affected by institutional, per-
sonal and academic factors. Students’ subjective opinion testifies about good readiness for the studies; they do
not encounter problems that would hinder successful studies.
Initially, students associate the adaptation to studies in the first term more with the “soft” transferable skills:
personal well-feeling and integration in the university study environment. Thus, students consider the personal
and institutional factors the key factors that affect their studies in the first term.
The academic factor is not considered as essential although the research findings reveal that exactly this
factor is the one that shows most differences between the adaptation of “high performing”, “medium performing”
and “low performing” students to the studies. The main differences lie in such issues as their skill to learn inde-
pendently at the university and the prior knowledge acquired at secondary school. Thus, the above-mentioned
aspects are significant for the students’ successful studies. If the students cannot learn independently, then they
encounter problems to acquire the new information well enough and the amount of this information seems too
large because obviously it is more connected with the surface learning – the memorization and recalling of facts
not with the deep learning – the understanding of interconnections. Students’ prior knowledge also determines
learning. Although students are certain that their prior knowledge is sufficient for studies, the research data prove
that in order to study successfully, the student needs to have equally good prior knowledge not only in some (the
chosen) school subjects but in the field of all natural sciences: biology, chemistry, physics and mathematics.
The studies of “low performing” students at the university are affected also by such personal factors as the
skill to learn acquired at school and the skill to plan one’s time, taking the responsibility for one’s studies that are
insufficient for these students. The lack of these skills also promotes the student dropout.
In general, it is possible to regard the mutual connection of the above-mentioned issues as the reason of the
big dropout numbers of the first-year students.
References
Abdullah, M. C., Elias, H., Mahyuddin, R., & Uli, J. (2009). Adjustment amongst first year students in a Malaysian university. Euro-
pean Journal of social sciences, 8(3), 496-505.
Aljohani, O. (2016). A review of contemporary international literature on student retention in higher education. International
Journal of Education and Literary Studies, 4(1), 40-52.
Bain, K. (2004). What the best college teachers do. Cambridge Massachusetts: Harvard University.
Birzina, R., & Cedere, D. (2017). The first-year students’ perceptions of higher studies: A case of University of Latvia. In V. Dislere
(Ed.), Proceedings of the 10th International Scientific Conference “Rural Environment. Education. Personality. (REEP)”, (pp. 40-49).
Jelgava: Latvia University of Agriculture.
Bone, E., & Reid, R. J. (2013). First course at university: Assessing the impact of student age, nationality and learning style. The
International Journal of the First Year in Higher Education, 4(1), 95-107.
Cattanach, K., & Watkins, R. (2017). Academic Adaptation to first year study: Student perceptions and staff engagement. Retrieved
15/02/2018, from http://www.enhancementthemes.ac.uk/docs/paper/8-2-3-academic-adaptation-to-first-year-study-
student-perceptions-and-staff-engagement.pdf?sfvrsn=4
Cēdere, D., Jurgena, I., & Targamadze, V. (2018). Interest of Latvian and Lithuanian students in science and mathematics. Journal
of Baltic Science Education, 17(1), 31-42.
Cherif, A., Movahedzadeh, F., Adams, G., & Dunning, J. (2013). Why Do Students Fail? Higher Learning, 227, 228.
Crisp, G., Palmer, E., Turnbull, D., Nettelbeck, T., Ward, L., LeCouteur, A., Sarris, A., Strelan, P. & Schneider, L. (2009). First year student
expectations: Results from a university-wide student survey. Journal of University Teaching and Learning Practice, 6(1), 11-26.
European Commission/EACEA/Eurydice. (2015). The European Higher Education Area in 2015: Bologna Process Implementation
Report. Luxembourg: Publications Office of the European Union.
General education. Retrieved 15/08/2018, from http://izm.gov.lv/en/education/general-education
Goldfinch, J., & Hughes, M. (2007). Skills, learning styles and success of first-year undergraduates. Active Learning in Higher Educa-
tion, 8(3), 259-273.
Grading system. Education in Latvia. Retrieved 10/01/2019 from http://www.aic.lv/portal/en/izglitiba-latvija/izglitibas-sistema
Harvey, L., Drew, S., & Smith, M. (2006). The first-year experience: A review of literature for the Higher Education Academy. York: The
Higher Education Academy.
358
ISSN 2538–7138 /Online/ (P. 349-361)
James, R., Krause, K. L., & Jennings, C. (2010). The first-year experience in Australian universities. Canberra: Department of Educa-
tion, Employment and Workplace Relations (DEEWR).
Kift, S., Nelson, K., & Clarke, J. (2010). Transition Pedagogy: A third generation approach to FYE – A case study of policy and practice
for the higher education sector. The International Journal of the First Year in Higher Education, 1(1), 1-20.
Krause, K. L., & Armitage, L. (2014). Australian student engagement, belonging, retention and success: A synthesis of the literature.
U.K.: The Higher Education Academy. Retrieved 15/02/2018, from https://www.heacademy.ac.uk/node/8683
Kumar, R. (2014). Research methodology: A step-by-step-guide for beginners. London: SAGE Publications Ltd.
Latvijas Universitātes 2017. gada pārskats (2018). University of Latvia Annual Report 2017 (2018). Riga: UL Press.
LUIS (2017). Authorized electronic study system of the University of Latvia. Retrieved 8/08/2017, from www.luis.lv
LU Studiju departaments (2017). Pirmā studiju gada studējošo aptauja par pirmo studiju pieredzi. Autorizēta piekļuve [UL Depart-
ment of Studies (2017). First study year students’ survey on the first study experience. Authorized access]. (In Latvian).
Retrieved 2/03/2018, from https://www.lu.lv/portals/
Marouchou, D. V. (2012). Can students’ concept of learning influence their learning outcomes? Higher Learning Research Com-
munications, 2(2), 18.
McCulloch, A. (2014). Learning from future track: Dropout from higher education. Retrieved 15/02/2018, from http://dera.ioe.
ac.uk/19643/1/bis-14-641-learning-from-futuretrack-dropout-from-higher-education-bis-research-paper-168.pdf
McPhail, R. (2015). Pre-university prepared students: a programme for facilitating the transition from secondary to tertiary
education. Teaching in Higher Education, 20(6), 652-665.
OECD (2007). PISA 2006: Science Competencies for Tomorrow’s World: Volume 1: Analysis. Paris: OECD Publishing. Retrieved
5/06/2017, from http://www.oei.es/evaluacioneducativa/InformePISA2006-FINALingles.pdf
OECD (2016). Education at a Glance 2016: OECD Indicators. Paris: OECD Publishing.
OECD (2018). Education at a Glance 2018: OECD Indicators. Paris: OECD Publishing.
Quinn, J. (2013). Drop-out and completion in higher education in Europe among students from under-represented groups.
Network of Experts on Social Aspects of Education and Training. Brussels: European Union.
Researchers’ Report 2014 Country Profile: Latvia. Retrieved 8/08/2017, from http://ec.europa.eu/euraxess/
pdf/research_policies/country_files/Latvia_Country_Profile_RR2014_FINAL.pdf
Romanenko, V., & Nikitina, G. (2016). Cognitive stress in pedagogics: general analysis. In V. Dislere (Ed.), Proceedings of the 9th
international scientific conference “Rural Environment. Education. Personality. (REEP)”, (pp. 121-127). Jelgava: Latvia University
of Agriculture.
Sevinc, S., & Gizir, C. A. (2014). Factors Negatively Affecting University Adjustment from the Views of First-Year University Students:
The Case of Mersin University. Educational Sciences: Theory and Practice, 14(4), 1301-1308.
Silva, E. (2009). Measuring skills for 21st-century learning. Phi Delta Kappan, 90(9), 630-634.
Sotardi, V., & Brogt, E. (2016). Common student problems with assessment during the transition to university: Support staff perspec-
tives, insights, and recommendations. Ako Aotearoa National Centre for Tertiary Teaching Excellence.
Sotardi, V.A., & Friesen, M.D. (2017). Student reflections on learning and studying during the transition to university: Implications
for tertiary learning advisors. ATLAANZ Journal, 2(1) 19-35.
Tennen, G., & Hagar, G. K. (2007). Helping Students to Study Effectively. Retrieved 8/09/2017, from http://www.4faculty.org/
includes/digdeeper/helpstudents/studying.jsp
TEQSA - Tertiary Education Quality and Standards Agency. (2017). Characteristics of Australian higher education providers and
their relation to first-year student attrition. Retrieved 02/02/2018, from https://www.teqsa.gov.au/sites/g/files/net2046/f/
attrition-report-june-2017-19dec2017.pdf?v=1513650539
Tinto, V. (2006). Research and practice of student retention: What next? Journal of College Student Retention: Research, Theory &
Practice, 8(1), 1-19.
Trujillo, G., & Tanner, K. D. (2014). Considering the role of affect in learning: Monitoring students’ self-efficacy, sense of belonging,
and science identity. CBE-Life Sciences Education, 13(1), 6-15.
UNESCO International Bureau of Education (2013). IBE Glossar y of Curriculum Terminology. Retrieved
0 2 / 0 2 / 2 0 1 8 , f ro m h t t p : / / w w w. i b e. u n e s c o. o r g / f i l e a d m i n / u s e r _ u p l o a d / Pu b l i c a t i o n s / I B E _ G l o s s a r y
CurriculumTerminology2013_eng.pdf
Vera, M. (2014). Sister Callista Roy’s Adaptation Model of Nursing. Retrieved 12/09/2017, from https://nurseslabs.com/sister-
callista-roys-adaptation-model/#person
Vossensteyn, H., Kottmann, A., Jongbloed, B.., Kaiser, F. Cremonini, L. Stensaker, B. Hovdhaugen, E., & Wollscheid, S. (2015). Drop-
out and Completion in Higher Education in Europe: Main Report. Luxembourg: Publications Office of the European Union.
Retrieved 10/01/2019, from https://supporthere.org/sites/default/files/dropout-completion-he_en.pdf
Williams, C. (2007). Research Methods. Journal of Business & Economic Research, 5(3), 65-71.
Wood, W. B. (2009). Innovations in teaching undergraduate biology and why we need them. Annual Review of Cell and Develop-
mental, 25, 93-112.
359
(P. 349-361) ISSN 2538–7138 /Online/
Appendix 1
Descriptive Statistics (N = 79) of issues arranged in descending order by means in every factor (codes of the
analysed issues marked in bold)
Code Issues Mean Median Mode SD

IF7 Communication with the administrative staff and study secretaries. 4.47 5.00 5 .78
IF1 Timetable of lectures. 4.28 4.00 5 .85
IF9 Communication with the curator (tutor). 4.20 5.00 5 .99
IF8 Communication with the academic staff and laboratory assistants. 4.16 4.00 5 .85
IF11 UL electronic systems (e-studies and LUISS). 3.95 4.00 5 1.26
IF12 Internet access. 3.91 4.00 5 1.30
IF6 Information about the requirements. 3.73 4.00 5 1.13
IF10 Provisions of the library resources. 3.71 4.00 5 1.32
IF2 Proportion of lectures to the practical classes. 3.68 4.00 5 1.15
IF3 Too long laboratory tasks. 3.37 4.00 5 1.40
IF4 Too few seminars. 3.10 3.00 3 1.38
PF9 Participation in seminars. 4.43 5.00 5 1.08
PF6 Family support. 4.32 5.00 5 1.09
PF12 Painful stress situation 4.25 5.00 5 1.28
PF7 Friends’ support. 4.13 5.00 5 1.21
PF8 Attendance of lectures. 4.03 5.00 5 1.35
PF4 Settling into the study environment. 3.95 4.00 4 1.06
PF13 Frequency of illness 3.81 5.00 5 1.57
PF1 Learning motivation. 3.47 4.00 5 1.38
PF2 Learning skill. 3.47 4.00 5 1.34
PF3 Planning the time for learning. 3.25 4.00 4 1.36
PF11 Stress about the organization of the studies. 2.94 3.00 1 1.45
PF10 Stress about the organization of the everyday life. 2.75 2.00 2 1.38
PF5 Work along the studies. 2.27 1.00 1 1.68
AF10 The use of free access e-materials. 4.32 5.00 5 1.02
AF11 Cooperation with course-mates. 4.18 5.00 5 1.15
AF1 The choice of the study program. 4.15 5.00 5 1.15
AF3 Support from the academic staff. 3.82 4.00 5 1.16
AF9 Useful tutorials. 3.81 4.00 5 .92
AF12 Prior knowledge in biology. 3.67 4.00 5 1.27
AF15 Prior knowledge in mathematics. 3.56 4.00 5 1.40
AF8 Time used preparing for the laboratory tasks. 3.53 4.00 5 1.26
AF14 Prior knowledge in physics. 3.52 4.00 4 1.39
AF13 Prior knowledge in chemistry. 3.46 4.00 5 1.48
AF7 Time used preparing for the seminars. 3.39 3.00 3 1.23
AF6 Boring and hard to understand lectures. 3.22 3.00 3 1.15
AF2 The amount of the acquirable information. 3.20 3.00 2 1.23
AF4 Skills to learn independently. 3.08 3.00 2 1.43
AF5 Adaptation to the lecturing style of the university teacher. 3.05 3.00 2 1.34
Note. IF- Institutional Factor. PF – Personal Factor. AF – Academic Factor
360
ISSN 2538–7138 /Online/ (P. 349-361)
Appendix 2
Spearman’s rho Correlation matrixs of issues influencing students’ adaptation to study environment (Figure 3).
Matrix 1
Code IF7 IF9 IF10 IF11 IF12 IF15
IF7 1
IF9 .43**
IF10 .30** .15
IF11 .21 .25* .51**
IF12 .31** .29* .40** .46**
IF15 .35 **
.15 .21 .23* .37**
PF4 .27* .41** .43* .41** .27* .33**
Matrix 2
Code PF2 PF3 PF4 AF1 AF2 AF3
PF2 1
PF3 .35**
PF4 .14 .33**
AF1 -.02 .16 .21
AF2 .35 **
.32 **
.19 .29**
AF3 .09 .30 **
.40 **
.23* .42**
AF4 .51** .45** .14 .13 .38** .38**
**. Correlation is significant at the .01 level (2-tailed).
*. Correlation is significant at the .05 level (2-tailed).
Received: April 10, 2018 Accepted: May 18, 2019
Rita Birzina Dr. paed., MSc. Biol., Leading Researcher, Faculty of Biology, University
(Corresponding author) of Latvia, Jelgavas street 1, Riga, LV-1004, Latvia.
E-mail: rita.birzina@lu.lv
Website: http://www.lu.lv
Dagnija Cedere Dr. chem, Associate Professor, Faculty of Chemistry, University of
Latvia, Jelgavas street 1, Riga, LV-1004, Latvia.
E-mail: dagnija.cedere@lu.lv
Liva Petersone MSc. Student of Biology Master Program, Faculty of Biology, University
of Latvia, Jelgavas street 1, Riga, LV-1004, Latvia.
E-mail: livapetersone2@inbox.lv
361
THE EFFECT OF THE USE
OF AUGMENTED REALITY
APPLICATIONS ON THE
ISSN 2538-7138 /Online/
ACADEMIC SUCCESS AND
MOTIVATION OF 7TH GRADE
STUDENTS
Abstract. The aim of this research was
to examine the effects of the Augmented
Reality (AR)-supported teaching activities
on the academic success, motivation of
students towards science learning and at- Esma Buluş Kirikkaya,
titudes of students towards AR application, Melek Şentürk Başgül
in the scope of the seventh grade science
course, “Solar System and Beyond” unit. For
this purpose, “Solomon Four Group Model”,
which controls the internal and external
validity, was used. Research was conducted Introduction
with 120 students in two different schools
and involved two experimental and two The rapid development of technology enables digital tools and applica-
control groups established by random tions to become an ordinary part of our social as well as learning experiences
method. Experimental (1) and Control (1) by allowing them to integrate more and more into daily lives. Augmented
groups received data collection tools as Reality (AR), which arises from these technological developments, is a technol-
pre-test and post-test, while Experimental ogy in which real and virtual objects are allowed to interact simultaneously
(2) and Control (2) groups received only and virtual-world objects are superimposed on real images (Azuma, 1997).
post-test. During the six-week research, the In other words, through these applications, an object that is not actually
students were taught in the experimental there is perceived as being there through the display of mobile devices. With
groups using the ‘Star-Tracker, Spacecraft, these developments in mobile devices, the AR technology is used in many
iSolarsystem, Space-4D’ mobile AR ap- sectors such as health, marketing, education, cinema, advertising and design.
plications, while the control groups were In engineering and architecture education, it is very important to use AR
taught with the activities envisaged in the applications since students need to animate objects in their minds in three
curriculum. “Solar System and Beyond Suc- dimensions. Quintero, Salinas, González-Mendívil and Ramírez (2015) de-
cess Test” and the “Students’ Motivation to signed an innovative and educational AR application aimed at strengthening
Science Learning” scale, were used as data mathematical cognitive skills for engineering students. With this application,
collection tools. Findings from the research it was determined that mathematical objects were learned more effectively
showed that teaching with AR applications than teacher drawings or verbal descriptions and students did not have dif-
significantly alters the success of students ficulties to imagine in three dimensions.
and motivation towards science learning. In the teaching of mathematics, the usage of AR in the classroom is
This showed that teaching with AR applica- quite important since teaching materials such as papers, pencils, ruler are not
tions is effective. sufficient to visualize geometric shapes three dimensionally in the minds of
Keywords: augmented reality, academic students. In a master thesis study, Gün (2014) carried out a semi-experimental
success, solar system, Solomon four-group study with 188 sixth grade students with the aim of investigating the effect
model, 7th grade students. of supplementing the mathematics course with AR applications named
BuildAR on the students’ spatial visualization abilities and academic success.
Three-dimensional shapes of rigid bodies were created with BuildAR during
Esma Buluş Kirikkaya application. As a result of the study, the students found the AR applications
University of Kocaeli, Turkey
Melek Şentürk Başgül fun and remarkable.
Science Teacher, Ministry of Education, Technology enhanced teaching equipment aids students in visualizing
Turkey physical concepts and understanding the real world situations and problems.
Özarslan (2013) conducted a semi-experimental study with 63 third-grade
362
ISSN 1648–3898 /Print/ THE EFFECT OF THE USE OF AUGMENTED REALITY APPLICATIONS ON THE ACADEMIC
SUCCESS AND MOTIVATION OF 7TH GRADE STUDENTS
ISSN 2538–7138 /Online/ (P. 362-378)
computer science students in order to determine the effect of using AR-enriched learning material on the success
and satisfaction levels of students. A functionally enriched application called OptikAR was developed. Eventually,
it was determined that the application positively affected student success and satisfaction levels.
In biology education, the usage of interactive AR applications in the teaching of many topics such as the di-
gestive system, circulatory system, plant diversity, insect diversity and classification, ecosystem, protein synthesis
resulted in more permanent and effective learning experiences. Yusoff, Zaman and Ahmad (2011) have worked
with 63 undergraduate students to determine their perception and acceptance of the AR technology. In the pro-
cess, the structure of the skin has been observed in three dimensions. As a result of the study, they concluded that
students would like to use this technology in the future. According to the result of another research, the students
have agreed that the AR platform was a new and useful tool that could be successfully applied for learning purposes
in the educational process (Lamanauskas & Bilbokaitė, 2009).
It is also possible to observe the atoms and molecules in three dimensions via AR technology. Nunez, Quiros,
Nunez, Carda and Camahort (2008) has carried out a study with 15 college students on inorganic chemistry using
AR applications. As a result of the study, it was seen that the three-dimensional crystal structures developed the
students’ spatial abilities and became an effective support tool that increased their success.
Significance and Aim of Research
It has become easier to bring AR applications into the class and laboratory as teaching tools using mobile
devices. In this way, mobile AR applications can be used as an educational tool for teaching of space and universe
learning as well as all other science fields. Observing the solar system and the planets clearly cannot be possible
in traditional classroom environment. For this reason, AR applications are seen as effective tools in providing ac-
cess to and visualization of the concepts in astronomy which are difficult to learn. Thus, the unknown features and
inaccessible locations of the universe are discovered by the students. They are provided with an opportunity to
experience and learn the universe in a realistic and better way. Şahin (2017) has designed supporting material and
developed AR activities for the “Solar System and Beyond” unit by consulting with educational design experts with
the aim of investigating the effects of supplementing science instruction with AR-enhanced teaching materials on
the success levels and attitudes of secondary school students. As a result of this semi-experimental study, it was seen
that there was a significant difference in the success levels and attitudes towards the course between the students
who took the course in a learning environment supported by AR technology and the success levels of students
who were taught in the traditional way. It was also determined that the AR application had a positive influence on
the attitudes of the students. In order to investigate the effect of the use of AR applications in the Solar System and
Beyond unit on the academic success of students, Kırıkkaya and Şentürk (2018) conducted a quasi-experimental
and pre-test & post-test controlled group study with 45 seventh grade students. As a result of the study, it was
determined that the AR applications applied in the experimental group increased the academic success.
Since the idea that AR can be used in educational environments has recently become widespread, many
applications have been designed to be used in science education. These include Spacecraft-3D, Star Tracker, Star
Chart, and Spacecraft -4D, Elements 4D, Dinosaur, Quiver-3D, Octaland-3D, Animal-4D, Augmenter, ARMolVis, Rapp
Chemistry, Animal Cell, Space Adventure, iSolarsystem and Solarsystem-AR applications.
It is known that the students of primary school age learn more easily about the objects they are able to see
and touch. It is especially important to be able to visualize objects such as the solar system, planets, telescopes
and spacecraft in order to comprehend the concepts in “Space and Universe” learning domain. AR technology is
a good alternative in terms of three-dimensional visualization of these concepts. The fact that the AR technology
provides interactive and reality-enhancing environments suggests that such practices can be used in teaching
subjects related to the world and the universe. Considering the fact that the usage of this technology is yet in
early stages and not yet widely used in classroom activities and the limited number of reports assessing its use in
science education in our country, it has been considered as worthwhile research topic. Given the limited number
of studies in which the Solomon four-group model was used, this research also filled a gap in the subject and the
AR technology as well as the method employed. In addition, while the researches for AR applications have only
just begun, the realization of a fully experimental work based on the Solomon four-group model which protects
both internal and external validity reveals the originality of the research. Through this perspective, it is aimed to
research the effect of teaching with AR-enhanced on the academic success and motivation of students in the context
of “Solar System and Beyond” unit in the Science course. In this respect, the research questions were as follows:
363
THE EFFECT OF THE USE OF AUGMENTED REALITY APPLICATIONS ON THE ACADEMIC
(P. 362-378) ISSN 2538–7138 /Online/
1. Is there a significant difference between students in the experimental and control groups in terms
of academic success and motivation upon application of the experimental procedure and pre-test
stimulation?
2. Does the experimental procedure lead to a significant difference between academic success and mo-
tivation pre-test and post-test scores of the groups stimulated by the pre-test in the science course?
3. When pre-test is controlled, does the experimental procedure make a significant difference between
the academic success and motivation post-test scores of the students in the pre-test groups?
Research Design
The use of experimental designs is recommended for strengthening scientific validity in educational research.
Experimental design is a research area where data are generated to explore the cause-and-effect relations between
variables under the control of researcher (Karasar, 2006). It is very important to ensure internal and external valid-
ity in this research area. By ensuring internal validity, it is determined whether the independent variable causes
the change in the dependent variable, whereas ensuring external validity allows the researcher to generalize the
results of the study to individuals in the population. This research conducted by using “Solomon Four-Groups
Design” which is one of the most powerful models that ensures internal and external validity at the same time
(Fraenkel & Wallen, 2009). This way, this research enabled the checking of the presence of pre-test sensitivity and
thereby added a higher external validity in addition to internal validity. In this context, a symbolic representation
of the Solomon four-group model was shown in Table 1.
As shown in Table 1, research groups included two experimental and two control groups. Pre-test scales were
applied to one of the groups whereas no pre-test was applied to the other experimental and control groups. In
other words, in addition to the effect of the independent variable on the result through the pre-test and post-test
applied to the first experimental and control groups, it was determined whether the pre-test has any effect on the
results by not applying the pre-test to the second experimental and control group (Rosnow & Rosenthal, 2002).
Table 1. A symbolic representation of the Solomon four-group model.
Groups Impartiality Pre-Test Treatment Post-Test
Experimental (1) R O1 X O2
Control (1) R O3 O4
Experimental (2) R X O5
Control (2) R O6
O1:Experimental (1) pre-test; O2:Experimental (1) post-test; O3:Control (1) pre-test; O4:Control (1) post-test; O5:Experimental (2) post-
test; O6:Control (2) post-test; X:Experimental intervention, R:Neutrality in the formation of groups
In experimental designs, multiple groups were used; usually experimental and control groups, and these
groups should be generated via unbiased assignment (Creswell, 2003). For this reason, research groups were
formed through random assignment in this research, so that any changes that could be observed in terms of
results between the experimental and control groups are unbiased. These steps were performed according to the
Solomon four-group model given in Figure 1.
As shown in Figure 1, experimental groups were educated with the help of AR applications appropriate for the
daily plans prepared by researchers, whereas control groups were instructed simply according to the daily plans.
“Solar System and Beyond Success Test” and “ Students’ Motivation to Science Learning “ scales were applied as
pre-test to the Experimental (1) and Control (1) groups prior to start of the study and as post-test to all groups at
the end of the research.
364
ISSN 2538–7138 /Online/ (P. 362-378)
Figure 1. The experimental design.
Research Groups
Before the experimental process, the groups to be included in the research were selected in two state schools
in the district of Dilovası, Kocaeli province in Turkey in the second semester of 2016-2017 education years. In the
selection attending groups, one-way analysis of variance (ANOVA) was done to determine whether there was a
significant difference between the year-end averages of science lecture of all seventh grades in these two schools
(Table 2).
Table 2. Results of One-way ANOVA for the science lecture report card averages of the seventh- grade students
in school A and B during the academic year of 2016.
Source of Variance Sum of Square df Mean of Square F p
Between Groups 3237.27 8 404.66 1.53 .15*

Within Groups 72452.61 274 264.43
Total 79073.66 282
*p>.05
365
(P. 362-378) ISSN 2538–7138 /Online/
As seen in Table 2, there was no significant difference between the average grade of Science lectures in the
2016 academic year of all seventh grade groups in two schools (F(1, 282)=1.53; p>.05). Since there was no statistically
significant difference between the groups, four randomly selected groups were included in the research. Experi-
mental and Control groups were assigned unbiasedly from these four groups. As a result, research was carried out
with 120 students which were grouped as experimental and control groups (Table 3).
Table 3. Demographic characteristics of students in the research group.
N
School Groups
Young Woman Young Man Total
Control (1) 17 13 30
School A
Control (2) 15 15 30
Experimental (1) 19 11 30
School B
Experimental (2) 13 17 30
Due to the lengthy research period, the researcher’s own activities and the need for a technical infrastructure,
the schools involved in the research were determined by the easy-to-reach sampling method. The reason for choos-
ing two different schools in the research was to ensure that the control group is not affected by the experimental
procedures. In this context, it was made a point of choosing the experimental groups from the same school. In ad-
dition, in the classrooms except the groups participating in the research in the same school where the experimental
procedures were carried out, the lessons were also taught with the support from the AR applications in order to
avoid the feeling of exclusion and to benefit from the same activities. The researcher has done the activities in all
classes in the school where the research was conducted.
Instrument and Experimental Process
Solar system and beyond success test (SSBST): The success test, which is formed of 20 questions, was developed
by Kırıkkaya and Şentürk (2018) in order to measure the success of students for the subjects of “Solar System and
Beyond” unit in the seventh grade. The test, with a confidence of .83, was evaluated over 100 points and details
are as follows: the total number of correct questions was multiplied by five, where one point was given for each
correct question and zero points for each wrong question. In this study, pre-test reliability coefficient for “Solar
System and Beyond Success Test” was .70; and the post-test reliability coefficient was .77.
Students’ motivation to science learning scale: In order to determine the motivation of primary school students
for science learning, “Students’ Motivation to Science Learning (SMSL)” scale, which was used in the study, was
developed by Tuan and colleagues in 2005. The Turkish adaptation study was conducted by Yilmaz and Huyuguzel
Çavaş (2007), and the reliability of the 5-point Likert type SMSL scale was found to be .87, and it was determined
that the reliability level for the measurement tools that can be used in the researches was provided. In this research,
reliability coefficient of pre-test was calculated as .82 for the SLMS scale and the final test was calculated as .89.
The experimental process was structured by the researchers with preparing lesson plans and activities. These
lesson plans and activities were organized in accordance with the AR applications and in line with the achievements
of the Science Education Program. The lesson plans prepared for the control groups were organized according to
the activities envisaged in the science curriculum. In addition, in order to provide the balance between the research
groups, it was found appropriate for the researcher to lecture the courses in both experimental and control groups.
The basic steps followed in the experimental process are shown in Figure 2.
According to Figure 2, SSBST and SMSL Scale was applied as pre-test before the experimental process and AR
technology was introduced to the experimental groups. It was necessary to inform students beforehand since they
had not already met with AR applications and that these applications required technological skills. The research
was conducted for a total of six weeks including four weeks experimental application, one week pre-test and one
week post-test applications. In this process, activities including AR applications were carried out by researchers in
the experimental groups. In the control groups, activities included in the curriculum were carried out (Figure 2).
366
ISSN 2538–7138 /Online/ (P. 362-378)
Time Realised Interventions. Groups

Determination of research groups.
Solar System and Beyond Success Test. (SSBST) Experimental (1)
Students' Motivation Scale for Science Teaching Control (1)
(SMSL)Pre-test application
1st. week
Introducing experimental process process (AG Experimental (1)
applications) to students. Experimental (2)
Performing activities supported by AG
applications in the Course of Science (performing
the experimental process).
Celestial bodies Space Adventure, Star Chart, Experimental (1)
Star Tracker, Nebulae, Experimental (2)
Solar System iiSolarsystem, Space 4D+, Solar
System 3D
Space Researches Spacecraft 4D, iSolarsystem,
2nd-5th Week Astronaut Control (1)
Processing of the course by the methods and Control (2)
techniques foreseen in the Science Education
Program.
Solar System and Beyond Success Test. (SSBST)

6th Week Students' Motivation Scale for Science Teaching All Groups
(SMSL)
Figure 2. Procedures followed during experimental process.
“Sky Bodies” subject was taught with Space Adventure and Star Tracker AR applications in experimental groups.
Students have been able to observe meteorite, meteor, meteor crater, star clusters (Ursa Minor, Ursa Major, and
dragon), Orion nebula, Andromeda galaxy, Milky Way galaxy in 3D observation. Moreover, owing to this applica-
tion, students had an opportunity to observe that the sun was a star and turned around its own axis and the stars
were not as the one on the Turkish flag, but were spherical in shape.
The “Solar System” subject was taught with iSolarsystem and Space 4D AR applications in the experimental
groups. One of the best achievements of these AR applications was that they showed very well that the planets
are turning around the sun in a certain orbit. Instead of learning the features of the planets from the books in
two-dimension, students have learned many of their features from the very beginning, such as proximity to the
sun, satellite numbers, magnitude, rotation speeds around orbits, daily temperature differences and number of
days to complete one revolution around the Sun. Moreover, students who studied the rotations of the Sun, the
Earth and the Moon with the application of “iSolarsystem” AR, could better perceive the concepts of time such as
a day, a year, a month.
“Space Research subject was taught with Spacecraft 4D and iSolarsystem AR applications in the experimental
groups. Through these applications, students were able to observe the structure of the satellites in three dimen-
sions via telescopes. During the experimental process, the students stated that they had heard the name of the
Hubble Space Telescope. They added that they were observing the sky realistically for the first time, and they felt
like astronauts and wanted to be astronauts. After four weeks of experimental process, “SSBST” and “SMSL” scales
were applied to all groups as the post-test.
367
(P. 362-378) ISSN 2538–7138 /Online/
Data Analysis
Cronbach’s alpha coefficient was used for the calculating the reliability of the scales. For the determination of
homogeneity and normality for data of obtaining from scales Levene Test, Shapiro Wilk, “Skewness and Kurtosis”
were used. For the determination of the independent and combined effect of pre-test and experimental process,
“Two-Way ANOVA for Independent Measurements”; for the interaction of pre-test and post-test in pre-test applied
groups, “Two-way ANOVA for mixed measurements (ANOVA) and one-way analysis of covariance (ANCOVA) were
used as statistical procedures. In addition, “Tukey Test” was used to determine the groups in which the significant
difference was obtained, and the “Eta Squared” value was used to determine the effect size of statistical significance.
The data collected in the research were analyzed through SPSS 21 package program.
Research Results
In the analysis of the data obtained from the research, it was first examined whether the data could meet the
basic assumptions of the parametric tests in order to decide which ones to be used. For this purpose, the Levene
test was performed to determine whether the variances of the data were homogeneous (Table 4).
Table 4. Homogeneity test results for pre-test and post-test of experimental and control groups.
Data collection tool Test Levene Test df1 df2 p
Pre-Test 3.50 1 58 .07

SSBST
Post-Test 2.31 3 116 .08
Pre-Test .04 1 58 .84
SMSL
Post-Test .51 3 116 .67
In Table 4, it is seen that the variance of the pre-test and post-test success scores of the experimental and
control groups showed a homogeneous distribution (Pre-test, F=3.50, p>.05; Post-test, F=2.31; p>.05). Likewise,
the variance of pre-test and post-test motivation scores of the experimental and control groups also showed a
homogeneous distribution (Pre-test, F=.04; p>.05; Post-test, F=.51; p>.05). It was determined that the data showed
a homogeneous distribution in terms of success and motivation variables via variance analysis. Then normality
analysis, which is another assumption of parametric tests, was done. In this research, the normality analysis of the
data was examined by considering the Skewness and Kurtosis values. According to Şimşek (2007), Skewness and
Kurtosis values should also be considered while testing the assumption of normality. Tabachnick and Fidell (2013)
indicated that the values of Skewness and Kurtosis should be within the range of +1.5 to -1.5, for admissibility
limits. The values of Skewness and Kurtosis were given in Table 5 for pre-test and post-test measurement scores
of this research.
Table 5. Values of Skewness and Kurtosis for pre-tests and post-tests of experimental and control groups.
Test
Test
Groups Skewness Kurtosis Groups Skewness Kurtosis
Experimental (1) -.12 -.70 Experimental (1) -,54 -,70

Pre Test
Pre Test
Control (1) .36 -.79 Control (1) -1.11 1.46

SSBST
Experimental (1) -.53 -.44 Experiment (1) -.33 -.59

SMSL
Post Test
Post Test
Control (1) -.38 -.72 Control (1) -.51 -.17

Experimental (2) .15 -.94 Experimental (2) -.54 1.09
Control (2) -.27 -.99 Control (2) -.99 1.41
Table 5 shows the values of Skewness and Kurtosis for the points that the research groups have taken from
the data collection tools. Considering the values determined by Tabachnick and Fidell (2013), it was decided that
368
ISSN 2538–7138 /Online/ (P. 362-378)
parametric tests should be used, because no dataset was found to exceed the range of +1.5/-1.5 and the data
showed normal distribution.
Results Related to the First Question
The first question of the study was “Is there a significant difference between students in the experimental
and control groups in terms of academic success and motivation upon application of the experimental procedure
and pre-test stimulation?”. That is to say, this question was related to examining the independent and combined
effects of the experimental procedure and pre-test. For this purpose, “Two-way Analysis of Variance (2x2 ANOVA)”
was conducted for independent measurements. Firstly, the arithmetic average and standard deviation of the SSBST
post-test scores of the research groups were examined (Table 6).
Table 6. Arithmetic mean and standard deviation values for the SSBST post-test scores of experimental and
control groups.
Data collection tool Group N X̅ SD
Experimental (1) 30 84.33 12.23

Control (1) 30 70.07 19.77
SSBST
Experimental (2) 30 81.33 13.19
Control (2) 30 69.67 16.66
As seen in Table 6, the average SSBST of Experimental group (1) after the application was X̅ post-test=84.33 and
Control group (1) X̅ post-test=70.07; Experimental group (2) X̅ post-test=81.33 and Control group (2) X̅ post-test=69.67. Two-
way ANOVA was then conducted for the SSBST post-tests of the research groups (Table 7).
Table 7. Results of two-way analysis of variance for independent measurements for SSBST post-test scores
of experimental and control groups.
Source of Variance Sum of Square df Mean of Square F p η2
Pre-test 86.70 1 86.70 .35 .56* .00

Experimental Process 5044.03 1 5044.03 20.34 .0001 .15
Pre-test * Experimental Process 50.70 1 50.70 .20 .65* .00
Total 34930.73 119
*p<.05
Without considering the experimental procedure, according to Table 7, there was no significant difference
between the groups who received and who did not receive the pre-test (F(1,116)=.35; p>.05). That is to say, the pre-
test stimulation of the students in the Experimental (1) and Control (1) groups before the experimental process,
did not lead to a significant difference in the success (SSBST) of the students in the research group. When the ef-
fect of the experimental process was examined, it was found that there was a significant difference between the
experimental group students and control group students’ academic success (SSBST) post-test scores (F(1,116)=20.34;
p<.05). In addition to the statistical significance in the difference between the groups, it could be said that the
experimental process reflected a wide range of effects (η2=.15). Tukey test was performed to understand between
which groups that significant difference was. According to this test, SSBST post-test average of the Experimental
group1 (X̅ =84.33) was significantly higher than the Control1 (X̅ =70.07) and Control2 groups (X̅ =69.67) (p<.05).
SSBST post-test average of the Experimental group2 (X̅ =81.33) was significantly higher than the Control1 (X̅ =70.07)
and Control2 (X̅ =69.67) groups (p<.05). According to this finding; it could be said that each Experimental group
differs from the control group in terms of SSBST post-test averages. There was no difference between Experimental1
(X̅ =84.33) and Experimental2 (X̅ =81.33) groups in terms of SSBST post-test scores, as Control group1 (X̅ =70.07) and
Control group2 (X̅ =69.67) (p>.05). These results showed that teaching the course with AR applications in experi-
mental groups was effective in increasing the success of the students. Moreover, it was found that the experimental
369
(P. 362-378) ISSN 2538–7138 /Online/
procedure and pre-test had no significant combined effect on students’ academic success (F(1,116)=.20; p>.05). After
conducting two-way analysis of variance for the success test, the arithmetic average and standard deviation values
of the post-test scores of the science learning motivation of research groups were examined (Table 8).
Table 8. Arithmetic mean and standard deviation values of post-test SMSL scores of the experimental and
control groups.
Data collection tool Group N X̅ SD
Experimental (1) 30 136.30 13.94

Control (1) 30 122.77 16.53
SMSL
Experimental (2) 30 131.47 15.35
Control (2) 30 125.87 17.48
As seen in Table 8, SMSL for the Experimental group (1) was X̅ post-test=136.30; Control group (1) was
X̅ post-test=122.77; the Experimental group (2) was X̅ post-test=131.47 and Control group (2) was X̅ post-test 125.87. Two-
way analysis of variance was then performed for motivation post-tests of research groups (Table 9).
Table 9. Results of two-way analysis of variance for independent measurements of post-test SMSL scores of
the experimental and control groups.
Source Sum of Square df Mean of Square F p η2
Pre-Test 22.53 1 22.53 .09 .77* .00

Experimental procedure 2745.63 1 2745.63 10.89 .0001 .09
Pre-test *Experimental Intervention 472.03 1 472.03 1.87 .17* .02
Total 32486.79 119
* p>.05
According to Table 9, there was no significant difference between the SMSL averages of the groups which
received the pre-test and did not receive the pre-test (F(1,116)=.09; p>.05). That is to say that pre-test stimulation of
the students before the experimental intervention did not lead to a significant difference in the motivation of the
students in the experimental (1) and control (1) groups. When the effect of the experimental process was examined,
it was found that the experimental group and the control group students had a meaningful difference between their
post-test scores of SMSL and this difference reflected moderate effect (F(1,116)=10.89; p<.05; η2=.09). Tukey test was
performed to see between which groups there were significant differences. Accordingly, the motivation post-test
average of the Experimental group (1) (X̅ = 136.30) was significantly higher than the Control (1) (X̅ =122.77) and
Control (2) (X̅ =125.87) groups (p<0.05). The SMSL scale post-test average of the Experimental group (2) (X̅ =131.47)
was significantly higher than the Control group (1) (X̅ =122.77) (p<0.05). According to this finding, it can be said
that each experimental group differs from the control groups in terms of motivation post-test averages. In addition,
there was no difference between SMSL scale post-test averages of Experimental (1) (X̅ =136.30) and Experimental
(2) (X̅ =131.47) groups. Nonetheless, there was no difference between SMSL scale post-test averages of Control
(1) (X̅ =122.77) and Control (2) (X̅ =125.87) groups (p>.05). These results showed that teaching the course with AR
applications in experimental groups was effective in increasing the motivation of the students. Moreover, it was
found that the experimental procedure and the pre-test had no significant common effect on students’ motiva-
tion (F(1,116)=1.87; p>.05).
Results Related to the Second Question
The second question of the research was declared as “Does the experimental procedure lead to a significant
difference between academic success and motivation pre-test and post-test scores of the groups stimulated
by the pre-test in the science course?”. That is to say that, this question was intended to determine whether the
experimental process is effective on groups received pre-test. For this purpose, “Two-way Analysis of Variance for
370
ISSN 2538–7138 /Online/ (P. 362-378)
mixed measurements (2x2 ANOVA)” was carried out to determine whether the scores of the experimental (1) and
control (1) groups received from the data collection tools significantly differed before and after the experimental
procedure. Firstly, for the SSBST, the arithmetic mean and standard deviation values for the pre-test, post-test, and
difference scores of experimental (1) and control (1) groups were examined.
Table 10. The arithmetic mean and standard deviation values for the SSBST pre-test, post-test, and difference
scores of experiment1 and control1 groups.
Pre-Test Post-Test Difference Score

Data collection
Group N
tool
X̅ SD X̅ SD X̅ SD
SSBST Experimental (1) 30 46.17 16.38 84.33 12.23 38.16 13.16

Control (1) 30 41.83 20.41 70.07 19.77 28.24 14.45
As shown in Table 10, before the application, the SSBST score of Experimental group (1) was X̅ pre-test=46.17; and
after the application, it was X̅ post-test =84.33. Before experimental process, the academic success score of the Control
group (1) was X̅ pre-test=41.83; while the post-test score was X̅ post=70.07, after the application. In addition, the differ-
ence between the post-test score and the pre-test score of the Experimental group (1) was X̅ post-test-X̅ pre-test=38.16;
while in the Control group (1), this difference was X̅ post-test-X̅ pre-test=28.24. Two-way analysis of variance for mixed
measurements for academic success was then performed (Table 11).
Table 11. Results of two-way analysis of variance for mixed measurements for SSBST pre-test and post-test
scores of experimental (1) and control (1) groups.
Between-Subjects Factors 28668.80 59

Experimental procedure 2594.70 1 2594.70 5.77 .02* .09
Within-Subjects Factors 43262.00 60
Difference between pre-test and post-test 33067.20 1 33067.20 202.85 .0001* .78
Difference between pre-test and post-test
740.03 1 740.03 4.54 .04* .07
*Experimental procedure
Total 71930.80 119
*p<,05
According to Table 11, it was seen that the SSBST scores of the experimental groups that had lessons with AR
applications and the control groups that had only lessons with the activities in the program showed a significant
difference after the experimental intervention, when compared to the start of the experiment; that is to say that
the increase of the score of the experimental group was significantly higher than the control group (F(1.,58)=4.54;
p<.05). Considering that the difference scores were in favor of the experimental group (1) (Experimental group (1),
X̅ post-test-X̅ pre test=38.16; Control group (1), X̅ post-test-X̅ pre-test=28.24); it could be said that the support of the program’s
activities with AR applications had a significant effect on the students’ success on the “Solar System” subject and this
effect was moderate (η2=.07). Since the aim of this study was to test the effectiveness of AR applications in increas-
ing students’ success, the combined effect of measurement and experimental process factors was emphasized.
In addition, the key factor tests of measurement and the experimental process were also examined. Accordingly,
the difference between the total scores of the post-test and pre-test of the experimental and control groups was
found to be significant (F(1,58) =5.77, p<.05). According to this finding, it could be said that the support of the envis-
aged program’s activities with AR applications had increased students’ success and reflected a moderate effect
(η2=.09). Finally, there was a significant difference between SSBST post-test and pre-test scores of all students in
experimental (1) and control (1) groups (without group distinction) (F(1,58) =202.85; p<.05). It could be stated that
no matter which activities are applied, the students’ success significantly increased throughout the experimental
process, from the pre-implementation to the post-implementation, and this increase reflected a broad effect
371
(P. 362-378) ISSN 2538–7138 /Online/
(η2=.78). Then, the arithmetic average and standard deviation values for SMSL pre-test, post-test and difference
scores of Experimental (1) and Control (1) groups were examined (Table 12).
Table 12. The arithmetic mean and standard deviation values for SMSL pre-test, post-test, and difference
Pre-Test Post-Test Difference Score

Scale Group N
X̅ SD X̅ SD X̅ SD
Experimental (1) 30 130.63 17.33 136.30 13.94 5.67 14.72

SMSL
Control (1) 30 121.70 18.94 122.77 16.53 1.01 21.69
As shown in Table 12, the SMSL scale score of Experimental group (1) was X̅ pre-test=130.63 before the applica-
tion, whereas X̅ post test=136.30 after the experimental process. The success score of the Control group (1) was X̅ pre-
test
=121.70, while the post- test was X̅ post-test=122.77. In addition, while the difference between post-test and pre-test
scores of experimental group (1) was X̅ post-tes-X̅ pretest=5.67, while the difference between post-test and pre-test scores
of control (1) group was X̅ post-test-X̅ pre-test=1.01. Two-way ANOVA was then conducted for mixed measurements to
determine students’ motivation for science learning and is given in Table 13.
Table 13. Results of two-way analysis of variance for mixed measurements for SMSL pre-test and post-test
Mean of
Source of Variance Sum of Square df F p η2
Square
Between-Subjects Factors 25097.30 59

Experimental procedure 3785.63 1 3785.63 10.30 .0001 .15
Within-Subjects Factors 11852.00 60
Difference between pre-test
340.03 1 340.03 1.74 .19* .03
and post-test
Difference between pre-test and post-
158.70 1 158.70 .81 .37* .01
test *Experimental procedure
Total 36949.30 119
According to Table 13, it can be seen that the SMSL scores of the experimental groups which were taught with
the AR applications and the control groups which were taught according to the activities in the envisaged program
did not show any significant difference after the experimental intervention (F(1,58)=.81; p>.05). When the difference
scores of the pre-test groups (Experimental group (1), X̅ post-test-X̅ pre-test=5.67; Control group (1), X̅ post-test-X̅ pre-test=1.01)
are taken into account; there appears to be less motivation score changes in both groups after the application.
The fact that SMSL was already high before the application, the student group was accustomed to the researcher,
and it was very difficult to change the affective variables such as motivation within a period of six weeks, may be
effective on having no significant difference. Although there was no significant difference, the increase in the dif-
ference scores in the experimental group shows that the AR applications are influential on the motivation of the
students. The difference between the sum of the post-test and pre-test SMSL scores of the experimental and control
groups was found to be significant (F(1,58) =10.30; p<.05). According to this finding, it can be said that supporting the
envisaged program’s activities with AR applications increased the SMSL and reflects a wide effect (η2=.15). Finally,
there was no significant difference between the SMSL post-test scores and pre-test scores of all the students in the
Experimental (1) and Control (1) groups (without group distinction) (F(1,58)=1.74; p>.05). This is due to the fact that
the difference between the sum of the post-test scores and the pre-test score of the groups which received the
pre-test was very small. As noted above, it is difficult to change affective variables, such as motivation, in a short
period of time, and the fact that students are already motivated in science lessons, even if they are not subjected
to any treatment, these factors can be shown to cause no significant difference.
372
ISSN 2538–7138 /Online/ (P. 362-378)
Results Related to the Third Question
The third question of the study was declared as “When pre-test is controlled, does the experimental proce-
dure make a significant difference between the academic success and SMSL post-test scores of the students in
the pre-test groups?”. That is to say, it is intended to determine whether the experimental process is effective or
not by controlling the pre-test applied to Experimental (1) and Control (1) groups. For this purpose, “Single Factor
Covariance Analysis (ANCOVA)” was conducted. In the analysis of covariance, the effect on the dependent variable,
namely pre-test, was determined as the variable to be controlled. Since the pre-test was checked on this count, the
actual effect of the experimental process was tried to be found. For the covariance analysis, pre-test and post-test
success scores of Experimental (1) and Control (1) groups showed homogenous and normal distribution and it
was identified that there was a linear relationship between pre-test success scores (control variable) and post-test
success scores (dependent variable) and that the slopes of the regression lines were equal. The post-test scores
corrected according to the pre-test scores of the Experimental (1) and Control (1) groups were given in Table 14.
Table 14. Corrected mean scores calculated by considering the SSBST pre-tests of experimental (1) and control
(1) groups.
Group N Mean Corrected Mean
Experimental (1) 30 84.33 83.43
Control (1) 30 70.33 70.97
According to Table 14, when the effect of the pre-test was controlled, the average success score of the Ex-
perimental group (1) decreased from 84.33 to 83.43, and the success score of the control group (1) increased from
70.33 to 70.97. The results of the covariance analysis (ANCOVA) conducted to understand whether the differences
between the adjusted average scores of the groups were meaningful, were given in Table 15.
Table 15. One-factor covariance analysis (ANCOVA) of the corrected post-test scores of SSBST of the experi-
mental (1) and control (1) groups.
Pre-test 3477.16 1 3477.16 16.26 .0001* .22
Total 17964.31 59
*p<.05
According to the results of the covariance analysis (ANCOVA) in Table 15, when the SSBST pre-test scores were
kept constant, it was found that there was a significant difference between the post-test scores of Experimental (1)
and Control (1) groups (F(1,57)=10.72; p<.05). These results showed that teaching the lecture with AR applications
reflects a broad effect in increasing students’ success (η2=.16). In addition, when the effect of the control variable
is examined in order to control the experimental process, it was seen that, there is a significant relation between
pre-test and post-test (F(1,57)=16.26; p<.05). Then for covariance analysis, it was determined that pre-test and post-
test SMSL scores of Experimental (1) and Control (1) groups showed homogeneous and normal distributions; it
was determined that there was a linear relationship between pre-test SMSL scores and post-test SMSL scores, and
the inclination of regression lines was equal. It has been identified. The post-test scores, which were corrected ac-
cording to pre-test scores of Experimental (1) and Control (1) groups, were given in Table 16.
373
(P. 362-378) ISSN 2538–7138 /Online/
Table 16. Adjusted mean scores calculated by considering the pre-tests of SMSL of experimental (1) and control
(1) groups.
Group N Mean Corrected Mean
Experimental (1) 30 136.47 135.14

Control (1) 30 122.77 123.93
According to Table 13, when the effect of pre-test was controlled, the SMSL average of Experimental group
(1) decreased from 136.47 to 135.14, and the SMSL average of Control group (1) increased from 122.77 to 123.93.
The results of the covariance analysis (ANCOVA) on whether the difference between the adjusted average scores
of the groups are meaningful, were given in Table 17.
Table 17. One-factor covariance analysis (ANCOVA) of SMSL scale on corrected post-test scores of experimental
(1) and control (1) groups.
Pre-test 1297.27 1 1297.27 6.03 .02* .10
Total 15325.22 59
*P < .05
According to the covariance analysis (ANCOVA) result in Table 17, when the SMSL pre-test scores were kept
constant, it was found that there was a significant difference between the post-test scores of Experimental (1) and
Control (1) groups (F(1,57)=8.24; p<.05).This suggests that teaching the course with AR applications reflects a mod-
erate effect in increasing the motivation of students (η2=.13). In addition, when the effect of the control variable
was examined in order to control the experimental process, it was seen that, there is a significant relation between
pre-test and post-test (F(1,57)=6.03; p<.05).
Discussion
Solomon four-group model enabled to determine whether pre-testing, conducted prior to the start of
the study, affects the academic success and motivation levels of students. Analysis results indicate that pre-
testing did not have a significant effect on the academic success and motivation levels of the study groups.
Similar results have been reported previously in the literature. Kartal (2013) has conducted a mixed study to
examine the impact of microteaching on the development of pedagogical subject matter knowledge of science
teacher candidates. Quantitatively, the study was based on Solomon four-group model and it was found that
pre-testing did not influence the results significantly. Doğan (2017) conducted an experimental study using the
Solomon four-group model to investigate the effect of teaching the subject “earthquake” via digital games in
the social studies course on the academic success. The results indicated that there was a significant difference
between the scores of experimental and control groups and pre-testing did not have a significant influence.
All et al. (2017) conducted a study using the Solomon method on the evaluation of game-based learning and
aimed to determine the advantages and disadvantages of pre-testing. As a result, it was determined that the
pre-test had no effect and the game-based learning was effective. However, they emphasized that the means
of pre-test groups were higher than those without pre-test and this finding should be taken into consideration
in other studies.
Although there is no significant difference between the study groups in terms of pre-test, it is observed
that the scores of the groups who are stimulated in terms of pre-test were higher when the final test scores of
the students were taken into consideration. This can be explained by the fact that students are familiar with
the structure and form of the questions. Previous studies conducted in our country using Solomon’s Four
Group Model were examined and it was observed that there was no clear information about the statistical
operations to be used in the analysis of the data of this experimental pattern. For this reason, the studies from
374
ISSN 2538–7138 /Online/ (P. 362-378)
abroad, describing Solomon’s Four Group Models, were taken into consideration. In this study, data obtained
by using Solomon’s Four-Group Model was analyzed rigorously and effort was made to transmit each step
clearly to the reader. Solomon’s four-group model, which allows the determination of whether the effect of
pre-test sensitivity is meaningful, has strengthened the internal validity and external validity of the study. The
lack of sensitivity to pre-test indicates that the changes in academic success and motivation of students are
due to the AR technology.
According to the two-way analysis of variance to determine the effect of experimental process, it has
been observed that the use of AR applications in the course process has a positive effect on students’ academic
success. One of the reasons for this positive effect is that the students encounter a new technology, find this
technology interesting, and perceive it as a magic, due to AR technology and objects appearing on the paper
in three dimensions. Previous studies also reported similar results that show that the use of AR increases the
academic success of students. Sırakaya et al. (2018) reported that the use of AR applications in the seventh
grade science class “Solar System” unit increased the academic success of students. Özarslan (2013) conducted a
quasi-experimental study and found that the AR applications have a positive influence on the academic success
of students. Baysan (2015) also reported that AR-supported learning environments increase the academic
success of students. Küçük (2015) investigated the effect of mobile AR applications on the academic success
and cognitive burden of medical students and found that the experimental group was more successful than the
control group. Chiang et al. (2014) reported that mobile-supported AR applications increase academic success
and motivation levels. Bacca et al. (2014) indicated that the AR is effective in improving the academic success.
Chien et al. (2010) found that AR training strengthens spatial memory by providing an interactive interface that
provides better anatomy learning. Maier et al. (2009) developed an AR program called “Augmented Chemical
Reactions” to teach chemical reactions in chemistry lessons, which facilitated learning of the underlying
processes and reduced the fear of chemistry.
The transfer of knowledge learned in the field of education in the direction of these studies is possible
with the interactive environment presented to the students by the AR applications. As a matter of fact, new
technologies integrated with education attract attention of students, activate them in the learning process,
increase their occupation and motivation, and help them to understand the subject more easily. The fact
that the world and the universe learning domain will be handled as the first units in the new 2017 Science
Education Program shows that this unit is now given a higher emphasis. Previously, this unit was not given the
attention it deserved due to it being the last unit and upcoming summer holiday. There is also a problem with
the implementation of the activities envisaged in the program concerned with this unit. This is because there
is no learning environment in which students can observe celestial objects. Textbooks are also insufficient in
terms of visualizing the solar system in three dimensions. The teaching of astronomy, which is usually carried
out in closed environments, is perhaps of interest to the students but obviously not very efficient. In this
study, three-dimensional visualization of astronomical phenomena that cannot be applied in a typical class
environment under normal conditions was achieved by the use of AR applications and thereby easier learning
was provided by establishing the spatial link between these concepts in accordance with the cognitive period
of seventh grade students.
When the effect of the experimental procedure on the motivation of the students is examined, it has been
observed that instruction with AR applications affects the motivation of students at a significant level. There is
also research in the literature showing that AR-enhanced learning increases the motivation of students and they
enjoy using these applications. When the studies carried out in this context are examined, Tan and Lui (2013)
found that using AR technology in teaching English vocabulary influenced the motivation of students positively.
Di Serio et al. (2013) has stated that the attention and motivation of the students increased significantly and
increase in their interaction with the learning environments has increased the motivation of the students when
AR was used in the visual arts lesson. Similarly, Mahadzir and Phung (2013) found that the use of AR technology
in books in foreign language teaching increased student motivation in vocabulary learning and Delello (2014)
stated that when prospective science teachers used AR technology in the science class, they stated that it had a
positive impact on the learning environment, increased students’ motivation and provided ease of application.
It is also important to mention some limitations of the AR technologies used in the research. These
limitations usually occur in the sense of follow-up, hardware, connection and improvement works are being
done to overcome these limitations. Especially, late-detecting of the object on the application by camera,
freezing of the camera and, the most importantly, excessive power consumption while the application is open
375
(P. 362-378) ISSN 2538–7138 /Online/
are the serious disadvantages. In order for AR technology to work properly, these technical problems need
to be addressed and for this purpose, work-force and time are needed. Although AR faces some problems, it
is possible to close this gap with the rapid development of digital and mobile technologies in recent years.
Another limitation observed in the study was that in order for the applications to operate in a healthy manner
and for each student to observe the objects in three dimensions, the study period in the experimental groups
should be longer than in the control groups. The probable reason, why students in the experimental group
finished their experience longer than others, could be that those students wish to observe the solar system
in three-dimension more than once, as well as the time needed for the technological tools to operate and for
the object recognition of the AR applications.
It is possible to say that mobile devices such as mobile phones and tablets are suitable environments for
today’s technology for AR experience. Mobile devices have advantages such as being portable, being widely
used, being easily recharged and including software projects as Play-Store and Apple-Store. When the progress
on mobile devices from past to present is examined, it is not wrong to predict that higher-capacity mobile devices
will be produced in near future. Indeed, this indicates that new AR applications will likely to be designed in
future. As such, mobile AR applications will become important material for educators. In this context, educators
can make innovative choices in the classroom, including AR applications in addition to tools such as textbooks,
board and chalk. Particularly, the AR technology will help students with weaker visual perception with their
three-dimensional thinking skills. Thanks to the AR technology, real images of virtual objects are constructed,
and stationary objects are transformed into multimedia, increasing the functionality of the learning environment.
The increasing diversity of educational technologies gives educators important duties and roles in recognizing
and using them effectively. For this reason, educators should follow the technological developments and try
to use the most appropriate tools for their fields (Akkoyunlu, 2002). The body of literature in this field shows an
increasing awareness of teachers, in particular towards the usage of AR technologies in education as well as an
increasing number of research studies. Taking this study and similar studies as references, further studies on
different topics and courses will be important to demonstrate the effectiveness of this technology in education.
Considering the compatibility of the use of mobile AR applications in teaching of other subjects in the
science course, this technology may as well be used for the teaching of other units that are similarly difficult
to visualize. The physical environments of schools should be designed in such a way that they can be utilized
in the best possible way to benefit from AR applications. Classrooms should be donated with the basic tools
and equipment required for this technology. In addition, studies using different methods and variables can
be done with a larger sample group on the effective usability of this technology. Considering that developed
countries aim to train individuals who are able to adapt technology rapidly, use technology efficiently and, most
importantly, produce technology while creating their education programs (Can & Kaymakci, 2017), the successful
integration of the AR technology into educational environments and the assessment of its advantages and
disadvantages with these studies will be useful to give an idea to educators who want to use this technology.
Finally, from the methodological point of view, it is suggested that further research is conducted to observe
the effect of pre-test and experimental procedure due to the lack of studies conducted on the use of Solomon
four-group model in science education in Turkey.
Conclusions
Thanks to the developments in mobile devices, the AR technology has been used in many sectors such
as health, marketing, education, cinema, advertising and design. Thus, it has become possible to bring AR
applications into the class and laboratory as teaching tools on mobile devices. In this study, AR applications
are thought to be effective tools to provide access and visualization of concepts that are difficult to learn
in astronomy. In this way, students will be able to discover the unknown properties of the universe and the
inaccessible places of it. Thus, students will have the opportunity to experience and learn the universe in a
realistic and better way. These experiences are expected to have a positive effect on their academic success
and increase their motivation.
The general conclusion of this research is that using augmented reality applications in science teaching
significantly contributes to the improvement of students’ achievement and motivation. Since this research was
conducted within the topic of Solar System and Beyond, the results cannot be generalized to the entire science
teaching. Of course, the use of AG technology is not the best possible solution. For future research, it will be
376
ISSN 2538–7138 /Online/ (P. 362-378)
useful to support science lessons with AG technology in order to have a significant effect on students’ science
achievement. This research was conducted by using “Solomon Four-Groups Design”, which is one of the most
powerful models that ensure internal and external validity at the same time. The Solomon four-groups design
method enabled us to obtain more valid and reliable results since the effect of pre-test and intervention could
be examined in detail. In the future, with the reporting of the successful integration of AG applications into
educational environments and the obvious advantages on the educational environment, it is expected to give
an idea and courage to educators who want to use this technology.
References
Akkoyunlu, B. (2002). Educational technology in Turkey: Past, present and future. Educational Media International, 39(2), 165-174.
All, A., Plovie, B., Castellar, E. P. N., & Van Kooy J. (2017). Pre-test influences on the effectiveness of digital-game based learning:
A case study of a fire safety game. Computers and Education, 114, 24-37.
Azuma, R. (1997). A survey of augmented reality. Presence-Teleoperators and Virtual Environments, 6(4), 355-385.
Bacca, J., Baldiris, S., Fabregat, R., Graf, S., & Kinshuk, (2014). Augmented reality trends in education: A systematic review of
research and applications. Educational Technology and Society, 17 (4), 133–149.
Baysan, E. (2015). The effect of augmented reality book (ar-book) on the students’ academic achievements and the opinions of students
about its use in educational environments (Master’s thesis). Gazi Üniversitesi, Ankara. (397402)
Chiang, T. H., Yang, S. J., & Hwang, G.-J. (2014). An augmented reality-based mobile learning system to improve students’ learning
achievements and motivations in natural science inquiry activities. Journal of Educational Technology and Society, 17(4),
352-365.
Chien, C. H., Chen, C. H., & Jeng, T. S. (2010). An interactive augmented reality system for learning anatomy structure. Proceedings
of international conference of engineers and computer scientists, (pp 370-37) Hong Kong.
Creswell, J. W. (2003). Research design: Qualitative, quantitative, and mixed methods approaches. California: Thousand Oaks, CA: Sage.
Delello, J. A. (2014). Insights from pre-service teachers using science-based augmented reality. Journal of Computers in Education,
1(4), 295–311.
Di Serio, Á., Ibáñez, M. B., & Kloos C. D. (2013). Impact of an augmented reality system on students’ motivation for a visual art
course. Computers and Education, 68, 586-596.
Doğan, E. (2017). The effect of teaching earthquake topic through digital games in social studies lesson to academic success (Master’s
thesis). Cumhuriyet Üniversitesi, Sivas. (462324)
Fraenkel, J. R., Wallen, N. E. (2009). How to design and evaluate research in education. New York: The McGraw Hill Companies.
Gün, E. (2014). Effects of augmented reality applications on students’ spatial abilities (Master’s thesis). Gazi Üniversitesi, Ankara.
(349802)
Karasar, N. (2006). Research method. Ankara: Nobel Yayın Dağıtım.
Kartal, T. (2013). The effect of microteaching on pre-service science teachers’ pedagogical content knowledge (Doctoral dissertation).
Gazi Üniversitesi, Ankara. (349124)
Kırıkkaya, E., & Şentürk, M. (2018). The impact of using augmented reality technology in the solar system and beyond unit on
the academic achievement of the students. Kastamonu Education Journal, 26(1), 181-189.
Küçük, S. (2015). Effects of learning anatomy via mobile augmented reality on medical students’ academic achievement, cognitive
load, and views toward implementation (Doctoral dissertation). Atatürk Üniversitesi, Erzurum. (389140)
Lamanauskas, V., & Bilbokaitė, R. (2009). Pedagogical evaluation of prototype 3 of the ar learning platform based on the results
achieved during the third ARISE summer school. Problems of Education in the 21st Century, 1, 86-103
Mahadzir, N., & Phung, L. F. (2013). The use of augmented reality pop-up book to increase motivation in English language learning
for national primary school. Journal of Research and Method in Education, 1(1), 26-38.
Maier, P., Klinker, G., & Tonnis M. (Eds.). (2009). Augmented reality for teaching spatial relations. Conference of the international
journal of arts and sciences. Ryerson University, Toronto.
Nunez, M. M., Quiros, R. R., Nunez, I. I., Carda, J. B., & Camahort, E. E. (2008). Collaborative augmented reality for inorganic chemistry
education. In 5th WSEAS/IASME International Conference on Engineering Education (EE’08), (pp. 271–277). Heraklion, Greece,
July 22–24
Özarslan, Y. (2013). The effect of augmented reality enhanced learning materials on learners’ achievement and learners’ satisfaction
(Doctoral dissertation). Anadolu Üniversitesi, Eskişehir. (331054)
Quintero, E., Salinas, P., González-Mendívil, E., & Ramírez, H. (2015). Augmented reality app for calculus: A proposal for the
development of spatial visualization. Procedia Computer Science, 75, 301-305.
Rosnow, R. L., & Rosenthal, R. (2002). Beginning behavioral research a conceptual primer. Englewood Cliffs, USA: Prentice Hall.
Sırakaya, Mustafa, & Kilic, Ebru. (2018). The effect of augmented reality use on achievement, misconception and course
engagement. Contemporary Educational Technology, 9(3), 297-314.
Şahin, D. (2017). Effect of science teaching with the augmented reality technology on secondary school students’ achievement and
their attitude towards the course (Master’s thesis). Atatürk Üniversitesi, Erzurum. (459513)
Şimşek, Ö. F. (2007). Introduction to structural equation modeling, basic principles and LISREL applications. Ankara: Ekinoks.
Tabachnick, B. G., & Fidell L. S. (2013). Using multivariate statistics, Boston, MA: Allyn & Bacon/ Pearson Education.
377
(P. 362-378) ISSN 2538–7138 /Online/
Tan, T., & Liu, Y. (2004). The mobile-based interactive learning environment (MOBILE) and a case study for assisting elementary
school English learning. Proceedings of the IEEE International Conference on Advanced Learning Technologies (pp. 530-534).
Joensuu, Finland: doi: 10.1109/ICALT.2004.1357471
Yılmaz, H., & Huyugüzel Çavaş, P. (2007). Reliability and validity study of the students’ motivation toward science learning (SMTSL)
questionnaire. Elementary Education Online, 6(3), 430-440.
Yusoff, R.C. Zaman H.B., & Ahmad A., (2011). Evaluation of user acceptance of mixed reality technology. Australasian Journal of
Educational Technology, 27(8), 1369-1387.
Esma Buluş Kırıkkaya PhD, Associate Professor, Kocaeli University, Education Faculty,
(Corresponding author) Department of Mathematics and Science Education, Kocaeli, Turkey.
E-mail: bulus@kocaeli.edu.tr
Melek Şentürk Başgül Science Teacher, Ministry of Education, Kocaeli, Turkey.
E-mail: melek__senturk@hotmail.com
378
RESEARCH TRENDS AND
ISSUES OF EDUCATION FOR
SUSTAINABLE DEVELOPMENT- ISSN 1648-3898 /Print/
RELATED RESEARCH IN SOUTH ISSN 2538-7138 /Online/
KOREA
Hyuksoo Kwon, Abstract. The aim of this research was

to explore research trends and issues in
Eunsang Lee Education for Sustainable Development
(ESD) in South Korea. To accomplish this,
we conducted an analysis of 222 peer-
reviewed journal papers related to ESD in
South Korea. The research concentrated
on identifying four characteristics of ESD
research: 1) key publication features,
Introduction 2) the research topics described, 3) the
research design employed, and 4) the types
Based on the benefits of science and technology, mankind has been of sample/subjects. The results showed
able to enjoy a convenient life incomparable to the lifestyles of any other that the initial ESD research began in
age. However, modern man stands face to face with gradually intensifying South Korea in 2005, and the number
problems, such as climate change, ecosystem problems, starvation, poverty, of the journal papers has increased
and the protection of human rights. It is necessary to resolve such problems, since 2010. Regarding research topics,
since these problems may affect future generations. In such circumstances, more research has been conducted on
the importance of sustainable development is being emphasized more. establishing conceptual or theoretical
The concept of “sustainable development” has transitioned into a foundations related to ESD. Moreover,
teachers’ attitudes or perceptions and
concept that comprises diverse topics since initially being mentioned in
program development associated with
a report by the World Commission on Environment and Development in
sustainable development were studied
1987 (WCED, 1987). Such sustainable development was mainly discussed
in depth. There were similar ratios of the
at the UN conference held in 1992 in Rio de Janeiro, Brazil to discuss the
theoretical research, quantitative research,
topic of “environment & development” (Chung, 2010). As Agenda 21, one
and qualitative research regarding
of the documents selected during this conference, emphasized the role research design but mixed methods studies
of education for sustainable development, such education for sustainable indicating the least ratio. Content analysis
development created a new global education paradigm. After 10 years, in was the most prevalent analysis technique
2002, at The Johannesburg World Summit on Sustainable Development held for quantitative/mixed ESD research.
in Johannesburg, the previous 10 years of how Agenda 21 was executed Regarding data sources or participants,
was reviewed. At this conference, through the Johannesburg Declaration, the most common data source was the
a resolution was reached to firmly execute sustainable development on a literature. Several research suggestions
regional/national/global level (Kim, Yoo, Shin, & Park, 2013). To accomplish were made for assessing ESD programs
this, it was repeatedly emphasized that education is an essential condition and in-service teachers’ ESD professional
for promoting sustainable development. At the US conference held in the development.
same year, the years from 2005 to 2014 were announced as the “UN Decade Keywords: ESD, South Korea, science
of Education for Sustainable Development (UNDESD)” with the purpose of education, technology education, research
executing what was agreed upon at the Johannesburg Conference (Barthes, trends.
Alpe, & Bader, 2013). Such an announcement by the UN can be seen as a
request to the world to make educational efforts to develop, reinforce, expand
and settle sustainable development.
Hyuksoo Kwon, Eunsang Lee
To achieve the goals of Education for Sustainable Development Kongju National University, Korea
(ESD), the Korean government established related institutions and various
legislations. Namely, in Korea, the Presidential Commission on Sustainable
379
RESEARCH TRENDS AND ISSUES OF EDUCATION FOR SUSTAINABLE DEVELOPMENT-
RELATED RESEARCH IN SOUTH KOREA
(P. 379-388) ISSN 2538–7138 /Online/
Development (PCSD) was established in 2000, and the institution established the ESD Execution Plan in 2006.
In 2008, a policy was promoted to materialize sustainable development in a way suitable for the circumstances
in Korea through 8.15 Green Growth Declaration. In June 2009, “ESDA Korea” was established under the Korean
National Commission for UNESCO (Yoo, Kim, Shin, & Park, 2013). In particular, in Korea, the legal grounds for ESD
were established through the Framework Act on Sustainable Development Law (2008.2.4), Environmental Education
Promotion Law (2008.9.22), Basic Act on Low Carbon Green Growth Law (2010.1.13), and Ministry of Education and
Science Technology Announcement (No. 2009-41) (Yoon, 2013).
Since prior research focused on defining the concept of ESD, it is difficult to specifically examine the information
that was studied in the developmental stages of ESD. In addition, it is necessary to analyze such research focused
on cases in Korea so that not only the flow of ESD being executed in Korea can be examined but also so that the
educational practice for ESD can be provided.
Accordingly, this research was expected to be used as the baseline data for seeking direction for ESD research
in the future through analyzing the trend of the ESD research in Korea at a time where the interest in ESD is high,
and, thereby, providing related researchers with information that is relevant to their research topics and methods.
Theoretical Background
ESD in South Korea
Through the United Nations Conference on Environment and Development held in 1992, all nations were
suggested to establish a National Commission on Sustainable Development with a purpose to actualize the Rio
Declaration and Agenda 21 (Chung, 2010). Accordingly, the President of South Korea announced the establishment
of the Presidential Commission on Sustainable Development (PCSD) in June of 2000. After the World Summit on
Sustainable Development (WSSD) was held in Johannesburg in 2002, the concept of sustainable development
started to expand globally. In June 2003, the PCSD issued the National SD Strategy Plan Report. In June 2005, the
President announced “a new nation where its economy, society, and environment make progress” as National SD
Vision. After a year, in October 2006, the National SD Execution Plan, which includes the specific execution and
follow-up plans for the National SD Vision was reported at the cabinet meeting (Chung, 2010).
Based on the efforts made by the Korean government, a number of SD-related legislations were established.
For example, in August 2007, the Framework Act on Sustainable Development was established. This act outlines
plans to participate in the global efforts for SD for the present and future generations to enjoy better lives.
A series of policies put out by the Korean government served as the baseline for ESD in Korea. When global
society announced the UN Decade of Education for Sustainable Development, Korea established a nation-wide
promotion strategy to efficiently execute ESD. Namely, in June 2005, the PCSD executed “National R&D Program
for UN Decade of Education for Sustainable Development”, and, thereby, participated in the flow of ESD. The PCSD
established the ESD Execution Plan in 2006 and promoted a policy to materialize sustainable development in a
way suitable to the circumstances in Korea through the 8.15 Green Growth Declaration in 2008. In June 2009, ESD
Korea was established under Korean National Commission for UNESCO (Yoo et al., 2013).
Such efforts were initially shown when Tongyeong was designated the initial Regional Centre of Expertise (RCE)
in Korea in October 2005. As Incheon was designated as the second RCE, ESD was concretely executed. ESD was
also applied to the national curriculums provided in Korea. For example, the Korean National Curriculum revised
in 2007 specifies to execute ESD and ESD-related education (e.g., environmental education, economic education,
energy education, education for international understanding and maritime education) in elementary and middle
schools (National Curriculum Information Center, 2008).
As described, Korea has been executing diverse ESD-related policies according to the flow of the global society,
and such policies have continued until today.
ESD-related Research Trends
Based on the trend in global society to encourage ESD, the number of ESD-related research gradually increased.
Accordingly, a number of research studies were conducted to analyze trends in such ESD-related research and
trends in the related educational significance. Initially, the studies conducted in Korea are as follows. Jeo (2012)
analyzed trends in ESD research based on the related theses and reports published in Korea over the last 10 years.
380
ISSN 1648–3898 /Print/ RESEARCH TRENDS AND ISSUES OF EDUCATION FOR SUSTAINABLE DEVELOPMENT-
ISSN 2538–7138 /Online/ (P. 379-388)
As a result, it was proposed that the number of studies dealing with ESD-related curriculum and teaching-learning
methods was greater than the number of studies dealing with other topics. In addition, it was also proposed that
a lot of these studies selected elementary school students as the subjects. Seo and Cho (2015) analyzed trends in
areas such as research periods, subject nations, content, and information, in 147 ESD-related journals and theses
related to the developmental stages of ESD. Through such analysis, it was confirmed that the number of ESD-related
studies has been continuously increasing. In addition, it was also confirmed that studies conducted overseas were
more focused on teaching methods and program development, whereas studies conducted in Korea were more
focused on theoretical areas. Kim and Choi (2014) analyzed areas of trends, such as research periods, research
content, methods, targets, and topics, in 155 ESD-related studies presented over the last 10 years. As a result, it
was proposed that the ESD studies conducted in Korea mostly consisted of qualitative and quantitative research
and that most of the ESD studies conducted in Korea selected elementary school students as the subjects.
The globally analyzed research trends in ESD are as follows. Davis (2009), having analyzed 22 ESD-related
studies that selected children as the research subjects, selected studies that included the concept of sustainable
development in the early childhood education, and divided these studies into the three categories: 1) within the
environment, 2) on the environment, 3) and for the environment. Davis (2009) confirmed that the number of
child-centric studies has been gradually increasing. Wright and Pullen (2007) analyzed trends in the ESD-related
theses published in the ESD-related academic journals from 1990 to 2005. They reported that the number of such
ESD-related theses gradually increased at a nonlinear rate, and they reported that the ESD-related research was
published in both interdisciplinary and traditional disciplinary journals. A research by Jeo (2012) was limited, in
that its uncertain description on the method used for analysis made it difficult to confirm reliability of the research.
In addition, studies by Davis (2009) and Seo and Cho (2015) were also limited, in that such studies only selected
the child-related ESD studies as the subjects. Research by Kim and Choi (2014) was limited, because it included
theses presented at symposiums. A research conducted by Wright and Pullen (2007) also showed limitations, due
to the fact that it cross-retrieved terms, such as “education,” “sustainable development,” “sustainability,” “sustain,”
and “sustainable” from Education Resources Information Center (ERIC), and, thereby, included theses unrelated to
“education of sustainable development” as intended by UNESCO.
Accordingly, it is necessary to carry out a research that can examine the flow of ESD executed in Korea and
provide direction for ESD by overcoming the limitations of previous studies.
General Description
This research explored ESD research trends for ESD-related academic journal papers published in South Korea
from 2005 to 2018. The research searched for academic papers related to ESD with keywords such as “educational
sustainability development” using websites, such as the Research Information Sharing Service (http://www.riss.kr), the
National Assembly Digital Library (http://www.dlibrary.go.kr), and DBpia (http://www.dbpia.co.kr). The searches were
performed as of January 3, 2019. First, a screening process for the searched papers was conducted by reviewing the
titles and abstracts of an individual academic papers. Additionally, selected papers were shortlisted after the first
screening process. Furthermore, among the selected research papers, conference proceedings or journal editorials
were excluded. As a result, a total of 222 papers were selected and analyzed for this research.
Criteria and Methods
The analysis criteria for the trends in Korean ESD studies have been developed by researchers by referencing
the trend analyses of prior studies (Gecit, 2010; Hallinger & Chen, 2015; Heimlich, 2007; Johnson & Daugherty, 2008;
Lee, Wu, & Tsai, 2009; McFarland, Williams, & Miciak, 2013; Williams, 2013). The criteria for analysis in this research
were four items: year, research topic, research design, and research subject.
First, year was selected to identify trends by year. Second, the research topics were selected in order to
determine what subjects were primarily dealt with in the paper. The sub-categories of research topics consist
of eight items: conceptual issues, teaching/curriculum, program development, student factors, teacher factors,
general factors, and combination of factors.
Third, research design was chosen to identify what kinds of research methods were used in ESD research
381
(P. 379-388) ISSN 2538–7138 /Online/
papers. The research design sub-category was composed of four items: quantitative research, qualitative research,
theory or conceptual research, and mixed research. For the specific analysis of these research methods, additional
analyses for the research design and statistical techniques were conducted. Specific research types of quantitative
research design were categorized as descriptive, quasi-experimental, and tools development and measurement.
Additionally, the statistical methods were categorized as descriptive statistics, testing, correlation, regression, and
factor analysis. For qualitative research, the specific methods were categorized as interview, observation, content
analysis, and mixed. The mixed category indicated qualitative research methods combining two or more qualitative
methods in this research.
Fourth, the research subject category was selected to identify the subject of research in papers. The research
subject sub-categories were kindergarten students, elementary students, middle/high school students, college
students, teachers/professors, parents, the public, combination, and documents.
Data Analysis
For this research, the researchers collected papers on ESD from August 2018 to November 2018 and examined
their contents. A preliminary analysis was conducted after establishing the analysis framework by reviewing the
previous research. The preliminary analysis was carried out by two analysts, a trend analysis research expert and
an ESD research expert, according to the analysis. The researchers randomly selected 50 papers for analysis and
calculated the agreement rate. The agreement between the analysts was 98.7%. After reviewing the results of the
analysis, researchers discussed the discrepancies, and revised the data analysis criteria by referring to the previous
research and books. The modified analysis framework was verified by one analytical expert professor. The researchers
collected additional updated papers by January 3, 2019 and analyzed the final 222 papers.
The SPSS 22.0 statistical program was used for the analysis and the results were presented as frequencies
and percentages.
Research Results
The Yearly Publication of ESD-related Research
The analysis of the yearly trends for Korean ESD is presented in Table 1. As a result of the analysis, it was
found that Korean ESD research was introduced in 2005 and fewer than 10 papers on ESD were published per year
until 2009. However, more than 10 papers were published starting in 2010, and more than 30 papers have been
published since 2014. Therefore, Korean ESD research started in 2005, and it has been confirmed that the amount
of research has gradually increased.
Table 1. Yearly publication of the ESD research.
Year N %
2005 2 0.9
2006 4 1.8
2007 2 0.9
2008 1 0.5
2009 6 2.7
2010 17 7.7
2011 13 5.9
2012 14 6.3
2013 19 8.6
2014 33 14.9
2015 31 14.0
2016 34 15.3
382
ISSN 2538–7138 /Online/ (P. 379-388)
Year N %
2017 33 14.9
2018 13 5.9
Total 222 100
Research Topics of ESD-related Research
The ESD research topic trends are presented in Table 2. Conceptual issues (38.7%) are the most popular,
followed by student factors (22.1%), teacher factors (16.7%), and so on.
Table 2. Research topic for ESD-related research.
Research topics N %
Conceptual issues 86 38.7

Student factors 49 22.1
Teacher factors 37 16.7
Program development 28 12.6
Teaching/curriculum 13 5.9
Public factor 5 2.3
Combination factor 2 0.9
Parent factor 2 0.9
Total 222 100
Research Methods of ESD-related Research
The research methods of ESD-related studies are presented in Table 3. The most popular method was the
theoretical or conceptual approach (34.7%), followed by qualitative research (30.2%) and quantitative research
(27.5%). Mixed research showed the smallest frequency (7.7%).
Table 3. Research methods for ESD-related research.
Research methods N %
Conceptual approach 77 34.7
Qualitative research 67 30.2
Quantitative research 61 27.5
Mixed research 17 7.7

Total 222 100
For a more specific analysis of the research methods, they were further analyzed to identify the research
design and statistical methods. First, the results of the analysis used in the research design in quantitative and
mixed research (a total of 78 papers) are shown in Table 4.
383
(P. 379-388) ISSN 2538–7138 /Online/
Table 4. Research design for quantitative/mixed methods research.
Research design N %
Descriptive research design 47 60.3
Quasi-experimental design 25 32.1
Tool or instrument development 6 7.7
Total 78 100
For quantitative and mixed methods research, descriptive research has been carried out the most (60.3%).
The statistical techniques used in quantitative research and mixed research are presented in Table 5. Testing has
been carried out the most (53.8 %).
Table 5. Statistical technique in quantitative/mixed methods research.
Statistical technique N %
Testing 42 53.8
Descriptive statistics 25 32.1
Factor analysis 8 10.3
Correlation 2 2.6
Regression 1 1.3
Total 78 100
The qualitative methods used in qualitative research and mixed research (a total of 84 papers) are presented
in Table 6. According to this, content analysis was carried out in 49 papers, accounting for the highest proportion
(58.3%), followed by mixed (19.0%), interview (15.5%), and observation (4.8%).
Table 6. Methods in qualitative/mixed methods research.
Methods N %
Content analysis 49 58.3

Mixed 16 19.0
Interview 13 15.5
Observation 4 4.8
Triangulation 2 2.4
Total 84 100
The Subjects or Participants of ESD-related Research
As a result of the analysis of the subjects or participants of ESD-related research, documents or literature were
used (51.4%) most frequently for the subjects of ESD-related studies. Additionally, teachers/professors (13.1%),
elementary school students (10.8%), multiple targets (8.6%), kindergarten students (5.0%), college students (5.0%),
the public (3.2%), middle/high school students (1.8%), and parents (1.4%) were recruited for research (Table 7).
384
ISSN 2538–7138 /Online/ (P. 379-388)
Table 7. Subjects or participants for ESD-related research.
Subjects or participants N %
Literature 114 51.4
Teachers/professors 29 13.1
Elementary school students 24 10.8
Combination 19 8.6
Kindergarten students 11 5.0
College students 11 5.0
Public 7 3.2
Middle/high school students 4 1.8
Parents 3 1.4
Total 222 100
In this section, the researchers analyzed the ESD-related research conducted in Korea through four items:
the year (Table 1), the topic (Table 2), the research method (Table 3), and the research subject (Table 7). Additional
analysis was also performed for the sub items (Table 4-Table 6) of the research method, and the results are presented.
Discussion
Research related to ESD in Korea was initially introduced in 2005. Fewer than 10 research studies were
conducted every year until 2009 and more than 10 studies were conducted per year after 2010. Namely, the
number of studies related to ESD was quite small in the early stages until a significant increase was seen in 2010.
Such results agree with the results the research by Kim and Choi (2014) that analyzed trends in ESD-related
research in Korea. In Korea, such research was initially introduced in Korea when the concept of ESD was being
globally adapted due to the announcement of US Decade of Education for Sustainable Development at the UN
conference in 2002. From this point, UNESCO and various overseas nations started to discuss ESD. However, even
after the concept of ESD was introduced in Korea, not that many studies related to ESD were conducted in Korea.
This was not only because teachers and Education Offices were unaware of ESD, but also because the competitive
entrance examination system in Korea made it difficult for middle schools to introduce ESD. The number of studies
related to ESD started to increase from 2010 because the Korean government and educational institutions started
making efforts to establish ESD-related policies and legislations from 2008 and also because the Korean National
Curriculum was revised in 2007 to emphasize the importance of ESD. Based on such background, as the results of
this research show, the amount of ESD-related research rapidly increased starting in 2010.
Secondly, the conceptual issue was mainly selected (38.7%) as the research topic in the ESD-related research
conducted in Korea. This is because Korea was quite unaware of the concept of ESD at the time ESD was introduced,
and, therefore, a number of studies were conducted to introduce the concept of ESD. In addition, many studies
were conducted to establish ESD-related educational strategies or to explore the possibility of ESD in Korea. Some
researchers introduced how ESD was executed in other countries, and some researchers focused on exploring
the ESD-related factors used in the textbooks and teacher’s textbooks in Korea. Prior studies with conceptual
issues provided a robust foundation for further research and practices, and the studies on students’ attitudes
toward sustainable development (22.1%), teachers’ perceptions (16.7%), and program development (12.6%) were
performed. It turned out that there were largely studies on the subjects and target for ESD. Comparatively, research
subjects, including parents, pre-service teachers, and communities, were not enough. For an efficient diffusion of
ESD, more studies on its value and significance should be conducted for parents or the public. Therefore, further
studies need to expand on research subjects like parents, the public, and pre-service teachers.
Thirdly, as a result of analyzing the studies related to ESD in Korea, there were relatively balanced figures
among qualitative research (34.7%), the conceptual approach (30.2%), and quantitative research (27.5%), in terms
of research design. However, the portion of mixed research design was relatively low, and most prior studies did not
385
(P. 379-388) ISSN 2538–7138 /Online/
clarify their research design. Globally, the mixed method design, combining quantitative and qualitative methods,
has been actively conducted (Kim, Kim, & Lee, 2011). Therefore, further studies using the mixed method design
should be performed for better interpretation and findings via the triangulation of collected data.
In this research, an additional analysis was conducted according to the research method. Initially, a total of
78 theses, consisting of 61 quantitative studies and 17 mixed method studies that used the quantitative method,
were selected as the subjects for additionally analysis of research design and statistical techniques. As a result of
analyzing the research design method used in the selected theses, it was discovered that the descriptive research
design method was used most frequently. This is because a number of studies conducted an ESD awareness
survey that targeted teachers and students or examined the actual status of ESD as applied in schools. As a result
of analyzing the statistical techniques used in the selected theses, it was found that the testing technique and
descriptive statistics technique were used the most frequently. On the other hand, tool or instrument development
was the least frequently used research design method. Accordingly, in future research, it is necessary to remodel
ESD-related test tools developed by other countries or newly develop a test tool that is suitable to circumstances
in Korea (Lee & Cha, 2016). In addition, it was found that only a few studies were conducted based on the methods
of statistical technique, factor analysis, and correlation and regression. This signifies that only a few studies were
conducted to validate the related test tools and examine the influential relationship of related factors. Accordingly,
it is necessary to conduct research that examines the influential relationship between variables related to the
subjects of science and technology (Na, 2015) and variables related to ESD or research that examines the structural
relationship between factors.
In this research, a total of 84 theses, consisting of 67 qualitative studies and 17 mixed methods studies, were
selected as the subjects to additionally analyze the qualitative research method. As a result, it was found that
content analysis was used most frequently. The researches that used such content analysis were mainly focused
on analyzing ESD-related factors presented in kindergarten/elementary school/middle school textbooks, teacher
textbooks, and curriculum provided in Korea. However, the interview method and observation method were used in
a relatively low number of quantitative studies. It seems that this is because the Korean researchers preferred using
the content analysis method, since this method allows a relatively convenient analysis of contents in comparison
to other qualitative research methods in Korea, most of the qualitative research that analyzes the cases where ESD
is applied selected children as the research subjects (Kim, Kim, & Yoo, 2016). Accordingly, it is necessary to conduct
research that qualitatively analyzes the cases where ESD is applied to middle school science/technology classes.
Fourthly, as a result of analyzing the subjects of this research, it was found that the literature covered the highest
percentage. As mentioned earlier, this is because ESD was introduced in Korea as a new education paradigm, and,
therefore, a number of research studies focused on examining related literature. One particular point confirmed
through this research is that almost no studies selected middle/high school students as the subjects. As pointed
out by Lee et al. (2005) and other researchers, this is because the entrance examination-based education system
in Korea made it difficult to apply ESD in a middle/high school context. To overcome this issue, middle school
teachers are required to enhance their capability to execute ESD. Namely, the difficulties arising out of educational
environments can be resolved through middle school teachers’ passion and capability to execute ESD. In particular,
the subjects of science and technology in Korea include many areas of ESD. Accordingly, it is necessary to provide
science/technology teachers and pre-service teachers with ESD-related training for them to acquire the ability to
include ESD in their curriculum. Moreover, it is necessary to provide support by integrating an ESD-related education
program into the subjects of science and technology so that ESD can be conveniently executed in schools.
Conclusions
In this research, ESD research published from 2005 to 2018 in journals from South Korea were reviewed and
analyzed to identify trends in ESD-related studies. The criteria of analysis in this research were four items: year,
research topic, research design, and research subject.
First, ESD research began in 2005, and the number of ESD journal papers has gradually increased since 2010 in
South Korea. It is a reasonable result that the Korean government accepted the ESD idea and supported educational
researchers who studied ESD.
Second, the analysis of research methods shows that theoretical or conceptual research has been the
most common method. South Korean professionals need a more stable foundation in ESD, and many studies
have presented the philosophy and direction of ESD. Additionally, the proportion of factor analysis, correlation,
386
ISSN 2538–7138 /Online/ (P. 379-388)
and regression used in statistical techniques has been very small. This means that the development of tools or
instruments to measure the effectiveness of ESD has not been fully achieved yet. In addition, this means that studies
measuring the relationships between related variables have not yet been carried out in the field of ESD. Therefore,
it is necessary to perform further research to adapt and develop various tools related to ESD. Moreover, research
needs to be conducted to determine the relationships between various science and technology educational
factors related to ESD.
Third, as a result of the analysis of the subjects or participants, the document category was found to be the
most common category. Many initial research studies have explored and introduced literature on ESD-related
content in South Korea. However, research targeting middle school or high school students is scarce. This means
that the practical and broad implementation of ESD targeting middle school or high school students in Korea
has been insufficient. Therefore, further research should focus on the development and application of programs
targeting these students.
Through this research, the researchers identified the trends of ESD research, methods of research, and subjects
of research. Specifically, the majority of the ESD concepts were strongly associated with the learning content of
science and technology curriculum. Therefore, science and technology educators need to relate the content in
their science and technology courses to ESD. For example, in Korea, ESD content started being presented as one
chapter in the middle school technology textbook from 2018. In addition, in middle school science textbooks,
ESD content is presented in a separate section in supplementary materials (reading materials, in-depth materials).
Researchers in countries that have not yet introduced ESD in science and technology courses need to add ESD in
their science and technology curriculum by referring to Korea’s approach.
Acknowledgements
This work was supported by the research grant of the Kongju National University (2016-0254-01)”
References
Barthes, A., Alpe, Y., & Bader, B. (2013). Questions and positions on education for sustainable development at university in France:
Example of short professional cycles. Environmental Education Research, 19(3), 269-281.
Chung, K. S. (2010). Emergence and development of education for sustainable development in Germany. Theory and Practice
of Education, 15(3), 153-173.
Davis, J. (2009). Revealing the research ‘hole’ of early childhood education for sustainability: A preliminary survey of the literature.
Environmental Education Research, 15(2), 227-241.
Gecit, Y. (2010). Primary tendencies in research on geography education. Educational Sciences: Theory and Practice, 10(2), 969-987.
Hallinger, P., & Chen, J. (2015). Review of research on educational leadership and management in Asia : A comparative analysis
of research topics and methods, 1995–2012. Educational Management Administration & Leadership, 43(1), 5-27.
Heimlich, J. E. (2007). Research trends in the United States EE to ESD. Journal of Education for Sustainable Development, 1(2), 219-227.
Jeo, E. H. (2012). Research trends in domestic ESD: Focused on the theses and the reports for about 10 years. Global Studies
Education, 4(2), 83-95.
Johnson, S. D., & Daugherty, J. (2008). Quality and characteristics of recent research in technology education. Journal of Technology
Education, 20(1), 16-31.
Kim, E. J., Yoo, Y. E., Shin, E. S., & Park, E. H. (2013). An inquiry into and implication of school education and early childhood
education for sustainable developments in Germany, England and Australia. The Journal of Korea Open Association for Early
Childhood Education, 18(4), 75-100.
Kim, J. S., Kim, E. J., & Yoo, H. O. (2016). Meaning of education for sustainable development experienced by children in the process
of inquiring a theme. The Journal of Child Education, 25(4), 161-185.
Kim, S. M., & Choi, D. S. (2014). The analysis on research trends related to education for sustainable development in Korea. Korea
Environmental Education Association Conference Catalog, 2014(12), 216-217.
Kim, Y. C., Kim, K. S., & Lee, H. C. (2011). Mixed research methods in educational research. The Journal of Elementary Education,
24(1), 305-328.
Lee, M. H., Wu, Y. T., & Tsai, C. C. (2009). Research trends in science education from 2003 to 2007: A content analysis of publications
in selected journals. International Journal of Science Education, 31(15), 1999-2020.
Lee, S. G., Lee, J. Y., Lee, S. C., Lee, Y. J., Min, G. S., Sim, S. G., ... Ha, G. R. (2005). The present status and implementation strategy of
education for sustainable development in Korea. Korea Environmental Education Association Conference Catalog, 2005(8),
11-23.
Lee, Y. S., & Cha, S. K. (2016). Research of development and validation of ecofriendly attitude scales for young children based on
the education for sustainable development. Korean Joural of Children‘s Media, 15(1), 133-169.
387
(P. 379-388) ISSN 2538–7138 /Online/
McFarland, L., Williams, J., & Miciak, J. (2013). Ten years of research: A systematic review of three refereed LD journals. Learning
Disabilities Research & Practice, 28(2), 60-69.
National Curriculum Information Center. (2008). Primary and middle school curriculum. Retrieved from http://ncic.go.kr/.
Na, Y. R. (2015). Child care teacher’s perception and attitude on environmental section of education for sustainable development.
Journal of Holistic Convergence Education, 19(3), 41-60.
Seo, H. J., & Cho, B. K. (2015). An analysis on trends of studies on early childhood education for sustainability. Korean Journal of
Early Childhood Education, 35(6), 241-269.
WCED. (1987). Our common future. Oxford, Oxford University Press.
Williams, P. J. (2013). Research in technology education: Looking back to move forward. International Journal of Technology and
Design Education, 23(1), 1-9.
Wright, T., & Pullen, S. (2007). Examining the literature a bibliometric study of ESD journal articles in the education resources
information center database. Journal of Education for Sustainable Development, 1(1), 77-90.
Yoo, Y. E., Kim, E. J., Shin, E. S., & Park, E. H. (2013). Analysis of the Korea educational policy and current curriculum based on the
education for sustainable development. Early Childhood Education Research & Review, 17(3), 319-341.
Yoon, J. H. (2013). A critical reflection and acception on education for sustainable development (ESD) in terms of „labor“: From
the focus on the 2009 national education curriculum of “practical art” subject. Journal of Korean Practical Arts Education,
19(1), 23-52.
Received: March 15, 2019 Accepted: May 20, 2019
Hyuksoo Kwon Ph.D. in Curriculum & Instruction, Associate Professor,

Kongju National University, 56, Gongjudaehak-ro, Gongju-si,
Chungcheongnam-do, Republic of Korea.
E-mail: hskwon@kongju.ac.kr
ORCID: 0000-0002-4013-1115.
Eunsang Lee Ph.D. in Technology Education, Assistant Professor, Kongju National
(Corresponding author) University, 56, Gongjudaehak-ro, Gongju-si, Chungcheongnam-do,
Republic of Korea.
E-mail: eslee@kongju.ac.kr
ORCID: 0000-0001-6932-6699.
388
LEARNING ENVIRONMENT,
STUDENTS’ BELIEFS,
AND SELF-REGULATION ISSN 1648-3898 /Print/
IN LEARNING PHYSICS: ISSN 2538-7138 /Online/
STRUCTURAL EQUATION
MODELING
Abstract. Studies of correlations

Maison, between students’ beliefs and various
Syahrial, aspects of student learning become one
of the fastest growing research areas in
Syamsurizal,
the field of education and psychology.
Tanti The aim of the current research was to
analyze the correlations between learning
environments, students’ beliefs, and self-
regulation in learning physics through
structural equation modeling (SEM). There
Introduction were 1010 students from the existing
five public high schools in Jambi city,
Physics is a branch of science that has its uniqueness and characteristics. Indonesia, participating in the research.
The uniqueness of physics lies in the existence of concepts that are abstract Three self-report questionnaires including
and require idealization through mathematical modeling. This makes physics (1) WIHIC, (2) CLASS, and (3) MSLQ were
conceptually justified as a difficult subject, both to be learned and taught used to collect the research data. The data
(Hofer & Pintrich, 1997; Mulhall & Gunstone, 2012). Some of the main obstacles analysis showed that students’ beliefs
faced by students in learning physics include: the low ability to explain the were significantly and positively correlated
with multiple dimension of self-regulation
principles of physics qualitatively (McDermott, 1993), misconception (Duit,
in learning physics (critical thinking
Niedderer, & Schecker, 2007), the low ability in solving physics problems (May
and peer learning); while the dimension
& Etkina, 2002), low conceptual understanding (McDermott, 1993; Osborne,
of sense-making and problem-solving
Simon, & Collins, 2003), and low motivation as well as low active involvement
ability significantly related to the affective
in learning physics (Tran, 2012).
component of self-regulation (test anxiety).
Researchers in the field of science education try to explore the factors However, the dimension of students’
that affect students’ learning process, especially in the field of physics by beliefs did not have any significant effect
not only submitting observational evidence, but also involving a multi- on all of the self-regulation components.
perspective framework to understand, describe, and convey the role of a Additionally, learning environment
social and individual aspect on students’ learning process (Otero, 2003). dimensions were significantly related to
One of the individual aspects that plays an important role in the process students’ beliefs about physics on the
of knowledge construction is a set of beliefs that students have about the dimension of conceptual connection
characteristics of knowledge and how to acquire that knowledge (May & and related to all of the self-regulation
Etkina, 2002). Youn (2000) defined beliefs as implicit assumptions held by dimensions.
students about the source and certainty of knowledge and how to obtain it. Keywords: learning environment, self-
This means that beliefs can be identified as a reference to learning, whereby regulation, students’ beliefs, structural
knowledge acquired by students is generated from the cognitive process. equation modeling.
Fishbein and Ajzen (2010) explained that these beliefs have a significant
impact on individuals’ attitudes, and finally, these attitudes influence
Maison, Syahrial, Syamsurizal
individuals’ behaviors. Jambi University, Indonesia
Hammer (1994b) categorized students’ beliefs about physics and Tanti
physics learning in a continuum on three aspects, namely (1) beliefs about Sultan Thaha Saifuddin State Islamic
University, Indonesia
the structure of physical knowledge (pieces - coherence), (2) beliefs about
389
LEARNING ENVIRONMENT, STUDENTS’ BELIEFS, AND SELF-REGULATION IN LEARNING
PHYSICS: STRUCTURAL EQUATION MODELING
(P. 389-403) ISSN 2538–7138 /Online/
the content of physics knowledge (conceptual understanding - formula), and (3) beliefs about the physics learning
process (authority - independent). Hammer (1994a) described that students with beliefs that the structure of
physics knowledge is coherent, emphasizing on conceptual understanding, and learning as a process of applying
and modifying knowledge itself, tended to have better performance in solving physics problems. Contrary to that,
students who have beliefs that physics is a collection of facts and formulas that must be memorized tend to fail to
replace misconceptions with scientific ideas and have the low reasoning ability (Qian & Alvermann, 2000). Similarly,
Sahin (2010) noted that students with sophisticated beliefs at the beginning of the semester tend to obtain higher
concept comprehension scores at the end of the semester than students with negative beliefs.
Research Problems
Students’ beliefs can be constructed, changed and strengthened. According to Tsai (2000) how the teacher
explains the scientific concept and organizes information plays a vital role in the construction of students’ beliefs.
Furthermore, Tsai (2000) explained that students desire a learning environment which provides opportunities
to interact with one another, integrate their prior knowledge and experience in the learning process, think
independently and solve problems related to everyday life. In the same vein, Madsen, McKagan, and Sayre (2015)
stated that teaching method and strategies which explicitly focus on developing models from the world of physics,
including instructional strategies centered on inquiry activities, modeling instruction, physics, and everyday
thinking resulted in a positive shift on students’ beliefs in pre-test and post-test. This learning model involves
students working in small groups to experiment and obtain evidence to build models from the world of physics.
Previous studies report that students’ beliefs will influence academic performance (Cano, 2005), motivation (Lin,
Deng, Chai, & Tsai, 2013), self-efficacy and attitude (Kapucu & Bahçivan, 2015), and learning strategies (Dahl, Bals,
& Turi, 2005). Research findings by Dahl et al. (2005) showed that the less students believe knowledge is organized
in a complex system (naive beliefs), the more they tend to report using rehearsal strategies, and the less they tend
to report using organization and metacognitive strategies. Although all of the previous studies have explored the
correlation between students’ beliefs with various learning outcomes, none of these studies have explored the
correlation between learning environment, students’ beliefs, and self-regulation simultaneously. Therefore, the
main purpose of this research was to study about the correlation between learning environment, students’ beliefs,
and self-regulation (motivation and cognitive and metacognitive component) in learning physics.
Based on the aforementioned research, it can be concluded that beliefs are a determining factor for students’
success in understanding and applying physics in everyday life. As Kortemeyer (2007) said students with positive
beliefs were students who understood the characteristics and process of construction of physics knowledge,
and were able to monitor, evaluate, and improve the learning process. Nevertheless, students’ beliefs could be
constructed, changed, and strengthened. Research in education showed the pivotal key that played an important
role in beliefs construction was the learning environment (Madsen et al., 2015; Ozkal, Tekkaya, Cakiroglu, & Sungur,
2009; Tsai, 2000). In addition, research has proven that students’ beliefs influenced their motivation and their learning
approach (Cano, 2005; Kapucu & Bahçivan, 2015; Ozkal et al., 2009; Tsai, Jessie Ho, Liang, & Lin, 2011).
Research Focus
Although the relation between learning environment, students’ beliefs, and learning approaches has been
analyzed in a wealth of studies, few studies have been done to explore the interaction between these variables in
the physics domain. Moreover, no research about which psychosocial factors of learning environment have the
most influence on beliefs’ construction and students’ self-regulation, as well as how these beliefs influence students’
self-regulation (motivation component and cognitive and metacognitive component) in learning physics has been
conducted. Therefore, the aim of this research was to explore a model of structural correlations between the learning
environment, students’ beliefs and self-regulation in learning physics, as guided by the following research questions:
1. Which psychosocial factors of the learning environment have the most salient influence on students’
beliefs about physics and learning physics?
2. Which psychosocial factors of the learning environment have the most salient influence on students’
regulation in learning physics, both in motivation and learning strategy aspects?
3. Which dimensions of student’ beliefs have the most prominent influence on students’ self-regulation
in learning physics?
390
ISSN 1648–3898 /Print/ LEARNING ENVIRONMENT, STUDENTS’ BELIEFS, AND SELF-REGULATION IN LEARNING
ISSN 2538–7138 /Online/ (P. 389-403)

General Background
The conceptual framework of this research is based on the cognitive social theory that forms the basis for the
development of constructive and cooperative learning models. According to the cognitive social theory, learning
process occurs due to the reciprocal triadic between personal factors (students’ beliefs), external factors (learning
environment), and behaviors (self-regulation). Based on this existing theoretical framework, the researchers
recommended a research model as seen in Figure 1.
Figure 1. Research model.
The researcher’s assumptions in this research were as follows: (1) psychosocial factors of classroom learning
environment positively correlate with students’ beliefs about physics; (2) students’ beliefs positively correlate with
the use of self-regulation in learning physics, (3) learning environment positively correlates with students’ self-
regulation in learning physics.
The researchers used the What is Happening in This Class (WIHIC) questionnaire to assess students’ perception
of the classroom learning environment (Aldridge, Fraser, & Huang, 1999; Fraser, Fisher, & McRobbie, 1996). The
WIHIC consists of seven dimensions of psychosocial factors of the learning environment (cohesiveness, teacher
support, investigation, involvement, task orientation, cooperation, and equity) that are developed based on human
environmental theory initiated by Moos and Trickett (1987). In the theory, Moos et al. (1987) divided the human
environment into three dimensions, namely relationship, personal development, and system maintenance and
change. Students’ cohesiveness, teacher support, and involvement fell into the relationship dimension according
to Moos’ scheme. While investigation, task orientation, and cooperation are parts of the personal development
dimension. The last component, equity was a part of the system maintenance and change dimension.
Students’ beliefs about physics and learning physics were assessed using The Colorado Learning Attitudes
about Science Survey (CLASS) which was developed by Adams, Perkins, Dubson, Finkelstein, and Wieman ( 2005).
The CLASS is developed based on the other established instruments that measure students’ attitudes and beliefs
about physics, such as the Maryland Physics Expectation Survey (MPEX), the Views about Science Survey (VASS),
the Epistemological Beliefs Assessment, and Fishbein’s theory of attitudes (Ajzen & Fishbein, 1977). The original
version of CLASS consists of eight categories, namely: real-world connection, personal interest, sense-making/
effort, conceptual connections, applied conceptual understanding, problem-solving (general), problem-solving
(confidence), and problem-solving (sophistication). In the current research, the researchers used CLASS in Bahasa
Indonesia version which was adapted and modified by Tanti et al. (2018). The Indonesian version of CLASS
questionnaire comprises three categories, namely personal interests, sense-making & problem-solving, and
conceptual connections.
391
(P. 389-403) ISSN 2538–7138 /Online/
The Motivated Strategies for Learning Questionnaire (MSLQ) developed by Pintrich, Smith, Garcia, and
McKeachie (1991) was used to assess students’ self-regulation in learning physics. The MSLQ is based on a general
social-cognitive view of motivations and learning strategies, in which students represent as active agents of
knowledge construction. The MSLQ consists of two categories, namely motivations which are divided into six
components (intrinsic goal orientation, extrinsic goal orientation, task value, control beliefs, self-efficacy, and test
anxiety) and learning strategies which are divided into nine components (rehearsal, elaboration, organization,
critical thinking, metacognitive self-regulation, time and study environment, effort regulation, peer learning, and
help-seeking).
Based on Figure 1, the researchers predicted that the seven components of the psychosocial learning
environment (cohesiveness, teacher support, investigation, task orientation, cooperation, equity, and involvement)
are positively and significantly related to students’ beliefs component (personal interest, sense-making & problem-
solving skills, conceptual connection). In addition, the researchers predicted that each component of the learning
environment is positively and significantly related to the self-regulation in learning physics, both in the motivation
components (intrinsic goal orientation, goal orientation, task value, control of learning beliefs, self-efficacy, anxiety
test), as well as cognitive and metacognitive strategy components (rehearsal, elaboration, organization, critical
thinking, metacognitive, time & research environment, effort regulation, peer learning, and help-seeking).
Sample
The respondents of this research involved 1,010 students from grade XI of the existing five public high schools
in Jambi City, Indonesia. All students registered in grades XI participated in the survey. Before the data analysis,
the researchers “cleaned and accounted” the data from errors and those of uncompleted ones (Creswell, 2012). The
process of the data cleaning included checking the students’ responses on each item in the research instrument
to make sure that all statements were completed by the respondents; the process of re-ranking of each negative
statement was also conducted simultaneously. As a result, 1,003 respondents were considered eligible for the
following data analysis phase. Table 1 summarizes the number of respondents based on genders:
Table 1. Respondents’ demography based on genders.
Gender Number Percentage
Male 346 34.5

Female 657 65.5
Total 1003 100
Instruments, Procedures, and Data Analysis
The researchers used three self-report instruments: CLASS to assess students’ beliefs about physics and learning
physics (Adams et al., 2006), WIHIC to assess students’ perception about the learning environment (Aldridge et al.,
1999) and MSLQ ( Pintrich, Smith, Garcia, & McKeachie, 1991) to assess students’ self-regulation. The CLASS and MSLQ
questionnaires were originally developed in English language. To obtain a valid and reliable measurement used in the
context of learning in Indonesia, the CLASS and MSLQ questionnaires were first adapted into the Indonesian version.
The translation process of these both questionnaires was conducted through standard translation methodology,
which includes translation, verification, and modification. All of the items were translated into Bahasa Indonesia.
The result was then validated qualitatively by two lecturers from the Department of Physics Education (bilingual).
The feedbacks provided by the validators were adopted as the basis for improving the translation versions of the
questionnaires.
For this research, the researchers conducted two stages of the data analysis, exploratory factor analysis (EFA),
and confirmatory factor analysis (CFA). Hair Jr, Hult, Ringle, and Sarstedt (2017) explained that EFA and CFA tests
cannot be done by using the same data-set as it would amount to mere data fitting rather than testing theoretical
constructs. Hence, the researchers divided the data into two parts, namely “odd” and “even” data. The odd data-
392
ISSN 2538–7138 /Online/ (P. 389-403)
set was used for factor analysis (EFA) test using SPSS version 21.0, while even data-set was used for confirmatory
factor analysis (CFA) test using PLS-based SEM (PLS-SEM). Table 2 summarizes the “odd” and the “even” research
data based on the respondents’ genders.
Table 2. Odd and even data based on genders.
Odd Even
Gender
Number Percentage Number Percentage
Male 167 33.2 179 35.80

Female 336 68.8 321 64.20
Total 503 100 500 100
The EFA was conducted on the CLASS questionnaire to measure students’ beliefs about physics and learning
physics. The EFA test on the CLASS questionnaire was performed for several reasons. First, the original version of
the CLASS questionnaire consists of 41 statement items, in which 26 of the 41 are grouped into eight overlapping
dimensions. The eight dimensions are real-world connections, personal interest, sense-making and effort, conceptual
connections, applied conceptual understanding, problem-solving general, problem-solving confidence, and
problem-solving sophistication. This means one item can fit into two or more dimensions. For example, item #11:
“I am not satisfied until I understand why something works the way it does”. This item falls into two dimensions,
namely personal interest and sense-making / efforts. The same tendency occurs in other items, indicating that
the resulting constructs are not unidimensional. While the remaining 16 items are not categorized into the
previous eight dimensions, because they have not received a response from experts. Second, the evaluation of
the psychometric characteristics of the CLASS questionnaire was only performed by one researcher (Douglas, Yale,
Bennett, Haugan, & Bryan, 2014). Douglas et al. (2014) stated that based on EFA and CFA analysis, there are 15 items
of validity statement and 26 categories of the CLASS questionnaire which are categorized into three dimensions,
namely personal application and relation to the real world, problem-solving/ learning, and effort/sense-making.
Based on these reasons, the researchers decided to conduct an analysis with the same stages as to the CLASS
questionnaire to obtain a valid and reliable instrument, which is used to measure students’ beliefs on physics and
physics learning by the context of socio-cultural conditions of Indonesia.
The second test conducted in this research was Confirmatory Factor Analysis (CFA). The CFA test was performed
to analyze the convergent validity of beliefs dimension produced through the EFA test, as well as to evaluate the
internal reliability of the MSLQ questionnaire used to measure the students’ self-regulation in learning physics. The
CFA test was also conducted to analyze the structural model of the correlation between the three latent variables
of the research, i.e., learning environment, beliefs, and self-regulation of students in researching physics.
The researchers also conducted EFA and CFA instrument validation. Cronbach’s alpha and composite reliability
for each instrument were examined. Finally, the researchers performed structural equation modeling (SEM), which
is based on variance (PLS-SEM) to analyze the fit of the proposed model in Figure 1. The PLS approach is Asymptotic
Distribution Free (ADF), meaning that the analyzed data do not possess a certain distribution pattern, it can be
nominal, category, ordinal, interval, and ratio.
Research Results
The initial assumption tests showed that the value of Kaiser-Meyer-Olkin was .862 and of Bartlett’s Test of
Sphericity was 2916.252, which was statistically significant (p<.001). Both values indicated that initial requirements
for factor analysis were fulfilled; since the value of KMO was >.5 and that of Bartlet’s Test of Sphericity was <
.05 (Pallant, 2011). Principal component analysis with orthogonal rotation (varimax) on 26 items of the CLASS
questionnaire resulted in 5 factors of students’ beliefs with Eigen value of >1 and total variance of 46.259%. However,
the analysis of the scree plot showed fractures over the three dimensions, as shown in Figure 2:
393
(P. 389-403) ISSN 2538–7138 /Online/
Figure 2. Screeplot of students’ beliefs.
Based on Figure 2 above, it was decided to take the three factors of students’ beliefs with a total variance of
40.896%. The examination of communalities indicated that there were five items that had a value of < 0.3; those
were items #1, #5, #10, #25, and #37. According to Cabrera-Nguyen (2010) communalities value of < 0.3 shows the
low relation between the items with the dimensions formed and should be excluded from the test; the remaining
21 items were included in the retest of factor analysis. Loading factors and reliability of each dimension can be
seen in Table 2.
Table 2. Loading factor and reliability of items from class questionnaire.
Component
Items
Personal Interest Sense-Making & Problem Solving Conceptual Connection
23 .649
33 .642
35 .635
13 .611
21 .585
32 .570
12 .558
3 .554
30 .540
14 .529
25 .482
22 .681
15 .641
7 .603
20 .574
24 .460
394
ISSN 2538–7138 /Online/ (P. 389-403)
Component
Items
Personal Interest Sense-Making & Problem Solving Conceptual Connection
28 .680
31 .641
36 .592
19 .568
6 .564
Eigenvalue 4.670 2.196 1.723
% Variance 22.236 10.457 8.203
Cumulative % 22.236 32.694 40.896
Reliability values .813 .643 .619
The next step was to analyze the convergence and discriminant validity on the 21 items of CLASS resulting
from EFA. The researchers also checked the convergence validity of MSLQ items. Convergent validity measures the
magnitude of the correlation between constructs with latent variables, including individual item reliability, internal
consistency, and average variance extracted (AVE). Individual item reliability was seen from the value of standardized
loading factor. According to Hair Jr et al. (2017) the loading factor value of ≥ .7 is said to be ideal, meaning that the
indicator validly measures the constructs it establishes. Another opinion is put forward by Haryono (2017), based
on empirical research, that the value of loading factor ≥ .5 is acceptable. Thus, the loading factor value of ≤ .5 must
be dropped from the research model. The CFA test result showed that out of the total of 21 items of the CLASS
questionnaire, there were several items that had a loading factor value of ≤ .5 so that they were dropped from
the model (item #3, #12, #13, #14, #19, #20, #21, #23, #27, #33, and #36) — resulted in ten valid items of the CLASS
questionnaire, which were categorized into three dimensions, namely personal interests (3 items), sense-making
& problem solving (4 items), and conceptual connection (3 items). Table 4 shows the item loadings, composite
variance, and average variance extracted of each of the CLASS and MSLQ dimensions.
Table 3. Item loadings, composite variance, and average variance extracted.
Average Variance Extracted

Latent Variable Item Loading Factor Composite Reliability
(AVE)
13 .861
Personal Interest 25 .658 .502 .746
33 .577
7 .655
15 .735
Sense Making &Problem Solving .517 .810
22 .757
24 .715
6 .657
Effort & Real-World Connection 28 .817 .564 .794
31 .769
1 .707
16 .749
Intrinsic Goal Orientation .545 .827
22 .734
24 .763
7 .725
11 .747
Extrinsic Goal Orientation .559 .835
13 .800
30 .715
395
(P. 389-403) ISSN 2538–7138 /Online/
Average Variance Extracted

Latent Variable Item Loading Factor Composite Reliability
(AVE)
10 .659
17 .670
Task Value 23 .752 .544 .856
26 .808
27 .786
2 .750
9 .640
Control of Learning .515 .808
18 .761
25 .712
12 .726
15 .736
20 .675
Self-Efficacy for Learning & Performance .507 .860
29 .719
31 .717
6 .698
8 .894
Test Anxiety 14 .743 .522 .754
28 .463
39 .711
46 .715
Rehearsal .519 .811
59 .709
72 .683
53 .669
64 .741
Elaboration 67 .736 .507 .837
69 .754
81 .655
32 .731
42 .854
Organization .546 .826
49 .669
63 .688
38 .710
47 .778
Critical Thinking .507 .804
51 .681
71 .676
36 .723
41 .705
44 .733
Metacognitive 54 .662 .502 .876
55 .739
56 .735
61 .659
35 .734
Time & Research Environment 43 .793 .557 .790
73 .709
48 .858
Effort Regulation .688 .815
74 .800
34 .685
Peer Learning 45 .761 .547 .783
50 .769
58 .693
Help-Seeking 68 .766 .520 .764
75 .702
396
ISSN 2538–7138 /Online/ (P. 389-403)
Based on Table 3, all values of loading factors are ≥ .5, so it can be concluded that the validity of the CLASS
questionnaire is good at the item level. Furthermore, at the level of construction, the composite value of reliabil-
ity is high with values ranging from .74 to .86. According to Hulland (1999), the value of the composite reliability
limit is the same as Cronbach’s alpha ≥ .7. The higher the composite reliability (CR) value, the higher the contribu-
tion of the construct in the measurement model. The final criterion of convergent validity is the measurement
of the average variance extracted (AVE) for each construct. The AVE value describes the variance or variability
of the manifest variables that the latent construct can have (Haryono, 2017). Hair Jr et al. (2017) recommends a
minimum of 0.5 AVE to indicate an excellent convergent validity measure. Based on Table 4 above, AVE values for
all components are above the minimum value, ranging from 0.5 to 0.6. So, based on the value of loading factor,
composite reliability, and AVE, it can be concluded that both CLASS and MSLQ have good convergence validity.
Discriminant validity measures the extent to which latent variable constructs are empirically different (Aldridge,
Afari, & Fraser, 2013). Table 4 and 5 show that the square root of the average variance (AVE) for each construct is
larger than inter-construct correlation.
Table 4. Discriminant validity of CLASS questionnaire.
Latent Variable CC PI SM & PS
CC .751
PI -.080 .709
SP&PS -.213 .416 .719
Table 5. Discriminant validity of MSLQ questionnaire.
Lat.
CL CT ER Ela EG HS IG Met Org PL Reh SE TO TV TA TSE
Var
CL .717
CT .419 .712
ER .419 .551 .829
Ela .423 .692 .581 .712
EG .632 .330 .429 .328 .748
HS .418 .549 .611 .628 .402 .721
IG .587 .593 .414 .565 .387 .397 .738
Met .440 .726 .533 .735 .283 .553 .616 .709
Org .408 .690 .529 .725 .336 .528 .626 .739 .771
PL .448 .640 .515 .608 .369 .546 .578 .694 .684 .739
Reh .417 .707 .568 .703 .333 .605 .569 .713 .691 .628 .720
SE .569 .611 .454 .599 .478 .419 .723 .623 .650 .628 .593 .712
TO .276 .313 .245 .303 .282 .266 .361 .309 .346 .259 .306 .397 .712
TV .647 .599 .467 .599 .439 .431 .781 .627 .635 .629 .593 .738 .353 .737
TA .527 .353 .314 .310 .571 .375 .405 .335 .321 .343 .321 .448 .226 .402 .722
TSE .485 .588 .621 .601 .455 .603 .524 .627 .636 .635 .644 .591 .364 .562 .388 .746
Note: the bold value in the diagonal are the square roots of average variance extracted, cc=conceptual connection, pi=personal
interest, sm &ps = sense-making & problem-solving ability, cl=control of learning, CT=critical thinking, ER=Effort Regulation,
Ela=Elaboration, EG=Extrinsic Goal, HS=Help Seeking, IG=Intrinsic Goal, Met=Metacognitive, Org=Organization, PL=Peer Learning,
Reh=Rehearsal, SE=Self Efficacy, TO=Task Orientation, TV=Task Value, TSE=Time & Research Environment
397
(P. 389-403) ISSN 2538–7138 /Online/
The amount of influence between constructs and interaction effects (moderation) is measured by the value
of the coefficient path (path coefficient). Path coefficient that has statistic value of ≥ 1.96 or has p-value of ≤ .05
expresses significance. The results indicate that 30 of the 168 possible correlation are statistically significant (p <
.05) and all of the statistically significant correlation are positive in direction. The Path coefficient and t-value for
each hypothesis are positive and significantly related in the research model, as shown in Table 6 below:
Table 6. Path coefficient.
Hypotheses t -values p values Conclusion
Cohesiveness -> Conceptual Connection 2.391 .017 Significant

Cooperation ->Peer Learning 2.642 .008 Significant
Cooperation -> Self Efficacy 1.993 .047 Significant
Equity -> Conceptual Connection 2.379 .018 Significant
Investigation ->Critical Thinking 1.995 .046 Significant
Investigation -> Elaboration 2.175 .030 Significant
Investigation ->Intrinsic Goal 3.904 .001 Significant
Investigation ->Metacognitive 3.178 .002 Significant
Investigation -> Organization 2.195 .028 Significant
Investigation ->Self Efficacy 1.999 .046 Significant
Involvement ->Peer Learning 1.984 .050 Significant
Involvement ->Time & Research Environment 2.172 .030 Significant
Personal Interest->Critical Thinking 2.249 .025 Significant
Personal Interest->Peer Learning 1.906 .057 Significant
Sense-Making & Problem-Solving ability->Test Anxiety 1.921 .055 Significant
Task Orientation -> Control of Learning 2.388 .017 Significant
Task Orientation -> Critical Thinking 3.312 .001 Significant
Task Orientation -> Effort Regulation 2.505 .012 Significant
Task Orientation -> Elaboration 3.103 .002 Significant
Task Orientation -> Extrinsic Goal 2.789 .005 Significant
Task Orientation -> Help-Seeking 2.879 .004 Significant
Task Orientation ->Intrinsic Goal 4.653 .000 Significant
Task Orientation -> Metacognitive 2.851 .004 Significant
Task Orientation -> Organization 3.837 .000 Significant
Task Orientation -> Rehearsal 3.219 .001 Significant
Task Orientation -> Self Efficacy 5.183 .001 Significant
Task Orientation -> Task Value 4.975 .000 Significant
Task Orientation -> Test Anxiety 2.327 .020 Significant
Task Orientation -> Time & Research Environment 4.345 .001 Significant
Teacher Support -> Conceptual Connection 2.211 .018 Significant
Teacher Support -> Control of Learning 1.980 .050 Significant
Teacher Support ->Extrinsic Goal 2.249 .025 Significant
Teacher Support ->Test Anxiety 2.407 .016 Significant
From Table 6, the researchers can conclude that three out of seven learning environment scales (cohesiveness,
equity, and teacher support) most likely influence students’ beliefs, especially on conceptual connection dimen-
sion. While, the learning environment scales that are closely related to students’ self-regulation in learning physics
is investigation, task orientation, and teacher support. The findings also indicate that teacher support is likely to
correlate both students’ beliefs and self-regulation in learning physics. Additionally, all dimensions of students’
beliefs significantly correlate with both motivation scale (test anxiety) and learning strategy scales (critical thinking
and peer learning) of students’ self-regulation in learning physics. All these statistically significant correlations are
represented in the model as in Figure 3.
398
ISSN 2538–7138 /Online/ (P. 389-403)
Figure 3. The t-value of the Structured Model (t > 1.96).
Discussion
This research aimed to analyze the structural correlations between the learning environment, students’ be-
liefs, and self-regulation in learning physics. The CLASS questionnaire developed by Adams et al. (2006) is used to
measure student beliefs about physics and learning physics. Before using, the CLASS questionnaire was adapted
into the Indonesian version. The validation process through the EFA and CFA obtained three factors of students’
beliefs, namely problem-solving abilities (4 items), conceptual understanding (3 items), and effort & real-world
connection (3 items). The values of composite reliability for the three dimensions of the beliefs were .812, .805
and .754 respectively for the factors of problem-solving ability, conceptual understanding, and effort & real-world
connection. The result of the validation of the Indonesian version of the CLASS questionnaire was in line with the
analysis conducted by Douglas et al. (2014). The evaluation of the psychometric factor of the CLASS questionnaire
conducted by Douglas et al. (2014) resulted in 15 valid items categorized into 3 dimensions, namely personal ap-
399
(P. 389-403) ISSN 2538–7138 /Online/
plication and relation to real world (6 items), problem-solving (5 items), and effort/sense-making (4 items), with
Cronbach alpha reliability values of .80, .73, and .69, respectively. The difference in naming the constructs and the
number of valid items were more due to the different cultural contexts that could have been understood differently.
The MSLQ questionnaires were used to measure students’ self-regulation in learning physics. The research-
ers adopted 15 self-regulatory scales of The Motivated Strategies for Learning Questionnaires (MSLQ) developed
by Pintrich et al. (1991). The 15 scales are categorized into two major components of self-regulation, namely the
components of motivation (6 scales) and components of cognitive and metacognitive strategies (9 scales). The CFA
test revealed the values of the composite reliability of 15 self-regulation scale ranging from 0.719 - 0.876. Similar
results were reported by Pintrich et al. (1991), with internal Cronbach’s alpha reliability values ranging from 0.52
to 0.93. Based on the findings, it was concluded that the Indonesian versions of CLASS and MSLQ were valid and
reliable to be used to measure beliefs and students’ self-regulation in learning physics.
The relation between the learning environment, students’ beliefs, and self-regulation in learning physics were
analyzed using structural equation-based variance modeling (PLS-SEM). The evaluation of the structural model
showed that there was a positive and significant correlation between learning environment, students’ beliefs, and
self-regulation, although this positive and significant relationship did not occur in all scales of research variables.
The scales of the learning environment that has a positive and significant influence on the formation or construction
of beliefs were cohesiveness, equity, and teacher support. These three scales of the learning environment had a
positive and significant effect on the formation of beliefs, especially in the conceptual connection dimension. The
conceptual connection measures the extent to which the beliefs the students have for the coherence of a physics
topic with other physics topics.
Hammer (1994a) explained that students who believe of physics knowledge structures are coherent, will
emphasize their learning on conceptual understanding, and thus tend to have better performance in solving
physics problems. Based on the result of structural relationship analysis, it can be concluded that students’ per-
ceptions toward the three aspects of the psychosocial learning environment, namely cohesiveness, equity, and
teacher support, have a positive and significant effect on their beliefs about the conceptual connection structure.
It means that students will be encouraged to be actively involved in the learning process when they feel there is
good social acceptance by peers, help from each other, and teachers’ support in their learning process. Students
will not be embarrassed to be laughed at by their classmates when they make mistakes in working on questions
or tasks assigned by teachers; besides, students will not hesitate to ask teachers when they encounter challenges
in understanding the concepts of physics.
The results of this research are in line with the studies conducted by Ozkal et al. (2009) and Tsai (2000). From
the results of their research, Ozkal et al. (2009) found that students who have students’ perceptions of the classroom
environments (including personal relevance, uncertainty, critical voice, and student negotiation) are positively
correlated with students’ beliefs that scientific knowledge is tentative. Further Ozkal et al. (2009) explained that
the classroom learning environment, which connects the learning process of science with the daily experiences of
students, provides opportunities for students to share their science ideas with others, as well as to provide hands-on
experience to students through inquiry activities in laboratories closely with the formation of tentative beliefs, i.e.
beliefs that scientific knowledge is tentative or evolving. The same thing was stated by Tsai (2000), that students
with perceptions of the classroom learning environment are constructivist, giving students the opportunity to
discuss with peers and integrate the students’ initial knowledge with new knowledge taught by teachers, tend to
have beliefs that science knowledge is constructivist.
Concerning with the relationship between students’ beliefs and self-regulation in learning physics, the cur-
rent research demonstrated that there is a positive relationship between students’ beliefs (personal interest and
sense-making/ problem-solving ability) and self-regulation both in learning strategy components (critical think-
ing and peer learning) and motivation component (test anxiety). The dimension of students’ beliefs, i.e. “personal
interest” has a positive and significant effect on the learning strategy components of self-regulation, namely critical
thinking, and peer learning. Critical thinking describes the extent to which students apply prior knowledge to new
situations in order to solve problems, while peer learning describes students’ perceptions of learning experience
where students share knowledge and discuss ideas (Pintrich et al., 1991).
According to Redish (1997) and Halloun and Hestenes (1998), because beliefs are closely related to students’
perceptions about the characteristics of knowledge and how to obtain that knowledge, personal interest is an
important part that will shape students’ beliefs about physics and the physics learning process. In other words,
students will have strong beliefs if they realize that studying physics provides benefits for them, because it is
relevant to what they experience in everyday life, and the skills needed to understand physical concepts such
400
ISSN 2538–7138 /Online/ (P. 389-403)
as reasoning abilities which will be useful in their lives; while students will have naïve beliefs, if they believe that
studying physics is not related to what they experience in everyday life.
The research findings above explain if students have sturdy beliefs on personal interests, they tend to feel more
challenging and have the confidence to understand and solve complicated physics problems. These findings are
consistent with the results of the previous research in which tentative beliefs were related to meaningful learning
strategies, and they were mainly motivated by their interests and curiosity about science (Tsai, 1998). Additionally
Chan, Ho, and Ku (2011), on their research about the relationship between epistemic beliefs and critical thinking
of Chinese students showed that epistemic beliefs had specific effects on critical thinking and cognitive ability.
Further, Chan et al. (2011) explained that students having beliefs that knowledge is being fixed and absolute (cer-
tain knowledge) tend to have poor performance in everyday evaluative thinking and reduced cognitive ability.
The research findings also show that the task orientation and investigation are the components of the learning
environment which contribute the most significant influence to the students’ self-regulation in learning physics,
both in the motivation and learning strategy component. Task orientation assesses the extent to which students
perceive that it is crucial to complete the subject tasks and understand the goals of the subject. According to hu-
man environment theory proposed by Moos et al. (1987), task orientation is part of the dimension of “personal
growth” which emphasizes the accessibility (opportunity) of students to develop themselves and improve self-
quality (self-enhancement) through various aspects such as achievement, competition, autonomy, and personal
status (Velayutham, Aldridge, & Afari, 2013). These results indicate that teachers should pay attention to students’
learning objectives and ensure that students understand what is needed to complete the task. The findings support
Velayutham et al. (2013) the suggestion that students need to be aware of the importance of completing planned
activities and stay focus on the subject matter.
Finally, the findings of the current research have also revealed that there are three learning environment scales
that are significantly related to students’ beliefs about physics. These factors are students’ cohesiveness, equity,
and teacher’s support. The findings also show that students’ beliefs correlate with their self-regulation in learning
physics. Another important result from the current research is that learning environment directly correlates with
students’ self-regulation in learning physics both in motivational and learning strategies component.
Conclusions
The purpose of the research was to study about the structural correlations between learning environment,
students’ beliefs, and self-regulation in physics learning. The findings show that three out of the seven psychosocial
factors in the learning environment (cohesiveness, equity, and teacher support) have a positive and significant
impact on students’ beliefs about physics and learning physics. The implication of these findings is that teachers
need to pay attention to how to create a conducive learning environment that facilitates students to be able to
work together and respect each other’s opinions. Also, the teacher must provide equal attention to students, not
to discriminate students, and support students to be successful in learning. The results revealed are significant in
that the current study is one of the few studies conducted in Indonesia applying structural equation modeling
(SEM) based on variance (PLS-SEM) which has commonly been used to develop a comprehensive model of correla-
tion between individual aspects of students (student trust), learning environment, and self-regulation in physics
learning. This research will contribute positively to all education stakeholders, curriculum developers, teacher
educators, teachers, and students. The results provide new insights to understand why a curriculum designed to
address students’ difficulties in understanding the concept of learning is not effective for some students. Information
obtained from this research can be a theoretical foundation for curriculum developers in designing and developing
science education curricula, especially in the field of physics so that the implementation of teaching and learning
can facilitate students to gain direct, contextual, and student-centered experience. Finally, the current research
will provide an important insight for physics teachers to explore students’ prior knowledge as a representation of
their beliefs about physics so that teachers can design learning processes that can facilitate shifting of students’
beliefs from naive beliefs to expert beliefs.
References
Adams, W. K., Perkins, K. K., Dubson, M., Finkelstein, N. D., & Wieman, C. E. (2005). The design and validation of the colorado learn-
ing attitudes about science survey. AIP Conference Proceedings, 790, 45–48. https://doi.org/10.1063/1.2084697.
401
(P. 389-403) ISSN 2538–7138 /Online/
Adams, W. K., Perkins, K. K., Podolefsky, N. S., Dubson, M., Finkelstein, N. D., & Wieman, C. E. (2006). New instrument for measuring
student beliefs about physics and learning physics: The colorado learning attitudes about science survey. Physical Review
Special Topics - Physics Education Research, 2(1), 1–14. https://doi.org/10.1103/PhysRevSTPER.2.010101.
Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological
Bulletin, 84(5), 888.
Aldridge, J. M., Afari, E., & Fraser, B. J. (2013). Influence of teacher support and personal relevance on academic self-efficacy and
enjoyment of mathematics lessons: A structural equation modeling approach. Alberta Journal of Educational Research,
58(4), 614–633.
Aldridge, J. M., Fraser, B. J., & Huang, T.-C. I. (1999). Investigating classroom environments in Taiwan and Australia with multiple
research methods. The Journal of Educational Research, 93(1), 48–62. https://doi.org/10.1080/00220679909597628.
Cabrera-Nguyen, P. (2010). Author guidelines for reporting scale development and validation results in the Journal of the Society
for Social Work and Research. Journal of the Society for Social Work and Research, 1(2), 99–103. https://doi.org/10.5243/
jsswr.2010.8.
Cano, F. (2005). Epistemological beliefs and approaches to learning: Their change through secondary school and their influence on
academic performance. British Journal of Educational Psychology, 75(2), 203–221. https://doi.org/10.1348/000709904X22683.
Chan, N. M., Ho, I. T., & Ku, K. Y. L. (2011). Epistemic beliefs and critical thinking of Chinese students. Learning and Individual Dif-
ferences, 21(1), 67–77. https://doi.org/10.1016/j.lindif.2010.11.001.
Creswell, J. W. (2012). Educational Research: Planning, conducting, and evaluating quantitative and qualitative research. In P. A.
Smith (Ed.), Pearson (4th ed.). 501 Boylston Street, Boston: Pearson Education.
Dahl, T. I., Bals, M., & Turi, A. L. (2005). Are students’ beliefs about knowledge and learning associated with their reported use of
learning strategies? British Journal of Educational Psychology, 75(2), 257–273. https://doi.org/10.1348/000709905X25049.
Douglas, K. A., Yale, M. S., Bennett, D. E., Haugan, M. P., & Bryan, L. A. (2014). Evaluation of Colorado Learning Attitudes about
Science Survey. Physical Review Special Topics - Physics Education Research, 10(2), 1–10. https://doi.org/10.1103/PhysRevS-
TPER.10.020128.
Duit, R., Niedderer, H., & Schecker, H. (2007). Teaching physics. In S. K. Abell & N.G. Lederman (Ed.), Handbook of research on sci-
ence education (pp. 599–629). London: Routledge.
Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: The reasoned action approach. 270 Madison Avenue, New York:
Psychology Press, Taylor & Francis Group.
Fraser, B. J., Fisher, D. L., & McRobbie, C. J. (1996). Development, validation and use of personal and class forms of a new classroom
environment instrument. In The Annual Meeting of the American Educational Research Association. New York.
Hair Jr, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2017). A primer on partial least squares structural equation modeling (PLS-
SEM) (2nd ed.). Los Angeles: SAGE Publication.
Halloun, I., & Hestenes, D. (1998). Interpreting VASS dimensions and profiles for physics students. Science & Education, 7(6), 553–577.
Hammer, D. (1994a). Epistemological beliefs in introductory physics. Cognition and Instruction, 12(2), 151–183.
Hammer, D. (1994b). Students’ beliefs about conceptual knowledge in introductory physics. International Journal of Science
Education, 16(4), 385–403. https://doi.org/10.1080/0950069940160402.
Haryono, S. (2017). Metode SEM untuk penelitian manajemen: Amos, lisrel & PLS [SEM method for management research: Amos,
lisrel, & PLS]. Jakarta: PT. Luxima Metro Media.
Hofer, B. K., & Pintrich, P. R. (1997). The development of epistemological theories: Beliefs about knowledge and knowing and
their relation to learning. Review of Educational Research, 67(1), 88–140. https://doi.org/10.3102/00346543067001088.
Hulland, J. (1999). Use of partial least squares (PLS) in strategic management research: A review of four recent studies. Strategic
Management Journal, 20(2), 195–204.
Kapucu, S., & Bahçivan, E. (2015). High school students’ scientific epistemological beliefs, self-efficacy in learning physics and
attitudes toward physics: A structural equation model. Research in Science and Technological Education, 33(2), 252–267.
https://doi.org/10.1080/02635143.2015.1039976.
Kortemeyer, G. (2007). Correlations between student discussion behavior, attitudes, and learning. Physical Review Special Topics
- Physics Education Research, 3(1), 1–8. https://doi.org/10.1103/PhysRevSTPER.3.010101.
Lin, T. J., Deng, F., Chai, C. S., & Tsai, C. C. (2013). High school students’ scientific epistemological beliefs, motivation in learning
science, and their relationships: A comparative study within the Chinese culture. International Journal of Educational De-
velopment, 33(1), 37–47. https://doi.org/10.1016/j.ijedudev.2012.01.007.
Madsen, A., McKagan, S. B., & Sayre, E. C. (2015). How physics instruction impacts students’ beliefs about learning physics: A
meta-analysis of 24 studies. Physical Review Special Topics - Physics Education Research, 11(1), 1–19. https://doi.org/10.1103/
PhysRevSTPER.11.010115.
May, D. B., & Etkina, E. (2002). College physics students’ epistemological self-reflection and its relationship to conceptual learning.
American Journal of Physics, 70(12), 1249–1258. https://doi.org/10.1119/1.1503377.
McDermott, L. C. (1993). Guest comment: How we teach and how students learn ̶ A mismatch? American Journal of Physics,
61(4), 295–298. https://doi.org/10.1119/1.17258.
Moos, R.H., & Trickett, E. J. (1987). Classroom environment scale manual (2nd ed.). Palo Alto, CA: Consulting Psychologists Press.
Mulhall, P., & Gunstone, R. (2012). Views about learning physics held by physics teachers with differing approaches to teaching
physics. Journal of Science Teacher Education, 23(5), 429–449. https://doi.org/10.1007/s10972-012-9291-2.
Osborne, J., Simon, S., & Collins, S. (2003). Attitudes towards science: A review of the literature and its implications. International
Journal of Science Education, 25(9), 1049–1079. https://doi.org/10.1080/0950069032000032199.
402
ISSN 2538–7138 /Online/ (P. 389-403)
Otero, V. K. (2003). Cognitive processes and the learning of physics part I: The evolution of knowledge from a Vygotskian per-
spective. Enrico Fermi Summer School in Physics, 409–445. Retrieved from papers2://publication/uuid/FF463D3B-83CA-
4440-8FD5-0D79E0564284.
Ozkal, K., Tekkaya, C., Cakiroglu, J., & Sungur, S. (2009). A conceptual model of relationships among constructivist learning envi-
ronment perceptions, epistemological beliefs, and learning approaches. Learning and Individual Differences, 19(1), 71–79.
https://doi.org/10.1016/j.lindif.2008.05.005.
Pallant, J. (2011). SPSS Survival Manual. A step by step guide to data analysis using SPSS (4th ed.). Crows Nest, NSW: Allen & Unwin.
Pintrich, P. R., AF Smith, D., Garcia, T., & McKeachie, W. J. (1991). A manual for the use of the Motivated Strategies for Learning Ques-
tionnaire (MSLQ). Michigan: National Center for Research to Improve Postsecondary Teaching and Learning.
Qian, G., & Alvermann, D. E. (2000). Relationship between epistemological beliefs and conceptual change learning. Reading &
Writing Quarterly, 16(1), 59–74. https://doi.org/10.1080/105735600278060.
Redish, E. F. (1997). Measuring student expectations in university physics: The MPEX survey. College Park, MD: Physics Education
Research Group.
Sahin, M. (2010). The impact of problem-based learning on engineering students’ beliefs about physics and conceptual under-
standing of energy and momentum. European Journal of Engineering Education, 35(5), 519–537. https://doi.org/10.1080/
03043797.2010.487149.
Tanti, Maison, Mukminin, A., Syahrial, Habibi, A., & Syamsurizal. (2018). Exploring the relationship between preservice science
teachers ’ beliefs and self-regulated strategies of studying physics : A structural equation model. Journal of Turkish Science
Education, 15(4), 79–92. https://doi.org/10.12973/tused.10247a.
Tran, V. D. (2012). Predicting the attitudes and self-esteem of the grade 9th lower secondary school students towards mathematics
from their perceptions of the classroom learning environment. World Journal of Education, 2(4). https://doi.org/10.5430/
wje.v2n4p34.
Tsai, C.-C. (2000). Relationships between student scientific epistemological beliefs and perceptions of constructivist learning
environments. Educational Research, 42(2), 193–205.
Tsai, C. C. (1998). An analysis of scientific epistemological beliefs and learning orientations of Taiwanese eighth graders. Science
Education, 82(4), 473–489. https://doi.org/10.1002/(SICI)1098-237X(199807)82:4<473::AID-SCE4>3.0.CO;2-8.
Tsai, C. C., Jessie Ho, H. N., Liang, J. C., & Lin, H. M. (2011). Scientific epistemic beliefs, conceptions of learning science and self-
efficacy of learning science among high school students. Learning and Instruction, 21(6), 757–769. https://doi.org/10.1016/j.
learninstruc.2011.05.002.
Velayutham, S., Aldridge, J., & Afari, E. (2013). Application of structural equation modeling in educational research and practice.
In M. S. Khine (Ed.), Application of Structural Equation Modeling in Educational Research and Practice (pp. 115–133). https://
doi.org/10.1007/978-94-6209-332-4.
Youn, I. (2000). The culture specificity of epistemological beliefs about learning. Asian Journal of Social Psychology, 3(1), 87–105.
https://doi.org/10.1111/1467-839X.00056.
Maison PhD in Science Education, Associate Professor, Faculty of Education and

Teacher Training, Jambi University, Jambi, Indonesia.
E-mail: maison@unja.ac.id
ORCID: https://orcid.org/0000-0002-4020-3051
Tanti MSc, Lecturer, Faculty of Education and Teacher Training, Sulthan Thaha
Saifuddin State Islamic University, Jambi, Indonesia.
E-mail: tanti.tjang@gmail.com
Syahrial PhD, Associate Professor, Faculty of Education and Teacher Training,
Jambi University, Jambi, Indonesia.
E-mail: syahrial.fkip@unja.ac.id
Syamsurizal Dr., Associate Professor, Faculty of Education and Teacher Training,
Jambi University, Jambi, Indonesia.
E-mail: syamsurizal68@unja.ac.id
403
A MODEL OF INTEREST IN STEM
CAREERS AMONG SECONDARY
SCHOOL STUDENTS
Abstract. Studies have shown that vari- Lilia Ellany Mohtar,

ous factors (the role of formal education, Lilia Halim,
informal education, and community)
contribute to the lack of participation in
Norshariani Abd Rahman,
STEM and STEM careers. This research Siti Mistima Maat,
aimed to understand the multi pathways Zanaton H. Iksan,
of factors contributing to the interest in Kamisah Osman
STEM careers (STEM careers in physical
sciences and STEM careers in life sciences).
This research was a survey research which
administered a questionnaire randomly
to 1485 secondary school students (14
years of age). Data analysis was based on
the Structural Equation Modeling (SEM) Introduction
approach using Analysis of Moment
Structures (AMOS) to test the hypothesized With the advent of Fourth Industrial Revolution (4IR), workers with skills
model. A model containing five constructs, in STEM are considered important more than ever. Not only knowledge and
namely environmental factors (learning skills in STEM are important for driving innovation but these knowledge
experiences, media, social influences), self- and skills are important in order to survive the impact of 4IR. It is projected
efficacy in science, technology, engineering (MOSTI, 2015) that Malaysia needs 500,000 workers in STEM in 2020. However,
and mathematics respectively, perceptions involvement in STEM among students in Malaysia is declining, not only in
of STEM careers and interest in physical schools but in higher education institutions as well (Halim & Subahan, 2016).
sciences and life sciences STEM careers was Nonetheless, this problem is not unique to Malaysia as it is also faced by other
proposed in this research. The results show countries. For instance, America announced having a tough time filling STEM
that students’ interest in life sciences based careers which are unfilled because of a lack of participation in STEM fields
careers is influence by their self-efficacy (Wyss, Heulskamp, & Siebert, 2012).
and perceptions of the career. Meanwhile, According to Bahar and Adiguzel (2016) and Hawley, Cardoso and Mc-
students’ interest in physical sciences based Mahon (2013), 20 percent of all vacancies in America require STEM related
careers is influence only by their self-effica- experts and some reports have indicated that there is a potential of 2.8 million
cy and not influence by their perceptions of job openings by 2018. Turkey likewise encounters a similar situation (Bahar &
the career. The need to improve students’
Adiguzel, 2016; Maltese, 2008) where 18 percent of all national vacancies in
self-efficacy through STEM learning experi-
the country are for STEM-related jobs. The disparity in growth in the gradu-
ences is imperative to ensure continued
ate and STEM-related vacancies demonstrates a widening gap, indicating
interest in STEM careers.
the increasing demand for STEM experts. In other words, the chances of
Key words: environmental factors, life
employments are higher for STEM graduates. However, while the demand
sciences STEM careers, perceptions of STEM
careers, physical sciences STEM careers, self-
for workers specializing in STEM-related fields is continuously growing, the
efficacy, social cognitive career theory. enrollment at the graduate level is consistently recording a decline, leading
to gaps in employment.
Tai, Liu, Maltese, and Fan (2006) argued that students tend to make
decisions about their career at middle or lower secondary school. According
Lilia Ellany Mohtar, Lilia Halim,
Norshariani Abd Rahman, to Eccles et al. (2003), teenagers have a tendency to investigate new things
Siti Mistima Maat, Zanaton H. Iksan, including making plans for their future. However, these students may not get
Kamisah Osman enough exposure to STEM careers. Thus, the students may make decisions
National University of Malaysia, Malaysia
about their career based on insufficient information of the career. Therefore, it
404
ISSN 1648–3898 /Print/ A MODEL OF INTEREST IN STEM CAREERS AMONG SECONDARY SCHOOL STUDENTS
(P. 404-416)
is imperative that students are provided with accurate information and appropriate guidelines about STEM careers
so that these lower secondary school students would be able to make effective decisions about their career choice
and path. Along these lines, it is therefore critical to encourage students’ attitudes, interest and perception that
would lead to career aspirations in STEM-related careers (Wyss, Heulskamp, & Siebert, 2012).
There are many STEM careers offered to graduates, for example physics, chemistry, biology, zoology, environ-
mental works, and computer science to name a few. In this research, the students’ choice in selected STEM occu-
pation is seen as students’ interest in STEM careers which is divided into two categories, namely physical sciences
STEM careers and life sciences STEM careers. Students’ interest in STEM careers is affected by various factors and
many studies have investigated the factors that influence interest in STEM careers (Bahar & Adiguzel, 2016; Franz-
Odendaal, Blotnicky, French, & Joy, 2016; Kauffmann, Hall, Batts, Bosse, & Moses, 2009). This research continues
the research of previous studies by adopting most of the factors investigated earlier but at the same time extends
the research to the context of a developing country, specifically Malaysia where STEM careers are still viewed as
a prestigious career. However, students are shying away from these STEM careers and one possible new factor is
related to students’ and parents’ perceptions of the prospect in STEM careers.
Literature Review
Various studies have examined the factors influencing interest in STEM careers. For example, Bahar and Adiguzel
(2016) examined the factors affecting students pursuing profession in STEM based disciplines. They employed the
Social Cognitive Career Theory (SCCT) theorized by Lent, Brown and Hackett (1994) as their theoretical framework.
In Bahar and Adiguzel’s (2016) research, the factors selected were founded on the hypothesis that students’ interests
in STEM jobs are formed by major constructs including self-efficacy, outcome expectations, and goal orientation
as suggested by SSCT. Apart from these major constructs, there are sub-constructs which include people (family,
teachers, peers, relatives, etc.), school-related factors (courses, clubs, competitions/fairs, classroom activities, outside
classroom activities, etc.), self-motivation and occupational expectations.
Moreover, prior studies have recognized a few important factors that directly and/or indirectly influence stu-
dents’ interest in STEM careers. For example, student’s own self-motivation, the quality of science and mathematics
courses offered at school, parents, teachers, and school-related factors have all been shown to have either direct
and/or indirect influence (Halim, Abd Rahman, Zamri, & Mohtar, 2018; Kauffmann et al., 2009). Franz-Odendaal et
al. (2016) found that factors influencing perception of students’ interest in STEM careers include the students’ own
perception about STEM careers, classroom culture and self-efficacy in STEM related fields. In addition, Franz-Odendaal
et al. (2016), Husin, Said, and Halim (2017), Kauffmann, Hall, Batts, Bosse, and Moses (2009) and Shahali, Halim, Rasul,
Osman, and Zulkifeli (2017), in their research found that factors influencing perceptions in STEM careers also include
employing the best curricular practices in the classroom, outside the classroom and in after-school activities. Based
on the various factors identified, this current research categorized the factors into three main constructs in order to
test the model. These factors affect secondary school students’ interest in STEM careers either directly or indirectly.
Environmental Factors
Vrankrijker and Napel (2017) defined environmental factors as the physical, social and attitudinal condition
in which individuals live and direct their lives. The factors can act as a catalyst and barrier. In this research, three
sub-constructs of environmental factors were tested in the model, namely learning experiences, social influences
and media. Learning experiences and media are under the category of services, system and policies while social
influences are under the category of support and relationship. Learning experiences include experiences inside
and outside the classroom. For learning experiences in the classroom teaching strategies that are students oriented
such as problem based learning are common in the STEM education literature g (Wyss, Heulskamp, & Siebert, 2012).
Meanwhile, informal STEM education can be implemented in places such as the museum and science center (Chi,
Dorph, & Reisman, 2014; Franz-Odendaal et al., 2016; Husin, Said, & Halim 2017). According to Chukwuemeka
(2013), STEM’s education-related media include reading materials such as books, magazines or encyclopedia, and
supplementary audio-visual materials such as radio and television. Media can also include any device with learning
substance or capacity that is utilized for learning such as STEM kits for construction or practical work. For social
influences, parents and friends are the closest inspiration to students and they may influence students’ perceptions
of STEM career and self-efficacy (Halim et al. 2018; Interactive, 2011; Kaiser, 2016).
405
(P. 404-416)
Self-efficacy in STEM
Burwell-Woo, Lapuz, Huang, and Rentsch (2015) stated that there are two aspects to address in order to encour-
age secondary students to plan for STEM jobs. The first is by expanding their initial enthusiasm and knowledge in
STEM disciplines and the second is to improve their STEM self-efficacy. Self-efficacy is defined as individuals’ beliefs
about their capability to execute a specific task that in turn affect their lives (Burwell-Woo et al., 2015; Rittmayer &
Beier, 2008). According to Rittmayer and Beier (2008), students with high science self-efficacy will motivate them-
selves to stipulate challenging goals and strive towards accomplishing those goals. In other words, STEM self-efficacy
is able to envisage one’s academic performance beyond one’s earlier attainment because assertive individuals are
inspired to succeed. Therefore, students’ STEM self-efficacy is positively related to STEM task performance interest
and engagement (Rittmayer & Beier, 2008).
According to Tracey (2010), career counsellors acknowledge the importance of using both the interest and
self-efficacy assessments in identifying and determining one’s possible career choice. Career counsellors acknowl-
edge the importance of using both the interest and self-efficacy assessments in identifying and determining one’s
possible career choice. Lent, Brown, and Hackett (1994; 2000) concurred that both self-efficacy assessments and
interest are crucial in career decisions but further argued that self-efficacy drives to interest development and hence
induces career preferences. In this present research, self-efficacy for STEM refers to the students’ beliefs about their
ability to perform in science, technology, engineering and mathematics respectively.
Perceptions of STEM Careers
STEM career requires individuals with skills in science, technology, engineering and mathematics and these
people often have better job prospects and a wider choice of rewarding careers (Holman & Finegold, 2010). How-
ever, the requirements for STEM skills, qualifications, requirements and job prospects are not presented clearly to
students (Holman & Finegold, 2010). Thus, students may not be interested in STEM careers (Wyss, Heulskamp, &
Siebert, 2012). On the other hand, Morgan, Issac, and Sansone (2001) argued that when an undergraduate college
student has positive perception of a career, it will most likely positively predict their career choice. Their research
demonstrated a clear link between undergraduate college students’ perception and interest in a career; however,
this is relatively unexplored among students of a younger age range especially at the secondary school level.
Students’ basic knowledge of career requirements, for example the skills needed and the job prospect, can
significantly impact their perception of what a particular STEM profession entails (Franz-Odendaal, Blotnicky, French,
& Joy 2016; Wyss, Heulskamp, & Siebert, 2012). In their research of 568 middle school children, Franz-Odendaal et
al. (2016) found that students will not follow careers that they are not familiar with. Wyss, Heulskamp, and Sibert
(2012) further argued that when students have negative views of STEM professions, the students may be left
out in developing interest in the process during the formal schooling. The research also revealed that students
are unaware of the types of skills suitable for various types of STEM careers. (Franz-Odendaal et al., 2016). This is
consistent with research findings by Wyss, Heulskamp, and Siebert (2012) who found that middle school students’
interest of particular occupations is linked to their perceptions of the career.
Interest in STEM Career
STEM encompasses a wide variety of career paths, ranging from medical science, forensic research, data
analytics and cyber security to game design and aviation just to name a few. However, students are not so familiar
with these kinds of jobs unlike jobs like doctors, engineers, and accountants. The research conducted by Bodzin
and Gehringer (2001) tried to make science authentic to students. Physicists were invited to talk to upper primary
students about physics and the work of a physicist specifically and scientist in general. In their research, the students
were asked to draw a scientist prior to the visit and four weeks after the visit. Studies such as the one by Bodzin and
Gehringer (2001) showed that students have a stereotypical view of a scientist. The students’ view was identified
through the students’ drawings of a scientist. The researchers suggested that students should be given a chance
to interact with scientists during formal schooling as it would influence students’ interest in STEM careers as well
as overcome students’ misconceptions of scientists as identified through their drawings.
According to Bodzin and Gehringer (2001), students are not interested in STEM careers because most schools
do not disclose information about STEM professionals to their students. On the other hand, Wyss, Heulskamp, and
406
(P. 404-416)
Siebert (2012) highlighted that if STEM professionals are brought to school, it might only meet a few of the students’
personal interests. Furthermore, there are several factors that may affect interest in STEM careers among students
such as self-efficacy (Burwell-Woo, Lapuz, Huang, & Rentsch, 2015; Rittmayer & Beier, 2008), perception of STEM
careers (Wyss, Heulskamp, & Siebert, 2012) and environmental factors (Chi, Dorph, & Reisman, 2014; Chukwuemeka,
2013; Interactive, 2011; Kaiser, 2016; Dorph et al., 2017).
Conceptual Framework
This research draws upon the social cognitive career theory (SCCT) (see Figure 1) which was developed by
Lent, Brown and Hackett (1994) based on Bandura’s social cognitive theory. SCCT claims that career aspirations
result from personal factors, environmental factors and behavior (Bahar & Adiguzel, 2016; Maltese & Tai, 2011). SCCT
has been applied in numerous researches to investigate the factors influencing interest in STEM careers (Bahar &
Adiguzel, 2016; Kier, Blanchard, Osborne, & Albert, 2014; Nugent, Barker, Welch, Grandgenett, Wu, & Nelson, 2015;
Sahin, Gulacar, & Stuessy, 2015). According to SSCT, career interest, choices, and educational and occupational
success are influenced by thoughts, beliefs, and personal and environmental factors (Petersen, 2014).
The theory draws upon factors related to cognition of a person (e.g., self-efficacy, outcome expectations,
goals) and on how the person’s cognition interacts with his environment that in turn forms one’s career develop-
ment. Thus, in this research, following on the theory and previous studies, five constructs have been proposed
for the tested model. The independent variables or factors include environmental factors, self-efficacy in STEM,
and perceptions of STEM careers while the dependent variables are interest in physical sciences STEM careers and
interest in life sciences STEM careers. According to the theory, interest in STEM careers is influenced by students’
perceptions of STEM careers and self-efficacy in STEM. Perceptions of STEM careers involve two perspectives, namely
job perspective and skills needed while self-efficacy involves four fields, namely science, technology, engineering
and mathematics which were studied separately. The theory also suggests that the environmental factors in the
framework are linked indirectly to interest in STEM careers (physical sciences and life sciences).
Figure 1. Conceptual framework.
The environmental factors involve three sub-factors, namely learning experiences, social influences and media.
For learning experiences, two aspects were studied, namely activities in the classroom and activities outside the
classroom while for social influences, two aspects were examined, namely influence by parents and friends. Media
influences consist of the internet, television programs, books, comics, magazines, newspapers, movies, social media
and the radio. According to SCCT, the environmental factors are linked indirectly to interest in careers through
self-efficacy expectations and outcome expectation (Lent, Brown, & Hackett, 2000). In this research, the factor
on perceptions of STEM careers was used instead of outcome expectation. In the Malaysian context, perception
of STEM careers has been the factor commonly associated with the decline of interest in STEM careers. Thus, for
this research, the environmental factors are linked indirectly to interest in STEM careers through self-efficacy and
407
(P. 404-416)
perceptions of STEM careers. Self-efficacy is the students’ beliefs about their task capability in science, technol-
ogy, engineering and mathematics separately. Perceptions of STEM careers take into consideration job prospects
and required skills in STEM. These factors are inter-related and influence interest in STEM careers and should be
considered in the context of Malaysia. This research was based on the assumption that students’ aspirations are
formed based on both personal and environmental factors. Previous research has reported that interest is one of
the strongest predictors of the choice of subjects and courses (Regan & DeWitt, 2015; Maltese & Tai, 2010). Therefore,
this research focused on interest in physical sciences STEM careers and interest in life sciences STEM careers as a
final predictor for choosing STEM careers. Physical sciences STEM careers such as electrical engineer, industrial,
agricultural, business scientists, and life sciences STEM careers such as pollution control analyst, geneticist, zoolo-
gist, geologist and medical scientist. Hence, these factors, namely self-efficacy in STEM and perceptions of STEM
careers are more relevant and have direct links to interest in STEM careers.
Hypothesis Model
Based on the conceptual framework, a hypothesis model (see Figure 2) was constructed using AMOS software.
Figure 2 shows the five latent constructs examined in this research: (i) environmental factors, (ii) perceptions of STEM
careers, (iii) self-efficacy, (iv) interest in physical sciences STEM careers, and (v) interest in life sciences STEM careers.
Three of the constructs (perceptions of STEM careers, environmental factors, self-efficacy) include sub-constructs
and all five constructs (including interest in physical sciences and life sciences STEM careers) are represented by
items in the questionnaire. In this model, the items that represent each construct and sub-construct were drawn
with an oval. Each item was measured based on a 10-point Likert scale.
Figure 2. Hypothesis model of factors that affect interest in STEM careers.
There are seven direct correlations in the model and these associations are the hypotheses to be tested. Each
hypothesis can be seen on each path and is labeled ‘H’. Therefore, this research posited the following hypotheses:
H1: Environmental factors have a significant effect on STEM self-efficacy.

H2: Environmental factors have a significant effect on perceptions of STEM careers.
H3: Self-efficacy has a significant effect on perceptions of STEM careers.
H4: Self-efficacy has a significant effect on interest in life sciences STEM careers.
H5: Self-efficacy has a significant effect on interest in physical sciences STEM careers.
H6: Perceptions of STEM careers have a significant effect on interest in life sciences STEM careers.
H7: Perceptions of STEM careers have a significant effect on interest in physical sciences STEM careers.
408
(P. 404-416)
The aim of this research was to develop a model on interest in STEM careers among students in Malaysian
secondary schools based on survey data which focused on three factors, namely environmental factors, self-
efficacy and perceptions of STEM careers. Specifically, the research was conducted to determine a model which
fits the hypothesized model and the causal directional effect between the five variables as shown in Figure 2.
For this purpose, goodness-fit index and statistically significant p-value < .05 were used to confirm the model
fit and to test the significance of the correlation between the variables.
Design
Survey research design was employed in the research in which a questionnaire was used as the research
instrument. The survey was conducted throughout Malaysia using cluster sampling technique. The cluster for this
research comprised the states in Malaysia where the states were divided into clusters. For Peninsular Malaysia,
the clusters were referred to as the north, east, west and south. For East Malaysia, both states, namely Sabah
and Sarawak were the clusters of the research.
Respondents
The questionnaires administered to the secondary schools in Malaysia were distributed based on the afore-
mentioned clusters. One state was selected to represent each cluster for Peninsular Malaysia. Thus, in total six
states were involved in this research. Three hundred (n=300) students of the age of 14 were involved for every
state, resulting in a total number of 1800 students as respondents. Of these 1800 students, 1780 returned com-
pleted questionnaires. However, for inferential statistical analysis, only 1485 respondents were included in this
research; the other 295 respondents were dropped because of missing data and extreme outlier. The removal
of outlier is in line with the requirement of the SEM- AMOS method of analysis. This parametric method requires
certain assumption of the data- namely the data needs to represent the normal distribution. If the extreme
outliers were not removed, then the research will not be providing the actual reflection of the phenomena that
is being examined (Aguinis, Gottfredson, & Joo, 2013).
Instrument
The questionnaire contained a total of 63 items after undergoing the exploratory factor analysis (EFA) and
confirmatory factor analysis (CFA) procedures. Table 1 shows the summary of the number and examples of items
for each construct and sub-construct tested in this research. For the learning experiences factor, items for the
sub-construct on activities in the classroom were developed by the researchers by focusing on teaching and
learning strategies related to STEM. Similarly, items for the sub-construct on activities outside the classroom were
developed by focusing on STEM-related activities such as visiting the science center and joining the carnival,
camp and STEM-related club. For items related to the social influence factor, the items were adapted from the
instrument developed by Nugent et al. (2015) and Kier et al. (2014). The items focused on the closest social fac-
tors influencing students’ interest in STEM careers. Two social influences were measured in this research, namely
parents and peers. Items related to media influences were also developed by the researchers with focus on the
influence of books, radio, newspapers, television, movies, the internet, social media, scientific magazines, comics
and digital games on interest in STEM careers
Self-efficacy refers to one’s perceived ability in completing an assignment task. In this research, self- efficacy
in STEM was measured through items on one’s self efficacy in Science, Technology, Engineering and Mathemat-
ics. Instruments from Nugent et al. (2015) and Kier et al. (2014) were referred to and adapted for this research,
especially for the items related to self-efficacy on science, technology, engineering and mathematics. In addition,
items from Buday et al.’s (2012) instrument were adapted for items related to self-efficacy on mathematics. For
the perceptions of STEM careers, items were entirely developed by the researchers by taking into account STEM
job prospects and the required skills in STEM careers. For the instrument on interest in STEM careers, items were
based on STEM careers as proposed by Faber et al. (2013). These STEM careers were then categorized into two
409
(P. 404-416)
main careers in the STEM fields: a) Life sciences (e.g. environmental works, biology earth science, medical sci-
ence, and chemistry) and b) Physical sciences (e.g. energy, engineering, entrepreneur and business scientists).
Table 1. Number and examples of items.
Number of
Construct Sub-construct Examples of Item
Items
Interest in Physical Sci- •• Energy: Electrical engineer, heating, ventilation and air conditioning.
3
ences STEM Careers •• Engineering: Civil, industrial, agricultural.
Interest in Life Sciences •• Biology and Zoology: Biological scientist.

5
STEM Careers •• Earth Science: Geologist, weather forecaster, archaeologist.
Perceptions of STEM Job Prospects 7 •• Careers in STEM fields are prestigious.

Careers
Skills Needed 6 •• Workers in the STEM fields require higher-order thinking.
Self-Efficacy Science 5 •• I can solve problems related to science concepts well.
•• I can handle everyday technological products easily (e.g., blender, micro-

Technology 5
wave, toaster, rice cooker).
Engineering 5 •• I can build electronic circuits.
Mathematics 5 •• I can solve mathematical problems properly.
Environmental Factors Media 10 •• I like reading books about STEM.
•• I conduct experiments or science projects in the laboratory or in the

Activities in the Classroom 3
school environment.
Activities Outside the

5 •• I visited the science center.
Classroom
•• My parents encourage me to pursue a career in STEM.

Social Influences 4
•• Most of my friends want to engage in STEM careers.
Total 63
SEM-AMOS Analysis
SEM approach was used to analyze data in order to evaluate the hypothesized model. For maximum likeli-
hood estimate, a set of goodness-of-fit index comprising chi-square (χ²), root mean square error of approximation
(RMSEA), comparative fit index (CFI), Tucker–Lewis index (TLI) and Incremental Fit Index (IFI) was used to evaluate
model fit. The significant p-value was used to explain the causal directional correlation while the regression weight
or known as beta value (β) was used to explain the strength of the correlation. Additionally, r squared (r2) value was
used to explain the variation of the model. All analyses were performed using AMOS18. Before the SEM analysis
was carried out, the items had been performed by validity and reliability procedures. Two types of validity were
involved in this research, namely content validity and construct validity. For content validity, two experts related
to STEM education in Malaysia were consulted to validate each item. Both the exploratory factor analysis (EFA) and
confirmatory factor analysis (CFA) were applied to determine the construct validity.
410
(P. 404-416)
Research Results
The structural model of interest in STEM careers is displayed in Figure 3. The normed chi square χ² was 3.847,
implying compatibility the hypothesis model fits with the survey data, χ² was greater than 1 and less than 5, and
the Root Mean Square Error of Approximation (RMSEA) was .044 which is less than .08. Thus, the model was con-
sidered good. The Comparative Fit Index (CFI) was .912, the Tucker-Lewis Index (TLI) was.908, and the Incremental
Fit Index (IFI) was.912. Each of the values obtained was larger than the suggested value of.90, showing a well-fit
model (Byrne, 2010; Hair et al., 2010). Thus, the above findings showed that the structural model fits the data well.
Figure 3. Structural model of interest in STEM careers.
For hypothesis testing, the overall findings showed that six out of the seven hypotheses were supported. Table
2 shows the results of the hypothesis testing for hypothesis 1 to 7 (direct effect path coefficient). Based on the
beta (β) standard value, the factor of self-efficacy was the greatest contributor to the interest in life sciences STEM
careers (β =.401; p <.001) and physical science STEM careers (β =.414; p <.001). This is followed by perceptions of
STEM careers (β =.217; p <.001) towards interest in life sciences STEM careers. Research findings also showed that
self-efficacy also affected perceptions of STEM careers in terms of job prospects and skills needed (β =.217; p <.001).
Environmental factors contributed 70% of the variance in self-efficacy (β =.838; p <.001) while 66% of the variance
in perceptions of STEM careers was predicted by environmental factors and self-efficacy. The environmental fac-
tors affected self-efficacy the most (β =.838; p < .001) compared to perceptions of STEM careers (β =.621; p <.001).
One important finding is that the direct correlation between perception on STEM career and interest in physical
sciences STEM careers was found to be not significant (β =.089; p <.001).
411
(P. 404-416)
Table 2. Significant effect path coefficient (direct effect).
Beta Standard Results of the

Path
value Error
t p
Hypotheses
Environmental
Self-efficacy <-- .838 .031 20.350 *** Supported (H1)
factors
Perceptions of STEM Environmental

<-- .621 .038 10.396 *** Supported (H2)
Careers factors
Perceptions of STEM
<-- Self-efficacy .217 .047 3.941 *** Supported (H3)
Careers
Interest in Life Sci-

ences
Interest in Physical
Sciences
Interest in Life Sci- Perceptions of STEM

<-- .217 .064 4.472 *** Supported (H5)
ences careers
Interest in Physical Perceptions of STEM

<-- .089 .093 1.713 .087 Not supported
Sciences careers
*** Significant at level p = .001
Discussion
This research determined the factors contributing to interest in life sciences and physical sciences STEM careers.
In general, the results contribute to our understanding of the correlation between environmental factors, internal
factors (self-efficacy and perceptions of STEM careers) and interest in life sciences and physical sciences STEM
careers. The findings indicated that environmental factors which consist of learning experiences, social influences
and media play an important role in triggering self-efficacy and good perceptions of STEM careers, which in turn
affect interest in STEM careers. These factors must be considered in planning STEM activities in order to enhance
students’ interest in STEM learning and occupations. However, an insignificant correlation was found between
perceptions of STEM careers and interest in physical sciences STEM careers.
Based on the model developed in this research, only 34% of the identified factors affected interest in life
sciences STEM careers and only 23% of the identified factors affected interest in physical sciences STEM careers.
Despite the low percentage of contribution to any of the interests in STEM careers, one could argue that since this
research is one of the early researches on modeling of factors contributing to interest in STEM careers, the cur-
rent finding is said to be acceptable. Another possible reason for the low percentage of contribution to interest is
that for this research, the factor of outcome expectancy that was proposed by Lent et al. (2000) was replaced by
perceptions of STEM careers (job prospects and skills needed in STEM careers). The adaptation of the model was a
result of the uniqueness of the factor, i.e. perceptions of STEM careers in the Malaysian context. Common reasons
cited for students shying away from STEM career is the low job prospect of STEM careers. This result empirically
proved that perceptions of STEM careers are not the main factor contributing to lack of interest in STEM careers.
In summary, self-efficacy affects interest in life sciences and physical sciences STEM careers while perceptions
of STEM career only affect interest in life sciences STEM careers. The findings from this research indicate that inter-
est in STEM careers is derived from task ability and good perception of STEM careers rather than good achieve-
ment in the subject matter. This idea supports the findings from a research by Maltese and Tai (2011) in which the
researchers argued that not all students choose STEM-related courses even though they have good achievement
in STEM subjects. The best students in school also choose other social science courses such as law, thus shying
away from the STEM pipeline. Previous research has reported that female students especially do not choose STEM
courses if they think that they are not proficient in STEM (Bamberger, 2014). The results of this research indicate
412
(P. 404-416)
that self-efficacy is crucial in cultivating interest in STEM careers and in strengthening the reason for students to
choose STEM trajectories.
Additionally, students’ beliefs about their capability in STEM provide good perceptions of norms as STEM
workers and in meeting the requirements of the STEM workforce. Therefore, students must be exposed to real life
STEM workers in order to give them understanding about STEM job prospects and the required skills in STEM fields.
Thus, schools must have good collaboration with universities and industries. Accordingly, the STEM mentor-mentee
program should be implemented in order to engage more students in STEM trajectories.
One of the reasons that led to the significant correlation between perceptions of STEM careers and interest
in life sciences STEM careers is due to job prospects that meet the students’ needs. For example, the career as a
doctor is the career most demanded by the students because they want to help other people and because such
a career gives them more satisfaction in terms of the possibility of contributing to the society. Additionally, the
job as a doctor is perceived to be a prestigious occupation. For the aspect of skills needed, students think that the
life sciences STEM careers require higher order thinking, creative problem solving and required work as a team.
While, the insignificant correlation found in this research between perceptions of STEM careers and interest in
physical sciences STEM careers questionnaire seems to suggest that students may be less aware of the kinds of
skills required in physical sciences STEM careers. This is in line with Depieri and Lopes’ (2014) research where they
explained that people generally have a reasonable idea of what a dentist or a doctor does. However, the scope of
the activities covered by the STEM careers in physical sciences such as engineering is more difficult to summarize.
A research by Shahali, Halim, Rasul, Osman, & Arsad, 2019) on students’ interest in STEM careers found that students
are unaware of the various types of responsibilities of an engineer. In other words, the diversity of STEM careers in
physical sciences such as engineering requires skills that the student does not know and is not aware of.
Thus, Depieri, and Lopes (2014) and Shahali et al. (2017) emphasized that in order to attract young
people to physical sciences STEM careers such as engineering, there is a need to build better understanding of
young people’s perceptions of STEM as well as to develop appropriate programs, activities or interventions that
will positively influence their perceptions of STEM. Such an intervention should take into consideration students’
perceptions of STEM careers especially on awareness related to physical sciences STEM careers.
Teachers and parents play a vital role in giving learning experiences and supporting students’ interest in STEM
learning and careers (Hall et al., 2011; Halim et al., 2018). Teachers must focus on task ability and emphasize skills
in STEM such as creative thinking, higher order thinking, making decisions wisely, working as a team, and being
able to create products based on STEM concepts. Teachers should organize activities in groups so that students
can collaborate with their peers and share their interests together. Furthermore, teachers should take into con-
sideration the different needs of students based on student diversity, particularly in relation to gender disparities
and the level of cognitive ability, in the STEM teaching and learning process.
Parents can help their children in STEM career selection by providing facilities such as putting them in additional
classes, allocating financial provisions to engage their children in outside of school science activities, promoting
the culture of science at home and encouraging their children to explore science-related careers. Accordingly,
information related to opportunities and demands in STEM should be disseminated to the parents so that they
could get accurate and the latest information as this would enable parents to provide their children with the neces-
sary information and the support their children need to develop interest in STEM careers (Halim et al., 2018). The
media should collaborate with the STEM stakeholders in providing quality materials to cultivate interest in STEM
learning and careers. The media should consider the diversity of students in terms of disparities related to gender
and ability. In sum, the final model of this research is presented in Figure 4.

413
(P. 404-416)
Figure 4. The final model of interest in STEM careers.
Further research could involve exploring students’ perceptions of identity in STEM careers. Interviews with
students and workers in the STEM fields would help to divulge how and why they make the decision to choose
STEM fields and careers. Venville et al. (2013) in their qualitative research reported that the passion for science and
curiosity about the world are important factors that contribute to students’ decision making about their trajecto-
ries. Thus, researching into the other factors that contribute to students’ decisions may be able to provide better
understanding of how students interact with their experiences to make decisions about their careers.
Conclusions
The model of interest in STEM careers developed showed that the environmental factors (learning experi-
ences, social and media) influence STEM self-efficacy and perceptions of STEM careers which, in turn, influence
interest in STEM careers except for the correlation between perceptions of STEM careers. Results suggest that in
developing students’ pathway to STEM careers, these factors should be taken into consideration in the intervention
effort of nurturing interest in STEM careers among students. In addition, the roles of social factors, namely teachers,
parents, and peers should be given emphasis as these individuals would be able to provide the necessary support
to the students in inculcating their interest in STEM careers. Additionally, the media should give support in terms
of providing the best quality materials for students to cultivate science learning as well as interest in STEM careers.
In order to be more efficient, the media should collaborate with STEM stakeholders to meet the students’ needs.
Overall, the factors are inter-related with each other and play a significant role in developing students’ interest in
STEM careers.

Acknowledgements
The authors sincere thanks to the Center for Research and Instrumentation (CRIM), National University of
Malaysia for supporting this research. This research is funded by Arus Perdana Grant (AP-2015-001), CRIM, National
University of Malaysia.
References
Aguinis, H., Gottfredson, R. K., & Joo, H. (2013). Best-practice recommendations for defining, identifying, and handling outliers.
Organizational Research Methods, 16(2), 270-301.
Ajzen, I. (2005). Attitudes, personality and behavior. (2nd ed.) Poland: Open University Press.
Bahar, A., & Adiguzel, T. (2016). Analysis of factors influencing interest in STEM Career: Comparison between American and Turkish
high school students with high ability. Journal of STEM Education: Innovations and Research, 17(3), 64.
414
(P. 404-416)
Bamberger, Y. M. (2014). Encouraging girls into science and technology with role model: Does this work? Journal of Science
Education and Technology, 23(4), 1–13.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman and Company.
Buday, S. K., Stake, J. E., & Peterson, Z. D. (2012). Gender and the choice of a science career: The impact of social support and
possible selves. Sex Roles, 66(3-4), 197–209.
Burwell-Woo, C., Lapuz, R., Huang, T., & Rentsch, N. P. (2015, June 14-17). Enhancing knowledge, interest and self-efficacy in stem
through a summer. Proceeding of 122nd ASEE Annual Conference & Exposition (pp. 1-15). Seattle, WA.
Byrne, B. M. (2010). Structural equation modeling with AMOS. Basic concepts, application and programming. New York: Taylor and
Francis Group.
Cavas, B., Cakiroglu, J., Cavas, P., & Ertepinar, H. (2011). Turkish students’ career choices in engineering: experiences from Turkey.
Science Education International, 22(4), 274–281.
Ceglie, R., & Settlage, J. (2016). College students’ persistence in scientific disciplines: A cultural and social capital as contributing
factors. International Journal of Science and Mathematics Education, 14(1), 169-186.
Chi, B., Dorph, R., & Reisman, L. (2014). Evidence and impact: Museum-managed STEM programs in out-of school settings.
Retrieved from http://sites. nationalacademies.org/cs/groups/ dbassesite/documents/webpage/dbasse_089887.pdf.
Chukwuemeka, O. (2013). Environmental influence on academic performance of secondary school students in Port Harcourt
local government area of rivers state. Journal of Economics and Sustainable Development, 14(12), 34-38.
Depieri, A. A., & Lopes, R. D. D. (2014, April 3-5). Students’ skills perceptions for engineering. Proceeding of Global Engineering
Education Conference (EDUCON), 2014 IEEE (pp. 402-407). Istanbul, Turkey.
Eccles, J., Barber, B., Stone, M., & Hunt, J. (2003). Extracurricular activities and adolescent development. Journal of Social Issues,
59, 865–889.
Faber, M., Unfried, A., Corn, J., & Townsend, L. W. (2013, Jun 23-26). Student attitudes toward STEM : The development of upper
elementary school and middle/high school student surveys. Proceeding of 120th ASEE Annual Conference & Exposition (pp.
1-26). Atlanta, Georgia.
Franz-Odendaal, T. A., Blotnicky, K., French, F., & Joy, P. (2016). Experiences and perceptions of STEM subjects, careers, and engage-
ment in STEM activities among middle school students in the maritime provinces. Canadian Journal of Science, Mathematics
and Technology Education, 16(2), 153-168,
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2010). Multivariate Data Analysis (6th ed.). New Jersey: Pearson
Prentice Hall.
Hall, C., Dickerson, J., Batts, D., Kauffmann, P., & Bosse, M. (2011). Are we missing opportunities to encourage interest in stem
fields? Journal of Technology Education, 23(1), 32-46.
Halim, L., Abd Rahman, N., Zamri, R., & Mohtar, L. E. (2018). The roles of parents in cultivating children’s interest towards science
learning and careers. Kasetsart Journal of Social Sciences, 39(2), 190-196.
Halim, L., & Subahan, T.M.M. (2016). Science education research and practice in Malaysia, In Chiu, M.H. (Ed.) Science education research
and practice in Asia: Challenges and Opportunities. Singapore: Springer.
Hawley, C. E., Cardoso, E., & McMahon, B. T. (2013). Adolescence to adulthood in STEM education and career development:
The experience of students at the intersection of underrepresented minority status and disability. Journal of Vocational
Rehabilitation, 39(3), 193-204.
Holman, J., & Finegold, P. (2010). STEM Careers Review. Report to the Gatsby Charitable Foundation. Retrieved from https://
warwick.ac.uk/fac/soc/ier/ngrf/stem/movingon/research/ 500stem_careers_review _nov_2010_holman.pdf
Husin, W.N. F., Said, S.M., & Halim, L. (2017). Enculturing STEM outside of the classroom. Bangi: Universiti Kebangsaan Malaysia.
Interactive, H. (2011). STEM perceptions: Student & parent study. Parents and students weigh in on how to inspire the next
generation of doctors, scientists, software developers and engineers (pp. 1–16). Rochester, NY: Study commissioned by
Microsoft Corp.
Kauffmann, P., Hall, C., Batts, D., Bosse, M., & Moses, L. (2009). Factors influencing high school students career considerations in
STEM fields. Proceeding of 2009 Annual Conference and Exposition (pp. 1-11). Austin, Texas.
Kaiser, K. (2016). Designing sewn circuits and STEM self-efficacy in middle school Girls (Doctoral dissertation). Retrieved from
https://scholarworks.uark.edu/etd/1515/.Khishfe, R., & Boujaouda, S. (2016). Lebanese students’ conceptions of and atti-
tudes towards science and related careers based on their gender and religious affiliations. International Journal of Science
and Mathematics Education, 14(1), 145-167.
Kier, M., Blanchard, M., Osborne, J., & Albert, J. (2014). The development of the Interest in STEM careers Survey (STEM-CIS). Re-
search in Science Education, 44(3), 461–481.
Kline, R. B. (2011). Principles and practice of structural equation modeling. (3rd ed.). London: The Guilford Press.
Lent, R. W., Brown, S., & Hackett, G. (1994). Toward a unifying social cognitive theory of career and academic interest, choice, and
performance. Journal of Vocational Behavior, 45(1), 79–122.
Lent, R. W., Brown, S., & Hackett, G. (2000). Contextual supports and barriers to career choice: A social cognitive analysis. Journal
of Counselling Psychology, 47(1), 36–49.
Maltese, A. V., & Tai, R. H. (2011). Pipeline persistence: examining the association of educational experiences with earned degrees
in STEM among U.S. students. Science Education, 95(5), 877–907
Morgan, C., Isaac, J., & Sansone, C. (2001). The role of interest in understanding the career choices of female and male college
students. Sex Roles, 44(5-6), 295-320.
415
(P. 404-416)
MOSTI. (2015). Science outlook action towards vision. Retrieved from: http://www.akademisains.gov.my/download/Science%20
Outlook%202015%20Invitation.pdf.
Nugent, G., Barker, B., Welch, G., Grandgenett, N., Wu, C., & Nelson, C. (2015). A model of factors contributing to STEM learning
and career orientation. International Journal of Science Education, 37(7), 1067-1088.
Regan, E., & DeWitt, J. (2015). Attitudes, interest and factors influencing STEM enrolment Behavior: An overview of relevant
literature. In Henriksen, E. K, Dillon, J., & Ryder, J. (Ed.), Understanding Students Participation and Choice in Science and Tech-
nology Education. (pp. 63-88). UK: Springer.
Rittmayer, A. D., & Beier, M. E. (2008). Self-Efficacy in STEM. SWE-AWE CASEE Overviews. Retrieved from http://www.AWEonline.org.
Sahin, A., Gulacar, O., & Stuessy, C. (2015). High school students’ perception of the effects of international science Olympiad on
their STEM career aspirations and twenty-first century skill development. Research Science Education, 45, 785-805.
Shahali, E. H. M, Halim, L., Rasul, M. S., Osman, K., & Arsad, N. M. (2019). Students’ interest towards STEM: A longitudinal study.
Research in Science & Technological Education, 37(1),1-19.
Shahali, E. H. M., Halim, L., Rasul, M. S., Osman, K., & Zulkifeli, M. A. (2017). STEM Learning through Engineering Design: Impact
on Middle Secondary Students’ Interest towards STEM. Eurasia Journal of Mathematics, Science and Technology Education,
13(5), 1189-1211. https://doi.org/10.12973/eurasia.2017.00667a.
Shahali, E. H. M., Ismail, I. & Halim, L. (2017). STEM Education in Malaysia: Policy, Trajectories and Initiatives. S&T Trends - Policy
Trajectories and Initiatives in STEM Education, 8(2), 122-133.
Tai, R., Liu, C., Maltese, A., & Fan, X. (2006). Planning early for careers in science. Science, 312(5777), 1143-1144.
Tracey, T. J. (2010). Relation of interest and self-efficacy occupational congruence and career choice certainty. Journal of Voca-
tional Behavior, 76(3), 441-447.
Venville, G., Rennie, L., Hanbury, C., & Longnecker, N. (2013). Scientists reflect on why they chose to study Science. Research in
Science Education, 43(6), 2207–2233.
Vrankrijker, M. K., & Napel, H. T. (2017). Environmental factors: classifications and measurement. Retrieved from https://unstats.
un.org/unsd/demographic-social/meetings/2017/new-york--disabilityegm/Session %206/ UNSD.pdf.
Wyss, V. L., Heulskamp, D., & Siebert, C. J. (2012). Increasing middle school student interest in STEM careers with videos of scien-
tists. International Journal of Environmental and Science Education, 7(4), 501-522.
Received: November 21, 2018 Accepted: May 23, 2019
Lilia Ellany Mohtar PhD Student (Physics Education), Faculty of Education, National
University of Malaysia, 43650, UKM, Bangi, Malaysia.
E-mail: lilelny@gmail.com
Lilia Halim PhD, Professor (Science Education), Faculty of Education & IDEA-
(Corresponding author) UKM, National University of Malaysia, 43650, UKM, Bangi, Malaysia.
E-mail: lilia@ukm.edu.my
Norshariani Abd Rahman PhD, Research Fellow, Institute of Islam Hadhari, National University
of Malaysia, 43650, UKM, Bangi, Malaysia.
E-mail: norshariani@ukm.edu.my
Siti Mistima Maat PhD, Senior Lecturer (Mathematics Education), Faculty of Education,
National University of Malaysia, 43650, UKM, Bangi, Malaysia.
E-mail: sitimistima@ukm.edu.my
Zanaton H. Iksan PhD, Senior Lecturer (Science Education), Faculty of Education,
National University of Malaysia, 43650, UKM, Bangi, Malaysia.
E-mail: zanaton.iksan@ukm.edu.my
Kamisah Osman PhD, Professor (Science Education), Faculty of Education, National
University of Malaysia, 43650, UKM, Bangi, Malaysia.
E-mail: kamisah@ukm.edu.my
416
THE RELATION OF GENDER
AND TRACK ON HIGH SCHOOL
STUDENTS’ ATTITUDE TOWARD ISSN 1648-3898 /Print/
CONVERGENCE ISSN 2538-7138 /Online/
Yustika Sya’bandari, Abstract. Preparing a positive attitude

toward convergence is essential to help
Minsu Ha, students grow into future generations
Jun-Ki Lee, with the ability to solve various complex
Sein Shin problems. This research aims to examine
high school students’ attitude toward con-
vergence and the relation to gender and
track. Responses from 1,186 Indonesian
students in tenth (10th) and eleventh (11th)
grade were purposively collected. Students
Introduction were administered twenty-three items of
attitude toward convergence covering five
Nowadays, problems in our daily life are getting complex, and it can constructs: knowledge, personal relevance,
only be understood by pulling together insights and techniques from dif- social relevance, interest, and self-efficacy.
ferent disciplines. Knowledge from several disciplines is required to solve Data were analyzed using IRT-Rasch analy-
the scientific and societal problems, and it has also been recognized as the ses, two-way ANOVA, Pearson correlation
critical issue of innovation challenge (Hacklin & Wallin, 2013). Knowledge analysis, and cluster analyses. The primary
integration is getting a keyword for the 21st century (Godemann, 2008), finding of this research presented that high
because it can unify the learning approach that fosters the connections school students’ attitude toward conver-
among disciplines in order to build the new knowledge in the form of gence on the construct of interest and self-
“integrative” concepts (Klein, 2005; Weinberger, 2007). An innovative study efficacy is significantly related to gender
integrates several disciplines through convergence (Hur et al., 2010). and track. Female students show to be
National Science Foundation (NSF) of the USA defined that conver- more interested in convergence than male
gence is knowledge, approach, and expertise, which is integrated deeply students. However, female students have
from various disciplines to construct broader and latest frameworks to lower self-efficacy. Additionally, science
solve the stratified problem. National Research Council (NRC; 2014) of the track students tend to be more interested
United States classified the terminologies of convergence: multidisciplinar- in convergence than humanities track and
ity, interdisciplinarity, transdisciplinarity. Multidisciplinarity brings together showing higher self-efficacy. After perform-
various disciplines, although all these disciplines keep on separated. It ing cluster analysis, students are divided
focuses on connecting topics and questions to foster broader information, into three particular groups according to
knowledge, and technique. Interdisciplinarity merges diverse expertise their attitude toward convergence. Finally,
and techniques from different disciplines to form new knowledge integra- customized learning is proposed to improve
tion. Transdisciplinarity exceeds disciplinary approaches through more students’ attitude toward convergence.
comprehensive frameworks. It is more than interdisciplinary combinations Keywords: attitude toward convergence,
which move beyond discipline-specific approaches to foster the modern female, humanities track, male, science
worldview. Many kinds of research have been developed from the integra- track.
tion of different disciplines through transdisciplinarity, interdisciplinarity,
and multidisciplinarity. Yustika Sya’bandari, Minsu Ha
Kangwon National University, Republic of Korea
The improvement of knowledge especially in science and engineering Jun-Ki Lee
makes convergence become very essential. Many researchers conduct the Chonbuk National University, Republic of Korea
convergence research to test many cases through collaborative working. Sein Shin
Chungbuk National University,
Collaborative working might generate the broader and newest idea due Republic of Korea
417
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
to integrating different insight of the group member especially if they come from the various field (Weinberger,
2007; Yuen et al., 2014). As recognized by Wuchty et al. (2007), there has been a big revolution in the research
field through joint research on various topics mainly in the science field. The number of team working has
increased immovably each year or even two times more from 1.9 to 3.5 authors per paper over 45 years. The
team research yet was cited more than research produced by an individual (Rousseau, 2001; Wuchty et al., 2007;
Valderas, 2007). The convergence research has improved significantly and provided many opportunities in the
future. Therefore, it is needed to encourage many people to recognize the important of convergence. One of
the ways is through convergence education.
Recently integrative approach and convergence in educational context has been emphasised in many coun-
tries. The most representative example would be ‘STEM Education’. The term ‘STEM Education’ has been broadly
implemented (Marginson et al., 2013; NRC, 2011). This field combines several disciplines, including science,
technology, engineering, and mathematics. It emphasizes science and math explicitly to increase technology
in school programs and recognition of engineering in K-12 education (Bybee, 2010; NRC, 2011). Moreover, the
importance of STEM education has been emphasized as STEM field seems very promising for future careers.
According to U.S. Bureau of Labor Statistics (2015) report in May 2015, 6.2% U.S jobs were represented from
STEM-related occupation. STEM employment increased by 10.5% between May 2009 and May 2015 compared
to 5.2% net growth in non-STEM. Based on the pattern, convergence related occupation will have given more
opportunity for a future career.
In order for STEM education to respond effectively to trends of convergence, it is important not only to
teach students already integrated knowledge, but also to help students to understand meaning and purpose
of convergent approach in this era. In addition, it is required to improve a positive attitude towards convergent
approach, in order for students can make a convergent attempt beyond the distinct academic boundaries in
STEM on their own in the future. Therefore, this study focuses on students’ attitudes toward convergence.
Particularly, this study was conducted with Indonesian students. STEM education has been implemented
in Indonesia. Several researches about STEM in Indonesia also have been conducted where there was a sig-
nificant result of students in attitude towards STEM (Suprapto, 2016; Winarno et al., 2017). The Programme for
International Student Assessment (PISA) in 2015 showed that Indonesia is one of the countries which has the
lowest performance of the results in science, reading and mathematics than OECD average (OECD, 2016). Trends
in International Mathematics and Science Study (TIMSS) 2015 International Results in Science also presented
that Indonesian students placed in 44th out of 47 participating countries in mathematics and science as well
as numeracy assessment for 4th-grade students (Martin et al., 2016). It is interesting when Indonesian career
motivation in STEM compared to Korean Indonesian high school students show higher career motivation in
STEM than Korean (Shin et al., 2018). Considering Indonesian students have lower knowledge, but they have
a significant interest in STEM. Therefore, this research is intended to examine Indonesian students’ attitudes
toward convergence.
Based on data of Pew Research Center Religion and Public Life, Indonesia is the country with the highest
Muslim population in 2010 with future growth projection for 2010-2050. Approximately 13.1% of the total
Muslim population in the world is from Indonesia. Gender is one of the most important influencing factors
since Muslim societies typically show a differentiation of gender inequality (Klingorová & Havlíček, 2015). The
educational outcome of gender remains a crucial issue to be discussed. Gender is an essential variable related
to convergence because as recognized by Patall et al. (2018) female is underestimated in science, technology,
engineering, and mathematics (STEM) track and careers. The previous research also showed that female seems
to be underrepresented in STEM educational program (Burke & Mattis, 2007). In Korea, gender significantly in-
fluenced the students’ attitude toward convergence, where a female has a lower attitude toward convergence
than male (Lee et al., 2017). The inequality of gender has been debated since the first biological innate gender
difference in cognitive capacity remains unclear (Good et al., 2008). However, the issue related to the women
demands for STEM professionals in the global marketplace is urgent, especially in facing the 21st-century chal-
lenges (National Academy of Sciences, 2007) has become a critical issue. Hence, gender becomes one of the
crucial variables that are involved in this research.
Senior High School students have been purposively chosen as the participants in this research because high
school students need to prepare their future career, and they have to be ready in facing the real world challenge.
A teacher needs to examine students’ attitude towards convergence and evaluate them to improve classroom
418
ISSN 1648–3898 /Print/ THE RELATION OF GENDER AND TRACK ON HIGH SCHOOL STUDENTS’ ATTITUDE TOWARD
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
instruction. The Indonesian high school student is divided into a different track. Track system in secondary
education divides students to a different academic course according to the students’ achievement and interest
to prepare them for higher education enrollment. Directorate of Development of Indonesia Senior High School
(2017) stated that there is academic specialization program aimed to develop students affective and cognitive
skills based on their interest skills competencies by following their interests, talents, and academic abilities in a
group of related subjects (track). Indonesia Senior High School divides the track into science, humanities, and
linguistics track. Hence, the track becomes the second fundamental variable that potentially influences students’
attitude toward convergence.
Preparing students’ positive attitude toward convergence becomes essential to help students grow into
future generations with the ability to solve creatively various complex problems. The authors evaluate effective
ways to improve positive students’ attitudes towards convergence by classifying students into different groups
and investigating each group’s characteristics. Clustering method is essential for grouping students and desi-
gning a learning model that can cover various constructs of students such as emotion and motivation to provide
adaptive learning support (Cha et al., 2006). Different learning style based on students preference in each group
should be accommodated to optimize the learning process (Cha et al., 2006). Mismatches between students
learning style and teacher common teaching will affect learning effectivity in the classroom where students can
become bored and inattentive and perform poorly on the test (Felder & Silverman, 1988). Felder & Silverman
(1988) have researched to classify students based on learning style to develop teaching strategies and evaluation
customized for students to finally reconfigure adaptively for accommodating students’ learning styles. Therefore,
this research is intended to uncover how gender and track are related to attitude toward convergence and by
classifying students, find effective ways to improve students’ attitude toward convergence.
Research Focus
Based on the background, this research focused on the following research objectives:
1. To analyze the validity of students’ attitude toward convergence instrument.
2. To explore the relation of gender and track to students’ attitude toward convergence.
3. To identify correlation among the five constructs of students’ attitude towards convergence based on
gender and track.
4. To examine how the group of student is classified based on their attitude toward convergence.
General Background
The quantitative survey was carried out in this research in August 2018 in Indonesia. The data of attitude
toward convergence were collected from Indonesian High School students,’ and it was validated by performing
IRT-Rasch analysis. The further analysis of relation and correlation of gender and track to students’ attitude toward
convergence was also examined. The clustering method was also performed in this research. The detailed explana-
tion of the research methodology was described as follows.
Participants
Data were collected from 1,186 Indonesian high school students in tenth (10th) and eleventh (11th) grade.
Students were purposively selected from two different tracks: humanities and science track. Among the total
participants, 570 (48.06%) were representative of the humanities track, and 616 (51.94%) were representative of
the science track. In terms of gender, students consisted of 471 (39.72%) male and 712 (60.03%) female students.
Three (0.25%) students did not report their gender. The participants were recruited from three private schools and
three public schools.
419
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
The instrument administered to students was the attitude toward convergence instrument for a high school
student developed by Shin et al. (2014b). Appendix 1 showed the instrument of students’ attitude toward con-
vergence, which has been translated into the English version. In this research, the convergence covered the five
constructs of attitude: knowledge, personal relevance, social relevance, interest, and self-efficacy. Twenty-three
items were graded in the 6-point Likert scale from strongly disagree (1) to strongly agree (6). The instrument is
beneficial for measuring students’ attitude toward convergence because the teacher can estimate the effective
teaching instruction of convergence in science learning. It can also measure whether the convergence teaching-
learning programs can significantly affect students’ attitude toward convergence or whether teaching science
through a convergence approach becomes mediating effect on students’ learning motivation for science (Shin
et al., 2014a).
Data Analysis

Item response theory (IRT) has been admitted as a more advanced psychometrics measurement than classical
test theory (CTT) (Embretson & Reise, 2013). In this research, IRT-Rasch analysis of item quality, item and person
reliability, and differential item functioning was performed by Winstep V.3.92.1 software. Besides IRT analysis, CTT
analysis was also performed to uncover the reliability (Alpha Cronbach) using IBM Statistics 24 software. Result
of CTT analysis was presented as consideration for the readers who are relatively new with IRT-Rasch analysis.
The item quality was investigated by examining the infit and outfit mean-square (MNSQ) value. The value
of MNSQ value for infit and outfit in productive measurement is from 0.7 to 1.4 (Wright & Linacre, 1994). If data
is beyond that range, it is called a misfitting item. The consistency of the item was examined by investigating
the item and person reliability and also Cronbach Alpha. The standard indicator of item and person reliability
interpretation follows the standard of Fisher (2007) that categorizes the item and person reliability into ‘poor’ if
the value is less than 0.67, ‘fair’ if the value is between 0.67 and 0.80, ‘good’ if the value is between 0.81 and 0.90,
‘very good’ if the value is between 0.91 and 0.94, and ‘excellent’ if the value is higher than 0.94.
The IRT-Rasch analysis also can be performed to analyze the item bias by testing differential item function-
ing (DIF). DIF measures whether the instrument could give the same treatment to the compared groups (Boone
et al., 2014). Differential item functioning will ensure that the instrument has the same standard to measure the
students across the group of gender and track. Boone et al. (2014) stated the DIF value, which is higher than
0.64 or less than -0.64 indicates the occurrence of item bias.
Additionally, the value of person measure (a quantitative measure in the Rasch scale that expresses the
ability of the respondent on a uniconstructal scale) was analyzed using IBM SPSS Statistics 24. It was used to
explore the relationship between gender and track on students’ attitude toward convergence through multi-
variate two-way ANOVA. The relation is categorized as significant if the p-value is less than 0.05. To examine
the correlation of each construct, the Pearson correlation coefficient was calculated. The construct might be
significantly correlated with one another if the p-value is less than 0.01 or 0.05.
A cluster analysis was examined to divide students’ group based on attitude toward convergence. Multivariate
clustering analysis (mclust) using R-software was performed. Mclust is very well-known for clustering based on
finite Gaussian mixture modeling. The EM algorithm is applied in Gaussian model-based clustering, which makes
easier to be implemented as it has a numerically stable algorithm and reliable global convergence under fairly
general conditions. Mclust can provide several mixing components and covariance parameterization selected
by using the Bayesian Information Criterion (BIC). The highest BIC will have a better classification (Scrucca et al.,
2016). The best model and the number of clusters will be found, and every student is enrolled in their cluster
(student’s group). Further analysis was conducted to investigate the characteristics of each group by comparing
the person measure of the group’s member using the IBM Statistics 24 software. Additionally, gender and track
as independent variables were applied in the Pearson Chi-square test analysis.
420
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
Research Results
The Validity of Students’ Attitude toward Convergence Instrument
The quality of the item was examined using IRT by analyzing the item fit with the Rasch model. Table 1 shows
infit, and outfit MNSQ values range from 0.72 to 1.38 logit in knowledge construct, 0.80 to 1.16 logit in personal
relevance construct, 0.78 to 1.24 logit in social relevance construct, 0.91 to 1.08 logit in interest construct, and 0.76
to 1.11 logit in self-efficacy construct. Overall, the MNSQ value ranged from 0.7 to 1.4 logit for all items in each
construct, which indicated no misfitting item. No misfitting item meant that the attitude towards convergence
instrument fit the Rasch model (Wright & Linacre, 1994). The items did not need to be revised because every item
had a rational function to measure what should be measured (Sumintono & Widhiarso, 2014).
Table 1. Psychometrical properties based on Rasch Modeling of students’ attitude toward convergence
instrument.
Item-fits
Item Person Cronbach
Construct
Outfit Reliability Reliability Alpha
Infit MNSQ
MNSQ
Lowest 0.73 0.72

Knowledge .99 .75 .764
Highest 1.38 1.38
Lowest 0.83 0.80
Personal Relevance .95 .74 .769
Highest 1.16 1.16
Lowest 0.81 0.78
Social Relevance .98 .76 .780
Highest 1.24 1.22
Lowest 0.95 0.91
Interest .98 .77 .805
Highest 1.08 1.06
Lowest 0.76 0.76
Self-efficacy .92 .78 .806
Highest 1.11 1.11
Reliability result was presented in Table 1. The IRT-Rasch analysis shows the lowest item reliability in the con-
struct of self-efficacy and the highest item reliability in the construct of knowledge (item reliability values of .92
and .99, respectively). Since the value of item reliability was more than .90, it indicated as very good to excellent
reliability (Fisher, 2007). High item reliability indicated that the sample was large enough to place items on the
latent variable (Boone et al., 2014).
Regarding the person reliability, the lowest value is in the construct of personal relevance, and the highest
value is in the construct of self-efficacy (person reliability value .74 and .78, respectively). The person reliability
value was between .74 and .78, meaning that it was categorized as fair reliability (Fisher, 2007). Person reliability
indicates how well the student performs based on the Rasch-model (Tornabene et al., 2018). Cronbach Alpha (CTT)
analyses show that the construct of knowledge has the lowest reliability value, and the construct of self-efficacy
has the highest reliability value (reliability value .764 and .806, respectively). Since reliability was over than .70, it
meant that the reliability value was categorized as fair to good reliability (Fisher, 2007).
Besides checking the misfitting item, item and person reliability, and the Cronbach Alpha, the IRT-Rasch model
also examined item bias or differential item functioning (DIF). It shows whether each item in the instrument can
generalize different groups of the test administered or not. The DIF value was shown in Table 2. DIF gender contrast
value in knowledge construct ranges from 0.00 to 0.29. It ranges from 0.00 to 0.11 in personal relevance construct,
0.05 to 0.13 in social relevance construct, 0.05 to 0.21 in interest construct, and 0.02 to 0.28 in self-efficacy construct.
These DIF gender contrast values in different constructs were lower than 0.64 and higher than -0.64 (Boone et al.,
2014). Therefore, there was no gender bias regarding students’ attitude toward convergence items.
421
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
Table 2. Differential item functioning (DIF) gender and track of students’ attitude toward convergence.
Construct of ‘Attitude toward

Code DIF Gender Contrast DIF Track Contrast
Convergence’
K1 0.16 0.08
K2 0.00 0.00
Knowledge
K3 0.08 0.15
K4 0.29 0.27
R_P1 0.11 0.51
R_P2 0.00 0.23
Personal Relevance R_P3 0.08 0.05
R_P4 0.08 0.58
R_P5 0.11 0.17
R_S1 0.05 0.20
R_S2 0.05 0.10
Social Relevance
R_S3 0.13 0.08
R_S4 0.11 0.00
A_I1 0.17 0.18
A_I2 0.05 0.30
Interest A_I3 0.19 0.11
A_I4 0.21 0.45
A_I5 0.16 0.23
A_SE1 0.02 0.07
A_SE2 0.28 0.16
Self-Efficacy A_SE3 0.13 0.00
A_SE4 0.02 0.19
A_SE5 0.08 0.25
DIF track contrast value ranges from 0.00 to 0.27 in knowledge construct, 0.05 to 0.58 in personal relevance
construct, 0.00 to 0.20 in social relevance construct, 0.11 to 0.45 in interest construct, and 0.00 to 0.25 in self-efficacy
construct. These DIF contrast values were also lower than 0.64 and higher than -0.64. Therefore, there was no track
bias regarding students’ attitude toward convergence items.
The Relation of Gender and Track on Students’ Attitude towards Convergence
The data of relation between the two independent variables and the five constructs of attitude toward con-
vergence based on multivariate two-way ANOVA were presented in Table 3. The significant value of gender and
track in relations with students’ attitude toward science was discussed.
Table 3. Result of two-way ANOVA of test gender and track.
Gender Track Gender x Track

Construct
F p-value ηp2 F p-value ηp2 F p-value ηp2
Knowledge 0.209 .647 .000 0.615 .433 .001 0.477 .490 .000
Personal Relevance 1.938 .164 .002 2.964 .085 .003 0.167 .683 .000
Social Relevance 0.278 .598 .000 23.757 .0001 .020 1.494 .222 .001
Interest 31.891 .0001 .026 57.591 .0001 .047 0.413 .521 .000
Self-efficacy 5.026 .025 .004 4.982 .026 .004 0.002 .965 .000
422
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
Knowledge. Based on the result, significant gender effect was not found regarding students’ knowledge on at-
titude toward convergence (F [1, 1179] = 0.209, p = .647, ηp2 = .000). Significant effect of track (F [1, 1179]
= 0.615, p = .433, ηp2 = .001) was also not found either. Based on the result of combined analyses, there
was no significant relation between gender and track on attitude toward convergence (F [1, 1179] = 0.477,
p = .490, ηp2 = .000). A non-significant result for the interaction between gender and track indicated that
students’ attitude toward convergence in knowledge construct was not related to gender or track. The
mean of person measure (logit) difference was shown in Figure 1. Male students in humanities track have
higher mean value than those in the science track. In contrast, female students in humanities and science
tracks have almost the same mean value.
Personal relevance. Based on the result, significant effect of gender was not found regarding student’s personal
relevance on attitude toward convergence (F [1, 1179] = 1.938, p = .164, ηp2 = .002). Significant effect of
track was not found either (F [1, 1179] = 2.964, p = .085, ηp2 = .003). In combined analyses, the relation
between gender and track was also not significant (F [1, 1179] = 0.167, p = .683, ηp2 = .000). A non-
significant result for gender and track indicated that students’ attitude toward convergence in personal
relevance construct was not related to gender or track. Mean differences in person measure (logit) were
shown in Figure 1. Both male and female students in the science track have higher mean value than those
in humanities track. However, the difference is not significant.
Social relevance. There was no significant effect of gender regarding student’s social relevance on attitude toward
convergence (F [1, 1179] = 0.278, p = .598, ηp2 = .000). In contrast, significant effect of track (F [1, 1179]
= 23.757, p = .0001, ηp2 = .020) was found. Results of combined analyses showed the relation between
gender and track was not significant (F [1, 1179] = 1.494, p = .222, ηp2 = .001). A significant result for track
indicated that a difference in the track related students’ attitude toward convergence in social relevance
construct. Mean differences in person measure (logit) were shown in Figure 1. Both male and female
students in the science track have both gap different mean value with humanities track.
Interest. Based on the result, significant effect of gender was found regarding student’s interest in attitude toward
convergence (F [1, 1179] = 1.938, p = .0001, ηp2 = .026). Significant effect of track (F [1, 1179] = 57.591,
p = .0001, ηp2 = .047) was also found. In combined analyses, the relation between gender and track was
not significant (F [1, 1179] = 0.413, p = .521, ηp2 = .000). This indicated that students’ interest in attitude
toward convergence was affected by gender and track. Mean differences in person measure (logit) were
shown in Figure 1. Female students have higher interest than male students. Science track students also
have more interest in attitude toward convergence than humanities track students.
Self-efficacy. Based on the result, gender affected significantly regarding students’ self-efficacy on attitude toward
convergence (F [1, 1179] = 5.026, p = .025, ηp2 = .004). Significant effect was also found for track (F [1, 1179]
= 4.982, p = .026, ηp2 = .004). In combined analyses, the relation between gender and track was not sig-
nificant (F [1, 1179] = 0.002, p = .9653, ηp2 = .000). A significant result on the interaction between gender
and track indicates that students’ attitude toward convergence in the self-efficacy construct was related
to gender or track. Mean differences in person measure (logit) are shown in Figure 1. Male students have
higher mean values in self-efficacy rather than female. Both male and female students in the science tracks
also have higher mean values than those in humanities track in self-efficacy construct.
423
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
Figure 1. The interaction between gender and track on students’ attitude toward convergence in each con-
struct.
Regarding the relation between gender and track on students’ attitude toward convergence, the essential
information emphasized that female students had a higher (significant) interest than male in attitude toward
convergence both in humanities and science track. It was interesting to see the evidence that female students
had a higher interest in the attitude toward convergence while they had lower self-efficacy in both science and
humanities track. In the term of the track, it showed a significant difference regarding attitude toward convergence
in interest and self-efficacy construct. Science track students had higher (significant) attitude toward convergence
than humanities track students both in interest and self-efficacy construct.
Correlations among Five Constructs of Attitude toward Convergence based on Gender and Track
Association of each construct was determined based on correlation coefficients. Students’ attitudes toward
convergence constructs are significantly correlated with each other if the r value was less than 0.01 (r < 0.01). Result
of correlation could be seen in Table 4.
424
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
Table 4. The correlation coefficient (r) for every variable.
Gender x Track
Construct Correlation Male Female
Humanities Science Humanities Science
K R_P .544** .528** .516** .517**

K R_S .364** .419** .368** .441**
K A_I .318** .337** .247** .246**
K A_SE .390** .424** .380** .424**
R_P R_S .528** .670** .584** .683**
R_P A_I .426** .504** .462** .432**
R_P A_SE .417** .411** .237** .398**
R_S A_I .476** .574** .417** .450**
R_S A_SE .333** .214** .261** .414**
A_I A_SE .398** .214** .274** .377**
Male students in humanities track show the highest correlation coefficient between knowledge and personal
relevance construct (r = .544) meanwhile, the lowest correlation coefficient is between knowledge and interest
construct (r = .318). Males in science track shows the strongest correlation coefficient between personal relevance
and social relevance construct (r = .670) meanwhile the weakest correlation coefficient is between social relevance
and self-efficacy and also between interest and self-efficacy construct (r = .214).
Both humanities and science track shows that female students have the strongest correlations coefficient be-
tween personal relevance and social relevance (r = .584 and r = .683, respectively). The weakest correlation coefficient
for female students in the humanities track is between personal relevance and self-efficacy construct (r = .237). For
those in science class, the weakest correlation coefficient is between knowledge and interest construct (r = .234).
Overall, correlation coefficients among five constructs of attitude toward convergence of male participants in
both humanities and science tracks were higher than those of female students in humanities and science tracks.
Regarding correlation coefficients among constructs of attitude toward convergence between the two tracks, sci-
ence class had higher correlation coefficients than humanities track.
Group Classification based on Students’ Attitude toward Convergence
The software suggested classifying students based on the highest value of BIC. The possibility of 1 – 10 groups
was inputted. Results showed that the highest BIC was two-groups. When students were enrolled in two groups, the
group was only divided into high and low person measure of the five constructs of attitude toward convergence.
Three-groups were analyzed for further analyses and resulted in the declination lines (BIC difference) of two-groups
and three-groups, which was not a significant difference. Three-groups resulted in the unique characteristics of each
group, as presented in Figure 2. The number of students in each group was also substantial enough to represent
their group. The author decided to use three-groups to classify the group of students based on their attitude toward
convergence. VEE model was best fitted to three groups, which indicated an ellipsoidal, equal shape, and orienta-
tion with BIC value of -22657.02. Characterization and naming were determined for every group by examining the
highest mean value of each construct. The distribution of each group by gender and track can be seen in Table 5.
425
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
Figure 2. Results of cluster analysis using mclust.
Group 1 consists of 15.18% (180) participants. It has a high mean value in knowledge (M = 2.66, SD = 1.14)
and the lowest mean value in interest (M = 2.10, SD = 1.10) of attitude toward convergence. Based on these
characteristics, group one was further called ‘High-Knowledge Convergence (HKC) group. Group 2 consists of
24.45% (290) participants. It has the highest mean value in interest (M = 3.44, SD = 2.47) and the lowest mean
value in knowledge (M = 1.52, SD = 2.67) of attitude toward convergence construct. Based on these charac-
teristics, group two was further called High-Interest Convergence (HIC) group. Names of these groups were
considered based on the difference in knowledge and interest. Personal relevance, social relevance, or self-
efficacy was not considered because the two groups did not have significant mean value in the result. Group 3
consists of 60.37% (716) participants. It had the lowest mean value among all groups, especially in self-efficacy
construct (M = 0.05, SD = 1.18). Based on these characteristics, group three was further called Low Self-Efficacy
Convergence (LSeC) group. The group members distribution based on their demographic was shown in Table 5.
Table 5. The student distribution in each group based on gender and track.
Gender Track
Group
Male Female Total Percentage Humanity Science Total Percentage
HKC 77 103 180 15.21 86 97 180 15.18

HIC 106 184 290 24.51 123 167 290 24.45
LSeC 288 425 713 60.27 364 352 716 60.37
Total 471 712 1183* 99.99 570 616 1186 100.00
*3 students didn’t report their gender
The group characteristic based on the demographic variable was analyzed by performing the Chi-squared
test. Attitude toward convergence of the groups was not related to students’ gender (χ2 = 2.047, p = .359).
However, it was significantly associated with the track (χ2 = 6.208, p =.045).
426
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
Discussion
Results of this research demonstrate that Indonesia senior high school students’ attitude toward conver-
gence was related to both gender and track in interest and self-efficacy construct. This section would discuss
evidence of this research regarding the four research questions.
The Validity of Students’ Attitude toward Convergence Instrument
One of the essential sections when applying an instrument is validation testing. Messick (1995) has stated
that validation was a process of evaluating empirical and theoretical judgment justification to support the score
interpretation precisely. Messick (1995) on psychological assessment proposed six aspects of validity such as
content, substantive, structural, generalizability, external, and consequential. This research applied several
constructs, which will be described as follows.
The content validity presents that the domain construct (content) is relevance and representative to me-
asure students’ ability and related to the purpose of the instrument (Messick, 1995). Students’ attitude toward
convergence instrument adopted from a journal published by Korean had conducted content validation. The
content validity of the instrument had been verified by experts in science education and science teachers in
high school (Shin, 2014). A total of 23 valid items in five constructs of students’ attitude toward convergence
were produced.
The second is a substantive construct which describes indirect data as a result of participants in response
to the measurement. The substantive aspect of validity includes item and person infit and outfit statistical
evidence engaged by students who are representative and relevant to the study assumed (Ravand & Firoozi,
2016). This research performed IRT-Rasch analysis to investigate infit and outfit MNSQ. Following the standard
of Wright & Linacre (1994), it is presented that the items did not produce the misfitting in students’ attitude
towards convergence instrument. No misfitting item in the instrument indicated that students’ cognitive pro-
cesses are appropriate to be measured with items included in the construct (Ravand & Firoozi, 2016). Items fit
with students’ ability indicated that they understand these items of the instrument (Fisher, 2000).
The consistency was presented in the form of a person and item reliability (IRT-Rasch) as well as Cronba-
ch alpha (CTT). Results showed that the item reliability was in the criteria of very good to excellent. The item
reliability value was more than 0.90, indicating that the participant was big enough to confirm the apparent
difficulties of the latent variable (Tornabene et al., 2018). Meanwhile, person reliability was categorized as fair.
If the person reliability value was bad, Linacre (2017) suggests that items might not be able to distinguish the
ability of students precisely, but this research was fair enough to present the performance of each student. As
a comparison point, the value of Cronbach alpha was also in the fair to good category.
Generalization ensures that the instrument is not violent to a particular sample or group (e.g., gender), but
it can be generalizable across the groups to make the valid data interpretation (Messick, 1995; Tornabene et al.,
2018). Differential item functioning (DIF) using the IRT-Rasch analysis was conducted in this study to investigate
whether the items can generalize across the group of gender and track. Results of DIF indicated invariance of
item calibrations across the population of students (Ravand & Firoozi, 2016). Following the standard of Boone
et al. (2014), no DIF was detected in this research, meaning that the instrument of students’ attitude towards
convergence was generalizable both for males or females in humanities or science class.
The Relation of Gender and Track on Students’ Attitude towards Convergence
This research examined the interaction between the two most significant contributing variables regarding
students’ attitude towards convergence: gender and track. These variables are significantly related to the attitude
towards convergence, particularly in the dimension of interest and self-efficacy. This section is going to discuss
the findings in the two discussion parts. The first, female students have significant interest but lower self-efficacy
than male students in attitude toward convergence for both in humanities and science track. The second is the
higher interest and self-efficacy of science track students than humanities in the attitude toward convergence.
427
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
It has been recognized that interest refers to a behavioral factor that affected students’ curiosity in learning,
and they will involve more in their assignments. It also encourages students to study and engage in a more
in-depth level of thinking (Hayden et al., 2011). This research indicates that female students tend more to get
themselves involved in convergence than male students. Female students are likely to enjoy creating a new idea
and work together with other people. Group working will allow students to share ideas and expertise to solve
a particular problem (Taylor & Greve, 2006). Hence, group working can contribute to knowledge integration
(convergence). The research by Stump et al. (2011) also shows that female students engage more when they
learn together in collaborative learning. The previous research supports the result of this research by Woolley
et al. (2010) which stated that the group performance is not determined by the average intelligence level of a
group member, but it is determined by the proportion of female in the group. Furthermore, the result of the
group performance improves when more females are in the group (Bear & Woolley, 2011; Woolley & Malone,
2011). It is because a female shows a better score on social sensitivity (Woolley & Malone, 2011) and conversa-
tional expropriations. Social sensitivity is the personal capability in accepting and perceiving others’ thinking,
mood, and behavior (Bender et al., 2012; Woolley et al., 2010). Woolley and Malone (2011) also state that the
most important thing for a collaborated group is having people with high social sensitivity, whether they were
males or females. However, previous research showed that females had a higher score on social sensitivity test
than males. This result of the research should be considered by teachers when designing gender proportion of
collaborative groups in the classroom to have better students’ convergence knowledge and how to improve
group performance.
It is interesting to have the evidence that females have a higher interest; in contrary, they also have lower
self-efficacy in attitude toward convergence for both science and humanities track. Females want to involve
their mind in the convergence, but they do not have any confidence to be more active or persist longer in their
effort. This might be because of the views that a female is underestimated in science, technology, engineering,
and mathematics (STEM) both in college majors and employment (Patall et al., 2018). Jordan and Carden (2017)
also report that females in the STEM track had less confidence in their academic ability than males. It is easy
for females to have stress with low self-efficacy (Goel & Bardhan, 2016). The research of Cavallo et al. (2004)
also show that female students have less self-efficacy than males in science learning, especially in physics. The
research of Glynn et al. (2011) has also claimed that, among both science track and nonscience track, male
students show higher self-efficacy than female students.
Students with low self-efficacy will lead them to have high anxiety (Glynn et al., 2011). It is also reported
that self-efficacy has a positive correlation to GPA. It influences students’ achievement because it can encourage
students to have learning persistence in achieving their academic goals (Britner, 2008). The low female student
self-efficacy in convergence remains a critical challenge for educators. Students’ experiences in high school
were very crucial in shaping beliefs and determining whether students pursue a STEM college track and career
or not. Thus, it should be considered by teachers when designing teaching instruction to discuss this issue to
increase female students’ self-efficacy in the classroom, especially in convergence topic.
Most students in the science track have a higher attitude toward convergence preferably those in huma-
nities track both in interest and self-efficacy construct. This is in line with the previous research, which stated
that students’ attitude toward convergence of science track was relatively higher than those of the humanities
(Shin, Ha, & Lee, 2014). Science is one of the tracks that involves many collaborative activities, and it might
influence positively to the students’ attitude towards convergence (Prince, 2004; Bowen, 2000). Collaborative
learning can provide a learning experience for students by effectively implementing knowledge integration
(Jeong & Chi, 2007; Willey & Gardner, 2012) so as students can generate the new idea (Yuen et al., 2014). Stu-
dents’ knowledge and experience to the particular learning program also could maintain the positive attitude
and improving the interest and motivation than those who do not implement the particular learning program
(Gibson & Chase, 2012; Nugent, 2010; Weinberger, 2007). The example of collaborative learning by applying the
STEM approach is the robotic project. This kind of activity imitates how engineers and scientists from different
disciplines work together on a single project (e.g., De Vault, 1998; Kitts & Quinn, 2004). This kind of activity
can encourage students’ interest in convergence as their idea and capabilities complement each other in the
group membership (Yuen et al., 2014) and the new product generated from work can increase their confidence.
The researchers in science fields also tend to have collaborative research with others. Even in the same area,
collaborative work among scientists all over the world is needed (Lee et al., 2013). The example of the research
that has been conducted was the research in the human genome project (Vermeulen et al., 2013), research on
428
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
SARS, and new CERN particle physics. Those research involved intellectual resources through the cooperation of
various countries in the world. Collaboration in scientific research has improved rapidly. For almost five decades,
90% of paper is dominated by the collaborative team authors. Generally, social science papers are written in pair
followed by a positive trend from more extensive team authors. In contrary, for about 90% research in art and
humanities are generated from the single author (Wuchty et al., 2007). Besides, science students also believe
that attitude toward convergence is beneficial to be applied in their field, especially for working in a team. It
is well recognized that social environment has a high influence on the formation of an individual’s self-beliefs,
especially their abilities to obtain their development goals and the collective intelligence (Bossche et al., 2006).
Correlations Among Five Constructs of Attitude towards Convergence based on Gender and Track
Correlations among all constructs need to be analyzed to know whether a difference in the other con-
struct follows the change in one construct. Results of this research showed that correlation among constructs
for gender and track exerted significant effects on students’ attitude toward convergence. Male students in
humanities and science tracks showed stronger correlations among constructs than female students. This result
suggested that when male students showed high value in one construct of attitude toward convergence, other
components might also have high scores. When female students show high value in one construct of attitude
toward convergence, other parts might not have high scores as male.
Regarding correlations among constructs for two tracks, those with science class had stronger correlations
among constructs than those with humanities class for both males and females. This result suggested that,
when science students showed high value in one construct of attitude toward convergence, other components
might also have high scores. When humanities students show high value in one construct of attitude toward
convergence, other parts might not have high scores as science track. The overall positive correlation of each
construct and gender and track indicated that the instrument associated well with the research variables.
Group Classification based on Students’ Attitude toward Convergence
Based on findings, there were three types of the group according to students’ attitude toward convergence,
namely High-Knowledge Convergence (HKC), High-Interest Convergence (HIC), and Low Self-Efficacy Conver-
gence (LSeC). Besides, the group is not differentiated based on gender, but it is significantly associated with
the track. These findings show the impact of the track on constructing students’ attitude towards convergence.
Science students dominated the group of HKC which the member has high knowledge of convergence. They
also dominated the group of HIC which the member has a high interest in convergence. However, group LSeC
with lower knowledge and interest, especially in self-efficacy of convergence, was dominated by students on
humanities track.
The previous discussion stated that science track had many collaborated activities in the classroom. It
showed the teaching-learning process at the school could build the knowledge and interest of students’ attitude
toward convergence. Therefore, it is suggested to provide a program of ‘customized learning’ by the school to
improve students’ attitude toward convergence, particularly for these three types of group. Customized learning
might adjust the learning process based on students’ various needs and their abilities to provide an effective
environment for education (Park & Kim, 2008). Customize learning is related to the differentiation instruction
program where the instruction is designed based on “whom the teacher teach” or the need of each group of
student. This learning can promote the students’ creative idea, problem-solving, respectful with a different
environment, challenges skill since it is designed based on students’ learning preference and personal interest
(Tomlinson & McTighe, 2006).ba
For these three cases of groups, a different learning strategy could be implemented by a teacher. The HKC
group is the group with high knowledge but low interest in attitude toward convergence. A teacher should
emphasize to improve students’ interest by encouraging them through motivation. Glynn et al. (2009) have
recognized career motivation as one of the extrinsic motivations, which could influence students’ conviction
regarding the relation of knowledge gained and future career. This future career motivation might increase
students’ interest in convergence because intrinsic motivation and career motivation were strongly correlated.
Therefore, it is recommended to involve career motivation in learning instruction (Glynn et al., 2011), especially
for HKC group.
429
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
The HIC group is the group with high interest but low knowledge of convergence. Many factors could
influence students’ knowledge (e.g., teacher’s knowledge, teaching model). Knowledge of a teacher is one of
the essential elements that influenced what is being conducted in classrooms, and it gives major effect on
what students learn (Fennema & Franke, 1992; Hill et al., 2005). That is why a teacher should be familiar not only
about the knowledge of convergence but also about how to implement it in the classroom (various learning
models in convergence). Furthermore, the LSeC group had lower mean values for all constructs compared to
other groups. The most important thing about this group is a large number of students (60.37%) with very
low self-efficacy in convergence. This group could combine the learning strategy from groups HKC and HIC.
However, teachers need to emphasize on how to improve students’ self-efficacy. Several types of research
showed the implementation of vicarious learning used to enhance students’ confidence (Albion, 1996; Ertmer
et al., 2003; Wang and Newby, 2004). However, Schunk (2001) suggested that increasing self-efficacy could be
conducted by evaluating self-performance and making realistic goals. It is because when students achieve a
realistic goal, they tend to be more confident to reach other goals.
Conclusions
Due to increasingly complex problems around the world such as global warming, integrated and multidisci-
plinary approaches that transcend boundaries to traditional academic fields are becoming increasingly important.
This movement of convergence has strengthened a variety of integrated approaches in the education field recently.
Particularly in current science education, different contents from previously separated scientific disciplines are
connected in teaching and learning through an integrated approach. Based on these approaches, students are
expected to grow into future generations with the ability to overcome disciplinary barriers and solve creatively
various complex problems in the world. However, to achieve this purpose effectively, it is crucial to focus on inte-
grated content and students’ positive attitude toward convergence such as knowing the meaning and purpose
of convergence well, understanding the relation to their life, and having interest and confidence in convergence.
Positive attitude toward convergence is expected to be a driving force for attempting a newly convergence.
This study uncovered the relationship of attitude toward convergence with gender and track in Indonesia
Senior High School students. The responses provided empirical data that gender was significantly related to
students’ interest in convergence, where female students had a higher interest in convergence than male students
because females had more social sensitivity in working collaboratively. This finding suggests that a learning
group will have a better result if more female students are in the group. However, female students were less
confident in attitude towards convergence. The teacher needs to be noticed when designing a teaching design
to improve female students’ confidence in convergence education. Furthermore, the track was also significantly
related to students’ interest in convergence where science track students had higher interest than humanities
track. It is because the science track had more collaborative learning activity which can make students more
interested in working together. Science class students also tend to believe that the attitude toward convergence
is very good and beneficial for their future career. Customized learning based on students’ class characteristics
is proposed in this study to optimize effective teaching-learning, specifically to create students’ good attitude
toward convergence. However, for effective improving students’ attitude toward convergence, further researches
regarding on educational environment and teaching program are needed to be conducted in the future.
Acknowledgments
This research was supported by the National Research Foundation of Korea (NRF-2017R1D1A1B03035881).
References
Albion, P. R. (1999). Self-efficacy beliefs as an indicator of teachers’ preparedness for teaching with technology. In Society
for Information Technology & Teacher Education International Conference (pp. 1602-1608). Waynesville, NC: Association
for the Advancement of Computing in Education (AACE).
Bear, J. B., & Woolley, A. W. (2011). The role of gender in team collaboration and performance. Interdisciplinary Science Re-
views, 36(2), 146-153.
Bender, L., Walia, G., Kambhampaty, K., Nygard, K. E., & Nygard, T. E. (2012). Social sensitivity and classroom team projects:
an empirical investigation. In Proceedings of the 43rd ACM technical symposium on Computer Science Education (pp.
403-408). New York, NY: Association for Computing Machinery (ACM).
430
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
Boone, W. J., Staver, J. R., & Yale, M. S. (2014). Rasch analysis in the human sciences. The Netherlands: Springer Science & Busi-
ness Media. doi: https://doi.org/10.1007/978-94-007-6857-4.
Bowen, C. W. (2000). A quantitative literature review of cooperative learning effects on high school and college chemistry
achievement. Journal of Chemical Education, 77(1), 116.
Britner, S. L. (2008). Motivation in high school science students: A comparison of gender differences in life, physical, and
earth science classes. Journal of Research in Science Teaching: The Official Journal of the National Association for Research
in Science Teaching, 45(8), 955-970.
Burke, R. J., & Mattis, M. C. (Eds.). (2007). Women and minorities in science, technology, engineering, and mathematics: Upping
the numbers. Northampton, MA: Edward Elgar Publishing.
Bybee, R. W. (2010). Advancing STEM education: A 2020 vision. Technology and Engineering Teacher, 70(1), 30.
Cavallo, A. M., Potter, W. H., & Rozman, M. (2004). Gender differences in learning constructs, shifts in learning constructs,
and their relationship to course achievement in a structured inquiry, yearlong college physics course for life science
majors. School Science and Mathematics, 104(6), 288-300.
Cha, H. J., Kim, Y. S., Park, S. H., Yoon, T. B., Jung, Y. M., & Lee, J. H. (2006, June). Learning styles diagnosis based on user inter-
face behaviors for the customization of learning interfaces in an intelligent tutoring system. In: Ikeda, M., Ashley, K.
D., & Chan, T. W. (Eds.), Intelligent tutoring systems. ITS 2006. Lecture Notes in Computer Science, vol. 4053 (pp. 513-524).
Springer, Berlin, Heidelberg. https://doi.org/10.1007/11774303_5.
De Vault, J. E. (1998). A competition-motivated, interdisciplinary design experience. In FIE’98. 28th Annual Frontiers in Edu-
cation Conference. Moving from ‘Teacher-Centered’ to ‘Learner-Centered’ Education. Conference Proceedings (Vol. 1, pp.
460-465). Piscataway, NJ: IEEE.
Directorate of Development of Indonesia Senior High School. (2017). Naskah pendukung implementasi Kurikulum 2013: Model
Peminatan dan Lintas Minat [Supporting manuscript for 2013 Curriculum implementation: Model of Specialization across
the Interest]. Retrieved from https://docplayer.info/58146016-Model-peminatan-dan-lintas-minat.html.
Embretson, S. E., & Reise, S. P. (2013). Item response theory. Hove: Psychology Press.
Ertmer, P. A., Conklin, D., Lewandowski, J., Osika, E., Selo, M., & Wignall, E. (2003). Increasing preservice teachers’ capacity for
technology integration through the use of electronic models. Teacher Education Quarterly, 30(1), 95-112.
Felder, R. M., & Silverman, L. K. (1988). Learning and teaching styles in engineering education. Engineering Education, 78(7),
674-681.
Fennema, E., & Franke, M. L. (1992). Teachers’ knowledge and its impact. In D. A. Grouws (Ed.), Handbook of research on
mathematics teaching and learning: A project of the National Council of Teachers of Mathematics (pp. 147-164). New
York, NY, England: Macmillan Publishing.
Fisher, J. W. (2000). Objectivity in psychosocial measurement: What, why, how. Journal of Outcome Measurement, 4(2), 527-563.
Fisher, W. P. Jr (2007). Rasch measurement transaction. Transaction of the Rasch Measurement SIG American Educational
Research Association, 21(1), 1095.
Glynn, S. M., Brickman, P., Armstrong, N., & Taasoobshirazi, G. (2011). Science motivation questionnaire II: Validation with
science track and nonscience track. Journal of Research in Science Teaching, 48(10), 1159-1176.
Glynn, S. M., Taasoobshirazi, G., & Brickman, P. (2009). Science motivation questionnaire: Construct validation with nonscience
majors. Journal of Research in Science Teaching: The Official Journal of the National Association for Research in Science
Teaching, 46(2), 127-146.
Godemann, J. (2008). Knowledge integration: A key challenge for transdisciplinary cooperation. Environmental Education
Research, 14(6), 625-641.
Goel, A., & Bardhan, S. (2016). Effect of stress on self-efficacy and emotional intelligence among college students of humani-
ties and sciences: A study on gender differences. International Journal of Applied Research, 2(12), 318-328.
Good, C., Aronson, J., & Harder, J. A. (2008). Problems in the pipeline: Stereotype threat and women’s achievement in high-
level math courses. Journal of Applied Developmental Psychology, 29(1), 17-28.
Gibson, H. L., & Chase, C. (2002). Longitudinal impact of an inquiry-based science program on middle school students’ at-
titudes toward science. Science education, 86(5), 693-705.
Ha, M., & Lee, J. K. (2012). Exploring variables related to students’ understanding of the convergence of basic and applied
sciences. Journal of the Korean Association for Science Education, 32(2), 320-330.
Hacklin, F., & Wallin, M. W. (2013). Convergence and interdisciplinarity in innovation management: A review, critique, and
future directions. The Service Industries Journal, 33(7-8), 774-788.
Hayden, K., Ouyang, Y., Scinski, L., Olszewski, B., & Bielefeldt, T. (2011). Increasing student interest and attitudes in STEM:
Professional development and activities to engage and inspire learners. Contemporary Issues in Technology and Teacher
Education, 11(1), 47-69.
Hill, H. C., Rowan, B., & Ball, D. L. (2005). Effects of teachers’ mathematical knowledge for teaching on student achievement.
American Educational Research Journal, 42(2), 371-406.
Hur, J., Nam, D., Kwon, S., & Lee, T. (2010). Design of the Convergence Study Program based Educational Robot. In Proceed-
ings of the 18th International Conference on Computers in Education, Malaysia (pp. 693-697). Jhongli: Asia-Pacific Society
for Computers in Education.
Jeong, H., & Chi, M. T. (2007). Knowledge convergence and collaborative learning. Instructional Science, 35(4), 287-315.
Jordan, K., & Carden, R. (2017). Self-efficacy and gender in STEM track. Modern Psychological Studies, 22(2), 8.
Kitts, C., & Quinn, N. (2004). An interdisciplinary field robotics program for undergraduate computer science and engineering
education. Journal on Educational Resources in Computing (JERIC), 4(2), 3.
431
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
Klein, J. T. (2005). Integrative learning and interdisciplinary studies. Peer Review, 7(4), 8-10.
Klingorová, K., & Havlíček, T. (2015). Religion and gender inequality: The status of women in the societies of world religions.
Moravian Geographical Reports, 23(2), 2-11.
Lee, J. K., Lee, T. K., & Ha, M. (2013). Exploring the evolution patterns of trading zones appearing in the convergence of
teachers’ ideas: The case study of a learning community of teaching volunteers steam teacher community. Journal of
the Korean Association for Science Education, 33(5), 1055-1086.
Lee, J. K., Shin, S., Rachmatuallah, A., & Ha, M. (2017). The relationship of engineering education accreditation program,
gender, and academic year with attitude towards convergence among engineering students: Application of latent
clases analysis. Journal of the Korean Association for Science Education, 37(1), 113-123.
Linacre, M. A. (2017). Users Guide to Winsteps/ministep Rasch Model Computer Programs 2017. Retrieved from https://www.
winsteps.com/manuals.htm.
Marginson, S., Tytler, R., Freeman, B., & Roberts, K. (2013). STEM: Country comparisons: International comparisons of science,
technology, engineering and mathematics (STEM) education. Final report. Melbourne, VIC: ACOLA.
Martin, M. O., Mullis, I. V. S., Foy, P., & Hooper, M. (2016). TIMSS 2015 International Results in Science. Chestnut Hill, MA: TIMSS
& PIRLS International Study Center, Boston College.
Messick, S. (1995). Validity of psychological assessment: Validation of inferences from persons’ responses and performances
as scientific inquiry into score meaning. American Psychologist, 50(9), 741.
National Academy of Sciences. 2007. Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and
Engineering. Washington, DC: The National Academies Press. https://doi.org/10.17226/11741.
National Research Council. (2011). Successful K-12 STEM education: Identifying effective approaches in science, technology,
engineering, and mathematics. Washington, D.C: National Academies Press.
National Research Council. (2014). Convergence: facilitating transdisciplinary integration of life sciences, physical sciences,
engineering, and beyond. Washington, D.C: National Academies Press.
National Science Foundation, USA. Convergence Research at NSF. Retrieved from https://www.nsf.gov/od/oia/convergence/
index.jsp.
Nugent, G., Barker, B., Grandgenett, N., & Adamchuk, V. I. (2010). Impact of robotics and geospatial technology interventions
on youth STEM learning and attitudes. Journal of Research on Technology in Education, 42(4), 391-408.
OECD. (2016). PISA 2015 Results: Excellence and Equity in Education, PISA (Vol. I). Paris: OECD Publishing.
Park, S., & Kim, Y. (2008). Applying petri nets to model customized learning and cooperative learning with competence.
International Journal of Computer Science and Network Security, Seoul, 8(2), 127-132.
Patall, E. A., Steingut, R. R., Freeman, J. L., Pituch, K. A., & Vasquez, A. C. (2018). Gender disparities in students’ motivational
experiences in high school science classrooms. Science Education, 102(5), 951-977.
Prince, M. (2004). Does active learning work? A review of the research. Journal of Engineering Education, 93(3), 223-231.
Ravand, H., & Firoozi, T. (2016). Examining construct validity of the master’s UEE using the Rasch model and the six aspects
of the Messick’s framework. International Journal of Language Testing, 6(1), 1-23. Retrieved from http://ijlt.ir/journal/
images/PDF/440-2016-6-1.pdf.
Rousseau, R. (2001). Are multi-authored articles cited more than single-authored ones? Are collaborations with authors from other
countries more cited than collaborations within the country? A case study. Berlin: Gesellschaft für Wissenschaftsforschung.
Schunk, D. H. (2001). Self-regulation through goal setting. ERIC Clearinghouse on Counseling and Student Service, University
of North Carolina at Greensboro.
Scrucca, L., Fop, M., Murphy, T. B., & Raftery, A. E. (2016). Mclust 5: Clustering, classification and density estimation using
gaussian finite mixture models. The R Journal, 8(1), 289.
Shin, S., Ha, M. S., & Lee, J. K. (2014a). Difference analysis between groups and the generalizability of the instrument for
measuring high school students’ attitude toward convergence. Journal of Learner-Centered Curriculum and Instruction,
14, 107-124.
Shin, S., Ha, M., Lee, J. K., Park, H., Chung, D. H., & Lim, J. K. (2014b). The development and validation of instrument for
measuring high school students’ attitude toward convergence. Journal of the Korean Association for Science Education,
34(2), 123-134.
Shin, S., Rachmatullah, A., Roshayanti, F., Ha, M., & Lee, J. K. (2018). Career motivation of secondary students in STEM: A
cross-cultural study between Korea and Indonesia. International Journal for Educational and Vocational Guidance,
18(2), 203-231.
Stump, G. S., Hilpert, J. C., Husman, J., Chung, W. T., & Kim, W. (2011). Collaborative learning in engineering students: Gender
and achievement. Journal of Engineering Education, 100(3), 475-497.
Sumintono, B., & Widhiarso, W. (2014). Aplikasi model Rasch untuk penelitian ilmu-ilmu sosial (edisi revisi) [Application of
the Rasch model for social sciences research (revised edition)]. Cimahi: Trim Komunikata Publishing House.
Suprapto, N. (2016). Students’ attitudes towards STEM education: Voices from Indonesian junior high schools. Journal of
Turkish Science Education, 13(3), 75-87.
Taylor, A., & Greve, H.R. (2006). Superman or the fantastic four? Knowledge combination and experience in innovative teams.
Academy of Management Journal, 49(4), 723–740.
Tomlinson, C. A., & McTighe, J. (2006). Integrating differentiated instruction & understanding by design: Connecting content
and kids. Alexandria, VA: ASCD.
Tornabene, R. E., Lavington, E., & Nehm, R. H. (2018). Testing validity inferences for Genetic Drift Inventory scores using Rasch
432
CONVERGENCE
ISSN 2538–7138 /Online/ (P. 417-434)
modeling and item order analyses. Evolution: Education and Outreach, 11(1), 6.
United Stated Department of Labor. (2015). Employment change and percentage employment change by type of STEM occupa-
tion, May 2009 to May 2015. Retrieved from https://www.bls.gov/spotlight/2017/science-technology-engineering-and
mathematics-stem-occupations-past-present-and-future/home.htm.
United Stated Department of Labor. (2015). STEM employment by type of STEM occupation, May 2015. Retrieved from https://
www.bls.gov/spotlight/2017/science-technology-engineering-and-mathematics-stem-occupations-past-present-
and-future/home.htm.
Valderas, J. M. (2007). Why do team-authored papers get cited more?. Science, 317(5844), 1496-1498.
Vermeulen, N., Parker, J. N., & Penders, B. (2013). Understanding life together: A brief history of collaboration in biology.
Endeavour, 37(3), 162-171.
Wang, L., Ertmer, P. A., & Newby, T. J. (2004). Increasing preservice teachers’ self-efficacy beliefs for technology integration.
Journal of Research on Technology in Education, 36(3), 231-250.
Weinberger, A., Stegmann, K., & Fischer, F. (2007). Knowledge convergence in collaborative learning: Concepts and assess-
ment. Learning and Instruction, 17(4), 416-426.
Willey, K., & Gardner, A. (2012). Collaborative learning frameworks to promote a positive learning culture. In 2012 Frontiers
in Education Conference Proceedings (pp. 1-6). Piscataway, NJ: IEEE.
Winarno, N., Widodo, A., Rusdiana, D., Rochintaniawati, D., & Afifah, R. M. A. (2017). Profile of Pre-Service Science Teachers
Based on STEM Career Interest Survey. In Journal of Physics: Conference Series (Vol. 895, No. 1, p. 012170). Bristol, UK:
IOP Publishing.
Woolley, A. W., Chabris, C. F., Pentland, A., Hashmi, N., & Malone, T. W. (2010). Evidence for a collective intelligence factor in
the performance of human groups. Science, 330(6004), 686-688.
Woolley, A., & Malone, T. (2011). What makes a team smarter? More women. Harvard Business Review, 89(6), 32-33.
Wright, B. D., & Linacre, J. M. (1994). Reasonable mean-square fit values. Rasch Measurement Transactions, 8(3), 370.
Wuchty, S., Jones, B. F., & Uzzi, B. (2007). The increasing dominance of teams in production of knowledge. Science, 316(5827),
1036-1039.
Yuen, T., Boecking, M., Stone, J., Tiger, E. P., Gomez, A., Guillen, A., & Arreguin, A. (2014). Group tasks, activities, dynamics, and
interactions in collaborative robotics projects with elementary and middle school children. Journal of STEM Education,
15(1), 39-45.
Appendixes
Appendix 1. The instrument of attitude toward convergence in the English version.
Construct of
‘Attitude toward Code No Statement in English Version
Convergence’
Knowledge K1 1 I can explain the meaning of “convergence” to others.
K2 2 We can mention the case of convergence.
K3 3 I can explain the difference between learning convergence with other learning.
K4 4 The attitude toward convergence can be used in everyday life.
Personal Relevance R_P1 5 Convergence is related to what I learned.
R_P2 6 Understanding convergence can help me in learning.
R_P3 7 Convergence will be related to my life in the future.
R_P4 8 Convergence helps in understanding the things I am curious about and interested in.
R_P5 9 Convergence will solve the problems of my daily life.
Social Relevance R_S1 10 Convergence will help to solve the complex problems of modern society.
R_S2 11 Convergence will help society to develop in the future.
R_S3 12 Convergence will help me understand modern civilization and society.
R_S4 13 Convergence will help us to prepare future society of the 21st century
433
CONVERGENCE
(P. 417-434) ISSN 2538–7138 /Online/
Construct of
‘Attitude toward Code No Statement in English Version
Convergence’
Interest A_I1 14 Finding the connection between a field of knowledge and other fields of knowledge is
interesting.
A_I2 15 I enjoy the process of getting to know various fields of science related to the subject that I
am interested in.
A_I3 16 Participating in the process of creating ideas with people who have different ideas is fun.
A_I4 17 It is very interesting if experts from various fields can collaborate to solve problems.
The experience of people who have applied their ideas creatively from one field to another
A_I5 18
is interesting.
Self-Efficacy A_SE1 19 I am good at finding links between knowledge in one field and other fields.
A_SE2 20 I know well the various fields of knowledge that are related to the subject I am interested in.
A_SE3 21 I am good at solving problems using knowledge from various fields.
A_SE4 22 I am good at seeing one problem from various points of view.
A_SE5 23 I am good at using knowledge from various fields when trying to solve the problems.
Received: January 21, 2019 Accepted: May 25, 2019
Yustika Sya’bandari S.Pd., Master Student, Division of Science Education, College of

Education, Kangwon National University, 1 Kangwondaehak-gil,
Chuncheon-si, Gangwon-do, 24341 Republic of Korea.
E-mail: yustikasya@kangwon.ac.kr
Minsu Ha Ph. D., Associate Professor, Division of Science Education, College of
Education, Kangwon National University, 1 Kangwondaehak-gil,
Chuncheon-si, Gangwon-do, 24341 Republic of Korea.
E-mail: msha@kangwon.ac.kr
Jun-Ki Lee Ph. D., Professor, Institute of Science Educaton and Division of
Science Education, Chonbuk National University, 567 Baekje-daero,
Jeonju, Republic of Korea.
E-mail: junki@jbnu.ac.kr
Sein Shin Ph. D., Assistant Professor, Department of Biology Education,
(Corresponding author) Chungbuk National University, Chungdae-ro 1, Seowon-Gu,
Cheongju, Republic of Korea.
E-mail: sein3027@gmail.com seinshin@chungbuk.ac.kr
434
MONODISCIPLINARITY
IN SCIENCE VERSUS
TRANSDISCIPLINARITY IN STEM ISSN 1648-3898 /Print/
EDUCATION ISSN 2538-7138 /Online/
Mitja Slavinec, Abstract. Contemporary society of the

fourth industrial revolution (Industry 4.0)
Boris Aberšek, increasingly requires the education system
Dino Gačević, (i.e., the school) to train competent, creative
Andrej Flogie and proactive professionals who will be
able to solve real life problems. If society is
to achieve this, some key paradigm chang-
es must occur in education. The school
must first prepare a competence-based
Introduction curriculum and, secondly, school practice
should move away from subject-based
In today’s increasingly results-driven and performance-oriented world, teaching, towards an interdisciplinary STEM
the school system – with the necessary distance, of course, – is trying to (and teaching approach. Obviously, to support
should) follow such requirements. The question of an appropriate distance is this, modern learning environments and
very much relevant here, as “fast” changes are not always good or successful. ICT solutions and tools have to be used.
In recent years (roughly in the period after the year 2000), especially under However, since the interdisciplinary STEM
the influence of modern technologies and, in the last decade, under the in- approach has already been implemented
fluence of the Industry 4.0 guidelines, educational paradigms are changing and integrated, it can be said that a new,
dramatically, or at least, they should be. To name only a few of these changes: integrated science discipline (STEM) has
already emerged, together with a transdis-
•• from traditional, teacher-centred teaching, to innovative, student- ciplinary approach to STEM learning and
centred forms of education, teaching.
•• from the passive learner (or, at a philosophical (paradigmatic) In the present research, a concrete case of
level, the behaviourist learning theory), to the active learner (or, designing, developing and producing a
the constructivist and cognitive theory), solar chimney was used to demonstrate an
•• from traditional learning methods, to brain-based learning (BBL), integrated approach to learning and teach-
which is based on findings from contemporary cognitive and ing, while emphasizing especially the ad-
neuroscience, vantages of such an interdisciplinary (trans-
•• from technology-based learning environments that required disciplinary) approach to teaching Science,
the learners and teachers to adapt to technology, to learner- Technology, Engineering and Mathematic
and teacher-centred environments (Žbona, 2016), where the content. The empirical research shows that
technology adapts to the needs of learners and teachers, and, such an approach produces incomparably
last but not least, better results, especially on higher cognitive
•• from subject-based teaching, to increasingly interdisciplinary ori- levels, in comparison to traditional ap-
ented forms of teaching, supported, among other, by contempo- proaches to learning and teaching.
rary learning strategies, such as problem-based, research-based, Keywords: industry 4.0, interdisciplinarity,
and project-based learning (Aberšek, 2018, Key Competences for solar chimney, STEM, transdisciplinarity.
Lifelong Learning, 2006, Mayer, 2010).
Innovative education places various contemporary and innovative forms Mitja Slavinec, Boris Aberšek, Dino
of organizing and performing teaching at its centre, where an integrative Gačević, Andrej Flogie
University of Maribor, Slovenia
principle and a systemic approach to the education process are constantly
emphasized:
435
(P. 435-449)
•• Pedagogy 1:1 (Aberšek et al., 2017) underlines the significance of individualization and differentiation
of teaching based on competence frameworks,
•• the 20 Keys philosophy in education, inspired by Kobayashi’s theory of organizing industrial processes
from the early 1970s (Kobayashi, 2003), highlights the importance of a continuous and even develop-
ment of all areas of a system, from the simplest, such as maintaining order and cleanliness, to the most
complex. He accentuates the idea that only an equal development of all areas of a system can lead to a
successful, optimised and thereby maximum effective production, supported by the total productivity
management (TPM) approach. A similar philosophy is being introduced to the education system via
the 21 steps system (Aberšek, et al., 2017, Clark & Svanaes, 2012; Microsoft Partners in Learning, 2010).
In focusing specifically on an integrated approach to performing teaching, the basic level of the education
process itself should integrate at least the learners, as subjects of education, teachers, as managers and operators
of the teaching process, and learning environments, which consist of learning technologies and methods and
strategies of learning and teaching.
This research focused only on the latter, i.e., learning environments, their impact on the process of learning
and teaching, or, in other words, their impact on the basic research question of this research, which is how can
changes in learning environments lead to improved results in both teaching and learning, and exert a positive effect
on the efficiency of the learning process as a whole? In order to achieve this, the (existing) learning process needs to
undergo significant changes, namely:
•• overcoming the division into individual subject areas and moving beyond the boundaries of individual
scientific disciplines, and
•• introducing methods and strategies that place students in an active role, making them co-responsible
for their success in acquiring knowledge and competences, and thereby increasing their level of interest
and motivation. Such methods are more or less based on research-based learning (RBL) and problem-
based learning (PBL) (Aberšek, 2018).
Therefore, in order to form a complex, competence-based view on knowledge, and in order to direct learn-
ers towards a comprehensive approach to solving real-life problems, all of the above mentioned methods and
strategies are connected into an integrated, complex strategy called project-based learning (PrBL). It needs to be
stressed that projects are a form of team work, a group effort that requires collaborative learning, which means
that learners involved in the process also develop their social competence and cooperativeness both on the level
of implementation and organization (distribution of roles in a project, communication between the participants,
negotiations, etc.). If in the context of RBL or PBL problems may still be “artificially” created, and solutions to the
problem are proposed within “laboratory and local” frameworks of a particular subject (of a single individual dis-
cipline), then it can be said that project-based learning (PrBL) no longer considers teaching and learning from the
standpoint of an individual discipline (i.e., monodisciplinary), but rather, it approaches the given problem in an
interdisciplinary manner. However, since STEM is a kind of interdisciplinary, interconnected discipline in itself, it
seems more appropriate to apply the term transdisciplinary to refer to the birth of a new discipline, which forms
its own working and research methods (Flogie & Aberšek, 2015). Thus, a transdisciplinary approach creates a new
perspective on learning and teaching, giving it a concrete expression and relating it to real-life situations and
problems, to the real-life social environment and the requirements of modern society (Industry 4.0), for which the
school should educate, whether creator of the school policy agree with it or not.
Course Development
General Background
In the proposed study the traditional approach to teaching physics on the example of renewable energy
sources and ways of exploiting solar energy was briefly summarized. Unfortunately, such an approach, which is still
predominant, has produced poor results in terms of enabling creativity and integrating knowledge for solving real-
436
ISSN 1648–3898 /Print/ MONODISCIPLINARITY IN SCIENCE VERSUS TRANSDISCIPLINARITY IN STEM EDUCATION
(P. 435-449)
life problems (Hussain, Azeem, & Shakoor, 2011, Deslauriers, Schelew, & Wieman, 2011). A project-based approach
to teaching similar subject-matter was presented in more detail, with a focus on using modern teaching technolo-
gies and contemporary learning approaches. An analysis and comparison of the two approaches was performed.
Development – A Case Study
The subject Applied Physics consisted of 60 contact hours and 60 hours of individual student work. On the
topic of Solar Energy, a course of 12 contact hours was organized. Short course description and/or study design is
presented in Table 1.
Table 1. Description of study design and procedures of the teaching/learning process.
Multidisciplinary STEM approach - project-based

Study Traditional approach (Example 1)
learning (Example 2)
Duration 12 contact hours 12 contact hours

12 hours of individual student work (according to curricula) ? hours of individual student work
Methods Problem-based learning Traditional teaching/learning methods such as lecture,
Research-based learning frontal teaching, homework, frontal instruction lesson,
Collaborative learning etc.
Team working
On a concrete example of designing and optimising a solar chimney, this research compared two learning
and teaching strategies, pointing out their advantages and disadvantages:
•• the traditional approach (Example 1), related to the individual discipline, i.e., the monodisciplinary
approach, and
•• the multidisciplinary (transdisciplinary – STEM) approach (Example 2), related to solving real-life prob-
lems and founded on the idea of project-based learning.
Example 1: Traditional education in Applied physics (monodisciplinary approach) – Theory

about energy and solar power
Focusing initially on the lesson about energy: it is common knowledge that two basic types of energy sources
exist in the world. There are conventional sources of energy, most of which are non-renewable, for example fossil
fuel, and renewable sources of energy, such as water or wood. Among the renewable ones, solar energy is the
most prospective. A wide range of existing power technologies can make use of the solar energy reaching earth.
Ways of harnessing solar energy can be divided into two basic categories: direct (transformed for use elsewhere
or utilized directly) and indirect (involving more than one transformation to reach a usable form).
The output power of the plant depends upon various parameters presented simply by the following equation:
Pout = Qsolar*ηcoll*ηtower*ηturbine = Qsolar*ηplant (1)
If the temperature rise in the collector is (Ti – To) then it can be easily expressed as:
ηcoll*I* Acoll = ṁ* Cp* (Ti – To) (2)
The input energy from the sun, Qsolar, can be expressed as:
Qsolar = I*Acoll = I*(π/4)*D2 c (3)
One of the possibilities to study the use of solar energy is through the sun tower (sun chimney), which converts
the heat-flow produced by the collector into kinetic energy (convection current) and potential energy (pressure
drop at the turbine). Thus, the density difference of the air caused by the temperature rise in the collector works as
437
(P. 435-449)
a driving force. The lighter column of air in the tower is connected with the surrounding atmosphere at the base
(inside the collector) and at the top of the tower, and thus acquires lift. A pressure difference Δptot is produced
between tower base (collector outlet) and the ambient (White, 1999):
Δptot = g* ∫0H(ρ0-ρi) dH (4)
This is simplified to:
Δptot = g(ρ0-ρi) Ht (5)
The static pressure difference drops at the turbine, the dynamic component describes the kinetic energy of
the airflow. With the total pressure difference and the volume flow of the air at Δp s = 0 the power Pout contained
in the flow is now:
Pout = Δptot*vt*Acoll (6)
The mass flow rate is:

ṁ = ρi*At*vt=ρi*(π/4)*D2 t *vt (7)
Thus, without the turbine installed, the total power available to the turbine can be obtained from equation
(6). And also, the velocity at the entrance is found by,
vt= √{2gHt(Ti – To)/To} (8)
Nomenclature:
Simbol Meaning and units
Qsolar Solar power input to the plant (W)
Ηcoll Collector efficiency
Ηtower Tower efficiency
ηturbine Turbine efficiency
ηplant Plant efficiency
I Solar Intensity (earth surface) (W/m2)
Δptot Total pressure difference (N/m2)
g Gravitational acceleration (9.8 m/s2)
To Ambient/outside temperature (oC)
Ti Temperature at tower entrance (oC)
ρ0 Outside air density (kg/m3)
ρi Air density at tower entrance (kg/m3)
Ht Height of the tower (m)
vt Air velocity at tower entrance (m/s)
Dc Diameter of the collector (m)
Acoll Area of the collector (m2)
Dt Diameter of the chimney (m)
At Area of the chimney (m2)
438
(P. 435-449)
It is only after the students have mastered the theoretical basics of physics, that they can move on to calcu-
lations and optimization of the solar chimney, i.e., to suitable experiments, to the technology and engineering
part. This process, however, is time-consuming and requires a high degree of abstract thought and good spatial
awareness skills.
Example 2: Project-based learning about energy (STEM approach)
The lesson could start by asking questions such as:
•• How have people harnessed solar power through history?

•• What must we know about using solar power?
•• What kinds of elements does a solar power plant contain?
•• Etc.,
and finally:
•• How can we use solar power today? → Answer could be: By Solar chimney
Research start: A solar chimney – often referred to as a thermal chimney – is a way of improving the natural
ventilation of buildings by using convection of air heated by passive solar energy. A solar chimney operates on
the same principles as a fireplace chimney in houses. A simple description of a solar chimney is that of a vertical
shaft utilizing solar energy to enhance the natural stack ventilation through a building. But is there another way
in which the solar chimney can be used? It can also be used as a power plant to produce electricity!
What is needed to construct a solar chimney? How to build such a solar power plant? With these questions, the
lesson could start, students continue their research and try to come up with an optimal solution.
Figure 1. Simple diagram showing the functional principle of solar chimney plants.
Figure 2. Simple diagram showing the parameters for theory.
439
(P. 435-449)
Research: What do we have to know? → The theory is learnt subsequently!
In the following stage, students begin designing the construction with the help of CAD programs. Possible
solutions are shown in Figures 3 and 4. Students build the construction and then start experimenting using differ-
ent engineering tools (for example ANSYS) for simulation – changing temperature, pressure, etc. and observing
what happens to the speed and pressure of the air flow (for examples, see Figures 6–11), as they attempt to find an
optimal solution. Simple physical models can be built by the students, or they can use numerical tools for calculat-
ing physical values. A draft description of the simulation (experimental) procedure is provided below:
1) the outside temperature T0 was the input data, it was measured by thermometer;
2) temperature Ti was simulated by a computer program (it could also have been measured at the same
time (as the outside temperature) by thermometer);
3) the air velocity at the entrance of the tower vt was calculated/simulated, as shown in Figure 7 (it could
also have been taken by using the anemometer);
4) the densities ρi and ρo were simulated automatically by means of numerical simulation (they could also
have been taken corresponding to the temperatures Ti and T0 respectively);
d) steps 1 to 4 were repeated in any new simulation (this could have been done experimentally, by repeat-
ing them once every half an hour for the time period of one day);
e) in this way, data was collected for six (6) simulations (experimental days) and in turn, the necessary
calculations were made.
Simulation as a Learning Strategy
Presented experimental procedure can be simulated (Žbona, 2016) by means of Finite Element Analysis (FEA),
the Finite Element Method (FEM), with various computer programs (such as ANSYS), in order to provide an optimal
solution easily and quickly.
Using one of the various engineering tools for modelling 3D constructions (Computer-Aided Design – CAD), it
is possible to model any desired form of construction, in the case of this research, a solar chimney. The advantage of
using CAD tools is in that they allow flexibility and rapid changes in the design, depending on the simulations per-
formed and the calculations obtained. An example of an initial construction of a solar chimney is shown in Figure 3.
Construction of the solar chimney
Construction Absorber
Figure 3. Simulation with using C AD possibilities (using SolidWorks).
Once the basic construction has been modelled with CAD tool, we can use this model as input model in
FEA, where first, surfaces need to be specified (Figure 4), to which boundary conditions will apply, before further
simulation using the Finite Element Method (FEM). For surface A (Figure 4), the pressure of 1 bar was prescribed,
and for surface B the pressure prescribed was lower by due to a height difference, which was
2,8 m between the two surfaces. Thus, the pressure difference was Pa. No value was prescribed to
other surfaces in this simulation.
440
(P. 435-449)
Figure 4. Defining boundary and initial conditions.
The next step is creating a volume mesh, see Figure 5. For the purposes of this research, a volume mesh was
created consisting of poly cells. For the creation of a volume mesh, it is necessary to prescribe the size of cells, the
same size across the entire mesh. The cell size in this research was defined so that approximately 5 to 10 cells fit
the thickness of the structure (0.1 m).
Figure 5. Volume mesh.
On the basis of such a volume mesh, various causal relationships can now be identified. Thus, the pressure
field shown in Figure 6 shows the pressure in the selected cross-section of the computational network. Figure 6
shows areas of elevated pressure at the centre of the chimney, which occur due to the merging of flows in that area.
441
(P. 435-449)
Figure 6. Pressure field in a solar chimney.
It is also possible to simulate what happens in terms of the velocity field, which shows the speed of the air
flow. Thus, for example, Figure 6 shows that the maximum speed of the air flow is where the solar roof joins the
main part of the chimney, which is also the optimal location for placing the turbine(s).
Figure 7. Velocity field of the air flow: a) inside a solar chimney and b) in direction y.
The velocity field in Figure 7 also shows that upon merging, part of the air flow is directed downward, which
is undesired (see Figure 8).
Figure 8. Air flow in the direction of the z axis.
442
(P. 435-449)
This undesired issue could be resolved by closing the bottom part of the chimney, while an even better solu-
tion would probably be to alter the way in which the entire construction is closed, which is shown in Figure 9 (see
square on bottom).
Figure 9. Theoretical (anticipated) air flow in a solar chimney.
Focusing more closely on the area where turbines could be placed for the production of electrical energy, the
course of the air flow can be demonstrated in various ways, including streamlines, which are shown in Figure 10.
Figure 10. The field of air flow inside a solar chimney.
The streamlines show the turbulence of air inside the solar chimney. This is confirmed by the turbulence field,
which is undesired in this case. Various simulations are used to determine what kind of form will result in a laminar
field. Figure 11 shows where the turbulent flow is most likely to occur.
443
(P. 435-449)
Figure 11. Turbulence field and turbulent flow occurrences.
From a comparison between the turbulence field in Figure 11 and the streamlines in Figure 10, it can be
concluded that the turbulent flow occurs mainly in the areas with a higher level of turbulent energy. It can also
be seen that the turbulent field is largely symmetrical (with regard to the symmetry of the chimney’s main pipe),
which is the expected result. On base of such analyses and simulation it is possible to find out optimal solution.
Sample
The research was conducted during the subject Applied Physics, which is taught at the Faculty of Natural Sci-
ences and Mathematics of the University of Maribor in the fifth year of the study course. The research included 20
respondents: 9 male students and 11 female students. The sample size depended on the total number of students
in this class, and represents nearly 80% of all students enrolled in this class.
General Background
According to the chapter Course Development lecture was presented to the students by means of a tradi-
tional, frontal instruction lesson. This is referred to as Example 1 in the following section. The teacher provided an
explanation of the topic and relevant formulations, while the students took notes and studied from them. During
the next lecture, two days later, they filled out a test.
After the period of one month, the test was repeated with the same questions and based on this, the forgetting curve
was calculated. The standardized Ebbinghaus forgetting curve is shown in Figure 12 (Ebbinghaus, 1908, Schacter, 2001).
Figure 12. Ebbinghaus forgetting curve.
444
(P. 435-449)
After the test, the lesson was repeated as a problem-based learning according to the described STEM model
(Example 2). Just like with Example 1, the students were asked the same questions two days after the initial lecture.
The test was repeated again, just like with Example 1, in one month’s time.
In both cases, the answers were assessed and evaluated individually and independently (by two professors,
two assistants and two students) with grades from 1 to 10, depending on the correctness and completeness of the
answers. The overall grade was calculated as an average of all individual evaluators’ grades.
Instruments
The differences in the effects of teaching strategies between Example 1 and Example 2 were tested by means
of a questionnaire in terms of the following indicators:
•• knowledge of the subject matter dealt with,

•• understanding, which is demonstrated by relationships / attitudes in the group, the students’ personal
development, and the systemic characteristics of individuals,
•• the ability to analyse and synthesize (the ability to reach higher cognitive levels) and
•• digital competence (digital skills of providing an analysis and evaluation using modern engineering
tools).
The effect of the applied method of work after the end of the lectures was evaluated according to four individual
phases of solving the problem (idea, design, production, evaluation), resulting in a common assessment. The
evaluation was performed under conditions of statistical control using a t-test, a Cohen’s d effect size calculation,
and a graphic representation. Each of the dimensions contained a specific number of statistically and graphically
validated propositions in the form of questions and task.
With the questionnaire the effect of the method of work after the lectures was measured. The questionnaire
consisted of 12 descriptive questions:
•• idea (3 questions),
•• design (3 questions),
•• modelling (3 questions),
•• evaluation and optimisation (3 questions).
that were performed as part of lectures on the theme of Design, construction and analysis of a solar chimney. The
effect of the method of work used by students was analysed after the lectures had ended:
•• the students’ interest in methods of learning and teaching, expressed as the overall result of all dimen-
sions of interest, shown in Figure 13 and
•• the students’ interest in methods of learning and teaching, expressed as the individual result of each
dimension of interest (idea, design, production, planning) separately, shown in Table 2.
Research Results
A graphic representation of distribution of arithmetic means is at Figure 13, and t-test, Cohen’s d effect of the
following dimensions of interest is presented in Table 2.
445
(P. 435-449)
Figure 13. Arithmetic means of evaluations of students’ interest according to the method of teaching (Example
1 (EX_1) or Example 2 (EX_2)) for individual dimensions.
Table 2. t-test results for independent samples of testing differences between the overall and individual
results for representability according to the group EX_1 and EX_2 after the experiment.
Arithmetic Standard Test of variance Arithmetic mean

Cohen’s
mean deviation homogeneity* differences test
Group
x̄ SD F p t p d
EX_2 7.935 1.181 5.063

Idea 27.589 .0001 .0001 1.099
EX_1 6.109 2.142 approximation
EX_2 12.065 1.665 6.809

Design 16.342 .0001 .0001 1.460
EX_2 12.848 1.429 5.548

Production 10.776 .001 .0001 1.178
EX_2 12.674 1.431 5.973

Evaluation 12.679 .001 .0001 1.296
EX_2 45.522 3.863 7.456

Method effect 21.743 .0001 .0001 1.646
* In all cases, the assumption of homogeneity of variance is not justified (p < .05), hence results of the approximation method are
reported.
The degree of forgetting in both learning situations was measured as well, and the results were compared
to known results from traditional teaching settings (Ebbenhard, 1908) and superlearning (Ostrander & Schroeder,
2012). The results are shown in the Figure 14.
446
(P. 435-449)
Figure 14. Diagram of the three forgetting curves.
For the presented case, an additional test was conducted only three months later. Example 1 produced similar
results to those obtained by Ebbenhard in his research, while for Example 2, the results were slightly lower than
in the case of superlearning.
Discussion
The graphic representation in Figure 13 shows that the line of Example 2 is above the line of Example 1 for all
dimensions (idea, design, production, evaluation), whereby the differences are significant, which was confirmed by
the results of statistical difference testing. The most marked differences between the two Examples are observed
in the case of design (it is here that most innovative tools and approaches were used, from CAD to FEA), and the
smallest differences in the dimension of ideas.
The t-test proved (Table 1) that the difference between Example 1 and Example 2 after the lectures ended
is statistically significant (p < .05) for all dimensions of the effect, i.e., for the individual dimensions (idea, design,
production, evaluation), as well as for the overall effect (interest):
•• idea (p = .0001, d = 1.099),

•• design (p = .0001, d = 1.460),
•• production (p = .0001, d = 1.178),
•• evaluation (p = 0001, d = 1.296),
•• summary effect size of the method of teaching (p = .0001, d = 1.646).
As demonstrated by measuring the effect size of Cohen’s d (d > 0.80), the effect size of the method is different
for both examples. The greatest effect was observed in designing the product (d = 1.460), followed by evaluation
(d = 1.296), production (d = 1.178), and searching for ideas (d = 1.099), where the effect size was the smallest. The
obtained average values of the evaluations (x̄ ) of interest in Table 1 show that the students evaluated all dimen-
sions of efficiency higher for Example 2 in comparison to Example 1, which consequently affects also the overall
efficiency of the method of teaching. With regard to Example 2, the students did better in coming up with ideas,
designing the product (chimney), producing, and evaluating the product. This means that the students, as regards
Example 2, made more progress in the phase of designing, than in the phase of evaluating the product. Indeed,
more time was spent on designing the product, especially on CAD modelling and engineering calculations and
analyses (FEA) with the help of the Finite Element Method (FEM), which was a novelty to the students.
Great efficiency was achieved also in the sense of memory retention: if traditional teaching led to a degree of
forgetting almost as high as 75%, after one month then the proposed approach has reduced the level of forget-
ting to 25%, which is very close to the standard curve of superlearning (Ostrander & Schroeder, 2012). It needs
to be taken into account, however, that these were not two different groups of students; the group in the case of
Example 2 was left with only some (25% at best) of the previous knowledge, but the progress was nonetheless
447
(P. 435-449)
evident. What is especially important, higher and in-depth cognitive levels of knowledge were stimulated in the
students, as well as critical thinking and competences for solving real-life problems, which is the essential pur-
pose of both, problem-based approaches to teaching, as well as the proposed STEM model. Similar results were
reported by various other authors, including (Hussain, Azeem, & Shakoor, 2011, Deslauriers, Schelew, & Wieman,
2011) and many others.
Conclusions
It should be emphasized that the findings derived from this research are preliminary, the research was carried
out on a small sample (mostly due to a small number of students in the class) and the research would certainly need
to be continued including a larger sample (also international), which would allow the formation of two separate
groups, the experimental one, where the proposed STEM model would be applied, and the control group, where
lectures would be delivered under the existing approach. It would also be reasonable to track the course of forget-
ting over a longer period of time, which is rather difficult within the academic environment, as lectures only last
one semester. Therefore, it would be a good idea to organize the survey as a project in a way that would allow the
monitoring of (larger groups of ) students over a longer period of time.
The obtained results, however, clearly show that the proposed approach resulted in a number of positive
shifts. To name only the most important ones:
•• the students’ interest was increased;

•• the students’ acquired knowledge and understanding were enhanced and finally
•• the students’ digital literacy was increased, and, it was proven that problems can be addressed and
solved in different ways. At the same time, the students were encouraged to form a comprehensive
approach to problem-solving.
Nevertheless, some possible disadvantages need to be mentioned as well. A one-dimensional use of con-
temporary engineering tools in STEM education could lead to a superficial understanding of the problem dealt
with. This method requires not only focusing on the results, but especially on the procedure itself and the level of
understanding, to avoid arriving to incorrect conclusions.
STEM learning and teaching is a reality that enables the acquisition of competences, which are important
for successful problem-solving in the 21st century society. Of course, STEM teaching and learning requires a mind
shift with regard to methods of imparting knowledge and conceptual changes in education from the primary to
the tertiary level. The presented example represents a concept of teaching for the society of the fourth industrial
revolution. The sooner the needs for conceptual changes in education are realized, the less painful these necessary
adaptations are going to be.
References
Aberšek, B. (2018). Problem-based learning and proprioception. Newcastle upon Tyne: Cambridge Scholars Publishing.
Aberšek, B., Barle Lakota, A., Borstner, B., Bregant, J., Dolenc, K., Flogie, A., Gartner, S., Kordigel Aberšek, M., Rutar Leban, T., Šverc,
A., Štraus, M., & Ploj Virtič, M. (2017). Cognitive science in education and alternative teaching strategies. Newcastle upon Tyne:
Cambridge Scholars.
Aberšek, B., Borstner, B., & Bregant, J. (2014). Virtual teacher: Cognitive approach to e-learning material. Newcastle upon Tyne:
Cambridge Scholars Publishing.
Alexandrov, D., Boukhanovsky, A., Chugunov, A., & Koltosova, O. (2017). Digital transformation and global society. St. Petersburg:
Springer.
Bernardes, M. A. D. S., Voß, A., & Weinrebe, G. (2003). Thermal and technical analyses of solar chimneys. Solar Energy, 75 (6), 511- 524.
Clarke, B., & Svanaes, S. (2012). One-to-one tablets in secondary schools: An evaluation study. London: Family Kids & Youth.
Deslauriers, L., Schelew, E., & Wieman, C. (2011). Improved learning in a large-enrollment physics class. Science, 332 (6031), 862-864.
Dhahri, A., & Omri, A. (2013). A review of solar chimney power generation technology. International Journal of Engineering and
Advanced Technology, 2 (3), 1-17.
Ebbinghaus, H. (1908). Psychology: An elementary textbook. New York: Arno Press.
Flogie, A., Barle Lakota, A., & Aberšek, B. (2018). The psychosocial and cognitive influence of ICT on competences on STEM Stu-
dents. Journal of Baltic Science Education, 17 (1), 267-276.
Flogie, A., & Aberšek, B. (2015). Transdisciplinary approach of science, technology, engineering and mathematics education.
Journal of Baltic Science Education, 14(6), 779–790.
448
(P. 435-449)
Gannon, A. J., & Von Backström, T. W. (2000). Solar chimney cycle analysis with system loss and solar collector performance.
Journal of Solar Energy Engineering, 122 (3), 133–137.
Holman, J. P. (1992). Heat transfer. 7th edn. New York: McGraw-Hill.
Hussain, A., Azeem, M., & Shakoor , A. (2011). Physics teaching methods: Scientific inquiry vs traditional lecture. International
Journal of Humanities and Social Science, 1 (19). 269-276.
Key Competences for Lifelong Learning (2006). European Reference Framework. Luxemburg: Office for Official Publications of
the European Communities.
Kobayashi, I. (2003). 20 ključev [20 keys]. Ljubljana: Lisac-Lisac.
Mayer, R. E. (2010). The nature of learning, using research to inspire practice. Paris: OECD.
Microsoft Partners in Learning (2010). Bringing a 1-to-1 Program to Life, a Handbook for Senior Secondary School Teachers. Wash-
ington: Microsoft Corporation
Pretorius, J. P. (2007). Optimization and control of a large-scale solar chimney power plant. (Dissertation). Retrieved from https://
www.crses.sun.ac.za/files/research/completed-research/solar/jp_pretorius.pdf.
Ostrander, S., & Schroeder, L. (2012). Superlearning 2000: New triple fast ways you can learn, earn, and succeed in the 21. century.
New York: Random House Publishing.
Schacter, D. L. (2001). The seven sins of memory: How the mind forgets and remembers. Boston: Houghton Mifflin
White, F. M. (1999). Fluid mechanics. Singapure: McGraw-Hill.
Wikström, N. (2007). Alternative assessment in primary years of international baccalaureate education. Stockholm: The Stockholm
Institute of Education.
Žbona, T. (2016). Navidezna resničnost kot proces realne izkušnje pri likovni umetnosti [Virtual reality as a process of real experi-
ence (Augmented Reality) in art education], AR: Arhitektura, raziskave, 2016(2), 6-15.
Mitja Slavinec PhD, Professor, University of Maribor, Faculty of Natural Science

and Mathematics, Koroška 160, 2000 Maribor, Slovenia.
E-mail: mitja.slavinec@um.si
Boris Aberšek PhD, Professor, University of Maribor, Faculty of Natural Science
(Corresponding author) and mathematics, Koroška 160, 2000 Maribor, Slovenia.
E-mail: boris.abersek@um.si
Dino Gačević Student, University of Maribor, Faculty of Natural Science and

Mathematics, Koroška 160, 2000 Maribor, Slovenia.
E-mail: dino.gacevic@gmail.com
Andrej Flogie PhD, Assistant Professor, University of Maribor, Faculty of Natural
Science and Mathematics, Koroška 160, and Institute Anton Martin
Slomšek, Vrbanska 30, 2000 Maribor, Slovenia.
E-mail: andrej.flogie@z-ams.si
449
THE EFFECT OF USERS’
BEHAVIORAL INTENTION ON
GAMIFICATION AUGMENTED
REALITY IN STEM (GAR-STEM)
EDUCATION
Abstract. As people attach increasing im- Chung-Ho Su

portance to the learning environment and
design of teaching materials in STEM, how
to design teaching materials that enable
students to seamlessly integrate the learn-
ing of scientific investigation, engineering Introduction
design, mathematical reasoning, and
technical skills has become increasingly With the rapid development of digital media in recent years, many
important. The purpose of this research textbooks have been equipped with the features of entertainment-oriented
media. The gamification design idea has been integrated with new scientific
was to explore the behavioral intention
technologies to facilitate teaching. In this way, the application of games in
of GAR-STEM teaching application users.
digital learning has been increased and the creative application of learning
The results showed that the best predic- materials has been restarted. In particular, the application of augmented
tor of practicability, entertainment, and reality gamification in STEM education has been reinforced with each pass-
media interactivity for intention towards ing day. As a combination of science (S), technology (T), engineering (E), and
GAR-STEM app design was usage attitude mathematics (M), STEM covers the experiences in daily life and applications
in the workplace. All these subjects (STEM) are interlinked and connected
(R2= 0.547), followed by intent to use
with the complicated problems facing individuals every day (Talley, 2016).
(R2 = 0.528), practicability (R2 = 0.408), and
The integrated knowledge education of STEM offers better solutions to tackle
entertainment (R2 = 0.186). The proposed the challenges in daily life and in the workplace (Webb, 2013). Therefore, it
model explained 52.8 % of the variance in is necessary for students in grades K-12 and college students to understand
behavioral intention. The overall findings the concepts of STEM and develop the ability to integrate and apply these
suggested that usage attitude and intent subjects so that they will have more chances to put STEM knowledge into
practice (Bybee, 2013; English, 2017). Science, technology, engineering, and
to use may augment its function as a key
mathematics are mutually independent, and the core concepts of each sub-
factor for the procedural intention toward
ject provide a solid foundation for further studies. As people pay increasing
GAR-STEM app design. attention to the application of STEM in K-12 and college education, inter-
Keywords: augmented reality, gamifica- disciplinary integration can help students gain a better understanding of
tion apps, importance-performance matrix subject concepts and transfer knowledge from one subject to another (Bell,
analysis, STEM education. 2016). According to the results of relevant studies, the combination of creative
and advanced technologies and teaching can enhance the STEM learning
participation of learners and improve their learning attitude towards STEM.
Moreover, a simulated learning environment can strengthen the cognition
and motivation of learning and make learning more effective (Kim et al.,
2015; Stull, Fiorella, Gainer, & Mayer, 2018). The integration of simulated STEM
learning environments is very important. Therefore, how to help students
Chung-Ho Su understand scientific concepts and solve interdisciplinary problems through
Shu-Te University, Taiwan creatively-simulated and seamless experience-based learning environments
is an essential topic and one of the challenges facing education researchers
(Katehi, Pearson, & Feder, 2009).
450
ISSN 1648–3898 /Print/ THE EFFECT OF USERS’ BEHAVIORAL INTENTION ON GAMIFICATION AUGMENTED REALITY IN
ISSN 2538–7138 /Online/ (P. 450-465)
Augmented reality (AR) is an information and communication technology with scientific applications and
has become a hot issue in scientific and education studies (Akçayır & Akçayır, 2017). With the help of scientific
technologies, AR incorporates virtual 2D or 3D computer simulations into a real environment (Lee, 2012). Since its
introduction, the technology has been widely applied in education. For instance, it has been used in the teaching
of the tissues and organs inside the human body and that of earth science. The knowledge of different subjects
has been incorporated into AR recognition labels on textbooks. The scanning of these labels leads to the display of
3D objects. Moreover, AR has been utilized to explore the models and performance of learning behavior, and it has
been shown that AR has the potential of being applied to teaching (Azuma et al., 2001; Lee, 2012). AR-supported
teaching features sensory organ immersion, simple operation, and a low cost, which together contribute to a greater
learning effect (Cheng & Tsai, 2013). Relevant researches showed that the application of AR to science, technology,
engineering, and mathematics (STEM) has potential educational value, including spatial understanding, technol-
ogy application, problem-solving ability, concept understanding, and scientific exploration and learning (Cheng
& Tsai, 2013; Wu, Lee, Chang, & Liang, 2013). Especially, AR is most welcomed when it is applied to such fields as
the humanities, art, and science (Liu & Tsai, 2013; Martín-Gutiérrez et al., 2010; Squire & Jan, 2007). A previous re-
search found that when AR-supporting teaching was applied in a physics lab for college students, it enhanced the
learning efficacy of the students and equipped them with a positive attitude towards their work (Akçayır, Pektas,
& Ocak, 2016). Some studies have revealed that AR has many advantages in an educational setting (Cheng & Tsai,
2013). For example, AR helps students with the exploration of the real world (Dede, 2009). By combining virtual
elements with real objects, AR enables individuals to observe items that cannot be observed with the naked eye
(Wu, Lee, Chang, & Liang, 2013). Hence, it can strengthen students’ learning motivation and help them develop a
stronger ability to explore things (Sotiriou & Bogner, 2008).
Gamification is defined as the usage of game mechanisms in daily activities. In 2010, there was a new trend
of gamification, in which game-based projects and game features were used in ways unrelated to a game (Radoff,
2012; Smith, 2012; Wu, 2012). Gamification refers to game elements being applied to a non-game environment
(Deterding, Dixon, Khaled, & Nacke, 2011). Aside from enriching the joyous experience in services and increasing
the intention to use, it can create interaction among groups (Hamari & Koivisto, 2015; Huotari & Hamari, 2017;
Liu, Santhanam, & Webster, 2017) and promote usability, productivity, and satisfaction (Saha, Manna, & Geetha,
2012; Rajanen & Rajanen, 2017), thus pushing forward publicity and generating a positive effect on performance
(Rodrigues, Costa, & Oliveira, 2014). In recent years, gamification has been increasingly applied to various subjects,
including education (Simões, Redondo, & Vilas, 2013; Kim, Song, Lockee, & Burton, 2018), environmental and ecologi-
cal behavior (Prestopnik & Tang, 2015), health and medicine (Fleming et al., 2017), creative business and marketing
(Roth, Schneckenberg, & Tsai, 2015), software development and machine learning (Chow & Huang, 2017), politics
(Santos et al., 2015), tourism (Saoud & Jung, 2018), and energy popularization and application (Nicholson, 2012).
Gamification is designed to adapt enjoyable and interesting game elements to the activities being publicized.
Applying these elements to teaching strengthens learning motivation (Su & Cheng, 2015; Su, 2017; Su, 2018).
The application of gamification to social activities is a highly pronounced category, especially in social network
websites where people share their daily lives. Social games and gamification are also common in other fields, such
as measurements of user loyalty, achievement (such as scores, virtual currencies, and levels) and employment
(Zichermann, 2011). Therefore, the gamification of education refers to game elements being applied to learning
environments, and it is often supported by ICT (information and communication technology).
Relation between Media Interactivity, Entertainment, and Practicability
The Media Richness Theory (Daft & Lengel, 1986) refers to the capability to carry information in media. Since
different types of media load various information capabilities, it causes differences in how users select media. Daft,
Lengel and Trevino (1987) divided media richness into the concepts of: (1) instant feedback; (2) multiple cues; (3)
personal focus; and (4) language variety. Rice (1993) indicated that different types of media present distinct char-
acteristics and result in various communication effects because of different objectives. Dennis and Kinney (1998)
mentioned that using computers as the communication media could provide instantaneous feedback, deliver
multiple cues, communicate with diverse languages, and focalize individuals. Wang, Hsieh, and Song (2012) pointed
out the positive effect of users’ perceived media richness on the perceived entertainment. Sullivan (1999) explained
that interactivity is the dialogue between information systems and users; in addition to the interaction, informa-
tion system users can respond to the information and content offered by the information systems in the process.
451
(P. 450-465) ISSN 2538–7138 /Online/
Previous literature (Coyle & Thorson, 2001; Lan & Sie, 2010; Lu et al., 2014; Peltokorpi, 2015; Zhao, Hen, & Wang,
2016) regarded the positive effect of interactivity on the practicability of information systems and the assistance in
better consistency between attitude towards use and use behavior. Thus, the following hypotheses were proposed:
H1: Media interactivity presents a positive effect on entertainment.
H2: Media interactivity shows a positive effect on attitude towards use.
H3: Media interactivity reveals a positive effect on practicability.
Relation between Entertainment Attitude and Behavioral Intention
The Theory of Planned Behavior states that behavior is controlled by behavioral intention and that other
possible factors might indirectly influence behavior through behavioral intention, which refers to an individual’s
intention to engage in certain behavior. Behavioral attitude, referring to the perceived positive or negative evalu-
ations of an individual executing certain behavior, is evaluated by behavior belief and results. An individual with
a more positive attitude towards a behavior will present higher behavioral intention, and vice versa. Subjective
norms refer to an influential or important person’s opinions about specific behavior, which are determined by
the beliefs and motivation of norms. Ajzen and Fishbein (1977) argued that attitudes and subjective norms are
the major factors in behavioral intention. From past research, The Theory of Planned Behavior can be applied to
explain different types of behavior. In the process of enjoyment, elements related to imagination, feelings, fun, and
symbol meanings often appear in entertainment (Hirschman & Holbrook, 1982), which could be festive or ludic
perceptions (Sherry, 1990). The characteristics of entertainment have been applied to research on behavior (Lee,
2009; Maes et al., 2014; Scarpi, 2005; Sherry, 1990; Wu & Chen, 2005) and present subjective and individual contents
(Babin, Darden, & Griffin, 1994). From the aspect of using a system, users perceiving the higher “entertainment” of
a decision system will show a higher intention of use (Gopal, Bostrom, & Chin, 1992). Accordingly, the following
hypotheses were proposed in this research:
H4: Entertainment reveals a positive effect on attitude.
H5: Entertainment presents a positive effect on behavioral intention.
Relation between Practicability, Attitude, and Behavioral Intention
The Technology Acceptance Model proposed by Davis (1989) is an extension of the Theory of Reasoned Action,
which aims to predict and understand human behavior. It is considered that individual behavior is determined by
behavioral intention, while behavioral intention is possibly affected by personal attitudes and subjective norms. Ac-
cording to Davis’ (1989) Technology Acceptance Model, user attitudes towards a system are influenced by perceived
ease of use and perceived usefulness, users’ behavioral intention is affected by attitude and perceived usefulness,
and perceived usefulness affects the perceived ease of use. This model has been broadly applied to various studies
on technology acceptance behavior, and the hypotheses in the model have been verified numerous times (Ahn,
Ryu, & Han, 2007; Al-Gahtani, 2016; Mohammadi, 2015; Moon & Kim, 2001; Sánchez & Hueros, 2010; Venkatesh &
Bala, 2008). The following hypotheses in the Technology Acceptance Model were therefore included in this research:
H6: Practicability presents a positive effect on attitude.
H7: Practicability shows a positive effect on behavioral intention.
H8: Attitude towards use reveals a positive effect on behavioral intention.
Therefore, to fill the gap left by the literature, this research adopted gamification augmented reality (GAR-
STEM) and the Media Richness Theory to explore usage behavior. In addition, IPMA was utilized to analyze the
features of GAR-STEM design, so as to find out its advantages and limitations. Matters worthy of attention and
advice improving the intention to use GAR-STEM in the future were also included. This research proposed the fol-
lowing three research goals:
1) To integrate the explanation power of the GAR-STEM app user intention model;
2) To understand GAR-STEM app user intention;
3) To integrate the analysis strategies of Importance-Performance Map Analysis (IPMA) for the matrix
analysis of the importance and performance of GAR-STEM app user intention in order to offer GAR-
STEM apps designers with better strategic design thinking.
452
ISSN 2538–7138 /Online/ (P. 450-465)
Partial Least Squares (PLS) is a form of component-based Structural Equation Modeling (SEM) that can deal with
smaller samples (30-100) and predict non-normal data with more variables, as well as deal with reflective indicators
and formative indicators, thus making it suitable for developing theories (Chin & Newsted, 1999; Gefen, Straub, &
Boudreau, 2000). Furthermore, PLS can overcome multicollinearity problems, effectively deal with moderating data
and missing values, and present favorable prediction and explanation capabilities. PLS was adopted in this research
because the retrieved data did not completely conform to a normal distribution and because only 177 samples
were retrieved. In addition, the research issue was about theoretical exploration and discovery. As the number of
research samples was small, analysis with PLS presented favorable prediction and explanation capabilities. Such
a small sample would also not be affected by the number of samples and the distribution of variables. For the
stable estimation of variables, 500 iterations of bootstrap resampling has been suggested for testing (Chin, 1998).
Research Structure
The integration of practicability, entertainment, and media interactivity was proposed to construct the mea-
surement model for the intention of using GAR-STEM apps. After establishing the hypotheses, the positive, direct,
and significant effect of media interactivity on entertainment, attitude towards usage, and behavioral intention
was as shown in Figure 1.
Figure 1. Model of behavioral intention to use GAR-STEM apps.
Operational Definition of Constructs
In order to test the above research hypotheses, a conceptual model was utilized, and the corresponding
questions were constructed in this research. According to the conceptual model, the research variables in this
research included media interactivity, entertainment, attitude towards use, practicability, and behavioral intention.
A structured questionnaire was presented to the participants and was scored using a Likert 7-point scale (with an-
swers ranging from extremely disagree to extremely agree) to collect relevant data. The results were processed and
analyzed using SPSS, and PLS was utilized to analyze the research model. The five latent variables in this research
were reflective indicators used for analyzing the relevance between variables and indicators. The definitions and
measurement of the variables, including the five dimensions of media interactivity, entertainment, practicability,
attitude towards use, and behavioral intention, as well as 16 questions, are listed in Table 1.
453
(P. 450-465) ISSN 2538–7138 /Online/
Table 1. Operational definitions of constructs.
Reference
Variable Operational definition Questionnaire item
The visual design of the app is attractive.

Overall speaking, the design of the app’s user interface is
Interactivity between app favorable.
Media interactivity users and the app in the usage (Coyle & Thorson, 2001)
process (MI) Good interactivity with other users on the app.
Being able to establish friendships with other users through
the app.
The app offers several funny and interesting functions.
Perceived entertainment of
(Csikszentmihalyi, 1975;
Entertainment app users about the functions I am pleased at using the app’s functions.
Moon & Kim, 2001)
offered by the app (EN)
I perceive fun in using the app.
The app offers the contents I desire.
App users’ perception of the Overall speaking, the app is useful. (Chen, Clifford, & Wells,
Practicability information or services offered 2002; Chen & Wells,
by the app (PR) The operation of the app is simple. 1999; Ducoffe, 1996)
The operation of the app is clear and understandable.
App users’ perceived preference Overall speaking, I like to use the app.
Attitude towards use (Davis, 1989)
for the app (ATT) I have a positive evaluation about the app.
I will continuously use the app.
App users’ evaluation of the
I will still choose to use the app even though there are other
Behavioral intention intention and possibility to (Davis, 1989)
similar ones.
continuously use the app (BI)
I will recommend the app to others.
Sample and Scale Analysis
(1) Formal questionnaire respondents

In the formal test, in consideration of the impact of regional disparities, a two-step sampling method was
applied in this research, namely, the stratified cluster sampling method and quota sampling. In step 1, the strati-
fied cluster sampling method was applied based on school size in consideration of regional disparities, and four
schools were selected as samples. In step 2, quota sampling was applied, and 186 samples were selected from the
sample schools by considering different sampling percentages for different grades and dispatch questionnaires
were dispatched. A total of 177 valid questionnaires were recovered, resulting in an availability of 95.1%. As the
late responses could be similar to those who did not respond, the early response (n = 106) and late response (n =
71) samples were tested for nonresponse bias (Armstrong & Overton, 1977). All measured variables and control
variables of the two groups of samples were compared with a t-test to judge the possible bias of the late responses.
The result showed no significant difference in the responses between the first and the second stage, revealing
insignificant nonresponse bias in this research.
(2) Test of Common Method Variance

Common Method Variance (CMV) refers to the man-made covariant relevance between independent vari-
ables and dependent variables resulted from using the same data sources, participants, measuring environments,
scale semantics, and scale characteristics. CMV can be used to test if a single factor is able to explain more than
50% of the variance of all variables in the initial solution during factor analysis (Podsakoff, 2003). For this reason,
Harman’s One-Factor Test was applied to test for CMV among the research variables in this research and factor
analysis was used to analyze the factors extracted from the initial solution in order to test for Common Method
Variance. A total of nine factors with eigenvalues larger than 1 were extracted using confirmatory factor analysis.
The explained variance of the first factor was 24.25% (maximum), which was less than 50%, and the accumulated
explained variance of the nine factors was 83.22%. Accordingly, there was no obvious common source bias in this
research (Saraf, Langdon, & Gosain, 2007).
454
ISSN 2538–7138 /Online/ (P. 450-465)
Research Results
In this section, the research results are presented in three stages. The reliability and validity of the measurement
model were analyzed during the first stage, the path coefficient test of the structural model and the estimation of
the model prediction indicator were analyzed during the second stage, and the Importance-Performance Matrix
Analysis (IPMA) of the structural model was performed during the third stage. The detailed steps and explanations
are described in the following sections.
Reliability and Validity Analysis of the Measurement Model
According to Fornell and Larcker’s (1981) principles of convergent validity, the test of convergent validity
of the designed questions needed to satisfy the following conditions. First, the factor loading (λ) should be sig-
nificant and higher than 0.5 (Chin, 1998). Second, the composite reliability (CR) should be larger than 0.6 (Chin,
1998). Third, the discriminant validity-average variance extracted (AVE) should be larger than 0.5 (Chin, 1998) and
the cross-loading matrix (Chin, 1998). The reliability analysis of the dimensions in this research was organized as
shown in Table 2. The factor loadings showed a significance of p=.001 and all questions were higher than 0.5, the
composite reliability was 0.80-0.96 and larger than 0.8, and the average variance extracted was between 0.67 and
0.85 and above 0.5, thus satisfying the above conditions. The measurement model showed that the factor loadings
of all questions were 0.75-0.96 and the p value reached the significance level of 0.05, thus indicating convergent
validity (Chin, 1998). The estimation of discriminant validity was tested from two directions (Henseler, Ringle, &
Sinkovics, 2009). One was the cross-loading matrix, which indicated that the individual question loadings of the
variables in this research were higher than the loadings of other variables (Fornell & Larcker, 1981). The other was
the Fornell–Larcker criterion; that is, the AVE of each variable needed to be larger than the square of the correlation
coefficient between paired variables to show the discriminant validity of the research variables (Fornell & Larcker,
1981; Segars & Grover, 1998). Table 2 presents the average variance extracted square roots of all variables, which
were larger than the correlation value between variables. Table 3 shows that the individual question loadings of
the variables were higher than those of other variables. Accordingly, the variables in this research had acceptable
reliability and validity.
Table 2. Reliability and validity analysis.
Mean SD Factor loading CR Cronbach’s α AVE BI EN ATT PR MI
BI 5.21 1.1 0.92 0.96 0.94 0.85 0.92
EN 4.83 1.3 0.87 0.94 0.92 0.75 0.61 0.87
ATT 4.61 1.2 0.73 0.80 0.79 0.69 0.63 0.54 0.83
PR 4.93 1.1 0.75 0.81 0.70 0.73 0.54 0.36 0.65 0.85
MI 4.74 1.4 0.87 0.86 0.77 0.67 0.48 0.43 0.64 0.64 0.82
Note: SD= Standard deviation; CR= Composite reliability
Table 3. Factor loading analysis of the variables.
BI EN ATT PR MI
BI 1 0.95 0.56 0.56 0.50 0.43
BI 2 0.95 0.60 0.60 0.51 0.43
BI 3 0.91 0.59 0.57 0.53 0.46
ATT1 0.45 0.51 0.76 0.43 0.43
ATT2 0.50 0.49 0.80 0.37 0.38
EN2 0.50 0.92 0.50 0.24 0.41
455
(P. 450-465) ISSN 2538–7138 /Online/
BI EN ATT PR MI
EN3 0.44 0.89 0.42 0.30 0.32
EN4 0.68 0.87 0.51 0.44 0.45
PR1 0.23 -0.01 0.38 0.69 0.22
PR2 0.23 0.09 0.26 0.70 0.14
PR3 0.44 0.32 0.45 0.76 0.62
PR4 0.43 0.37 0.53 0.71 0.49
MI1 0.26 0.14 0.35 0.37 0.75
MI2 0.53 0.53 0.63 0.70 0.86
MI3 0.30 0.27 0.51 0.39 0.84
MI4 0.33 0.29 0.58 0.51 0.86
Path Coefficient Test of the Structural Model and Prediction Indicator Estimation Model
PLS was used in this research to analyze the structural model for the correlation strength and direction between
research variables. If significance appeared after the test, the hypotheses would be supported as expected. PLS can
be used to analyze and explain the percentage of the variance of exogenous variables to endogenous variables
and to test the prediction capability of a research model. The structural model was evaluated based on six steps
proposed by researchers, including: Step 1 - evaluating the multicollinearity of the structural model (VIF); Step 2 -
setting the significance of the path coefficient as the standard; Step 3 - the test of and the path coefficient; Step
4 - evaluation of the effect size; Step 5 - evaluation of the predictive relevance ( ); and Step 6 - the goodness-
of-fit index (Hair, Ringle, & Sarstedt, 2011; Reinartz, Haenlein, & Henseler, 2009; Rigdon, 2012; Tenenhaus, Amato,
& Esposito Vinzi, 2004). First, the variance inflation factor (VIF) was utilized to evaluate the multicollinearity. The
result showed the maximal VIF of 1.32 for all variables, which conformed to a VIF between 0.2 and 5 as suggested
by Neter and Wasserman (Hair et al., 2011) and indicated there was no multicollinearity problem.
The structural model was further analyzed using PLS to list the variance explained ( ), standardized path
coefficient (β), and t value. and the path coefficient are the major indicators used to judge a model (Chin,
1998). The analysis of the structural model is shown in Figure 2. A complete analysis of the model was performed
in this research. The path coefficient between media interactivity and entertainment was 0.432 (t=4.164) and had
high significance, thus H1 was supported. That is, the higher media interactivity with GAR-STEM apps showed the
higher entertainment of the apps. The path coefficient between app media interactivity and attitude towards use
was 0.284 (t=2.961) and had high significance, thus H2 was partially supported. The stronger media interactivity
with GAR-STEM apps showed a higher usage attitude towards using the apps. The path coefficient between app
media interactivity and practicability was 0.638 (t=9.045) and had high significance, thus H3 was supported, reveal-
ing stronger app media interactivity and higher practicability. The path coefficient between entertainment and
attitude towards use was 0.293 (t=2.024) and had high significance, thus H4 was supported, indicating that the
higher the entertainment of GAR-STEM apps, the better the attitude towards use. The path coefficient between
practicability and usage attitude was 0.360 (t=3.374) and had high significance, thus indicating that H5 was sup-
ported. The result also showed a higher practicability of the app and better usage attitude. The path coefficient
between entertainment and players’ behavioral intention was 0.379 (t=2.949) and had high significance, thus H6
was supported. That is, higher entertainment will enhance players’ behavioral intention to use GAR-STEM apps.
The path coefficient between players’ attitude towards use and behavioral intention was 0.275 (t=2.459) and had
high significance, thus H7 was supported. In other words, players with a better attitude towards using an app will
present higher behavioral intention to use the app. Finally, the path coefficient between the practicability of an app
and players’ behavioral intention was 0.228 (t=1.692) and had high significance, thus H8 was supported, indicating
the higher the practicability of an app, the higher the behavioral intention of the players.
456
ISSN 2538–7138 /Online/ (P. 450-465)
Generally speaking, the of the endogenous construct in a model can be evaluated as the overall goodness-
of-fit in PLS (Hulland & Business, 1999). As shown in Figure 2, the R2 of the four endogenous constructs were 0.186
for entertainment, 0.574 for attitude toward use, 0.408 for practicability, and 0.528 for behavioral intention. Attitude
toward use and behavioral intention showed a high R2, representing a favorable explanation among behavioral
intention, the four antecedents, and a small unexplained residual. The result indicated a favorable goodness-of-fit
between behavioral intention and the antecedents. The R2 of entertainment was slightly lower, possibly because
the practicability and usefulness of an app are emphasized in GAR-STEM apps. Entertainment, therefore, is not a
major component of GAR-STEM apps. In sum, the four endogenous constructs showed a favorable explanation,
revealing a good explanation of the research model and conformity to the principles suggested by Cohen (1988).
Figure 2. Path coefficient of intention to use GAR-STEM apps.
In addition to evaluating the predictive power of the model, the effect size of the structural model was
another important evaluation indicator. In addition to evaluating the of all endogenous constructs by deleting
the change caused by specific exogenous variables in the model, the effect size can be used to evaluate
the effects of the deleted variables on the endogenous variables. Such a measurement is called the effect size.
The equation is:
(1)
is generally evaluated by having the values of 0.02, 0.15, and 0.35 represent the small, medium, and large
effect of the exogenous latent variables (Cohen, 1988). Table 4 shows the average of the effect size larger than
0.15 (the medium effect). In addition to , researchers should test Stone-Geisser’s value, which is an indicator
of a model’s predictive relevance, to evaluate the prediction accuracy of the correlation (Geisser, 1974; Stone,
1974). In regard to in the structural model shown in Table 4, the value of the endogenous latent constructs
larger than 0 revealed the predictive relevance between the path model and constructs. Tenenhaus et al. (2004)
and Tenenhaus et al. (2005) proposed the overall goodness-of-fit index (GoF) for PLS as “the possible program to
verify the overall model of PLS” (Tenenhaus et al., 2004). Wetzels et al. (2009) suggested using the communality
value of 0.50 (Fornell & Larcker, 1981) and the value (Cohen, 1988) to evaluate the goodness-of-fit of the PLS
structural model as GoFsmall (0.10), GoFmedium (0.25), and GoFlarge (0.36). Table 4 shows that the GoF of 0.56 in this research
was larger than 0.36, thus presenting an excellent goodness-of-fit of the structural model. The model predictive
457
(P. 450-465) ISSN 2538–7138 /Online/
indicators proposed by relative researchers and the goodness-of-fit organized in this research could be used as a
reference (Table 5) for successive researchers, including the model predictive indicators and the goodness-of-fit
checklist (Yang & Su, 2017).
Table 4. Analysis of Model criterion and IPMA.
EN PR ATT BI IPMA Model fit indicator
(D) (I) (T) (D) (I) (T) (D) (I) (T) (D) (I) (T) Performance R2 f2 Q2 GoF
MI 0.43 0.43 0.64 - 0.64 0.28 0.35 0.63 - 0.65 0.65 56.12 - - -
EN - - - - - - 0.29 - - 0.38 0.08 0.46 61.86 0.186 - 0.13
PR - - - - - - 0.36 - - 0.23 0.1 0.33 57.09 0.408 - 0.18 0.56
ATT - - - - - - - - - - - 0.28 60.55 0.574 0.19 0.26
BI - - - - - - - - - - - - - 0.528 0.21 0.43

Note: MI=Media interactive; EN=entertainment; ATT=Attitude; PR=Practicability; BI=Behavioral Intention
Table 5. Model predictive indicators and goodness-of-fit checklist.
Model
Model evaluation Criterion Suggestion value Result
value
Cronbach’s alpha (CA) Values must not be lower than 0.600 CA >0.80 Good
Composite reliability (CR) Values must not be lower than 0.600. CR >0.80 Good
Values should be significant at the .050 level and

Indicator loadings >0.70 Good
higher than 0.70
Values must higher than 0.70

Internal consistency reliability CR >0.80 Good
Reliability and Validity Composite Reliability (CR)>=0.70
& Measurement Model
Values must higher than 0.50
Convergent validity >0.5 Good
Average Variance Extracted (AVE) >= 0.500
Cross-loadings Factor Loading should higher than Cross Loading Higher Good
the square root of the AVE of each construct should be

Fornell-Larcker criterion Higher Good
higher than the construct’s highest correlation
Multicollinearity 0.2<VIF< 5 1.32 Good
Path Coefficient & 0.670 (High), 0.333(medium), 0.190(weak). 0.528 Good

Estimation Model
Path coefficients p value <0.05 <0.05 Good
0.02~0.15 (weak), 0.15~0.35 (medium), 0.19

Effect size Medium
Value higher than 0.35(strong) 0.21
Model fit indicator &
Use blindfolding, Q2 >0 indicate a predictive
Prediction Indicator Predictive relevance 0.13 Relevance
relevance.
Goodness-of-Fit Index (GoF) 0.1(small), 0.25(medium), 0.36(large) 0.56 Good
458
ISSN 2538–7138 /Online/ (P. 450-465)
Result of Importance-Performance Matrix Analysis (IPMA)
Importance-performance matrix analysis (IPMA) helps to gain additional insights by utilizing the extension
of the PLS-SEM results and provides additional dimensions (Hock et al., 2010; Kristensen et al., 2000; Slack, 1994;
Völckner et al., 2010, Su & Cheng, 2019). The analysis is established based on the estimation of the PLS-SEM path
relevance and by adding the dimension of the latent variable mean value. For the endogenous variables of the
specific target, IPMA compares the total effect (importance) of the structural model with the mean value of the
latent variable fraction (performance) to decide the priority of management (or the specific focus in the model).
More specifically, the result could point out the factors of high importance and relatively worse performance in
the model for the improvement of successive key design elements and management (Hock et al., 2010; Rigdon,
Ringle, & Sarstedt, 2010).
The calculation of importance in IPMA in this research required the total effect of all other constructs (i.e.
media interactivity, entertainment, attitude towards use, and practicability) on the intention of using the target
construct (in Figure 2) for judging the importance. The total effect between two constructs was the sum of all direct
and indirect effects in the structural model and could be represented as total effect=direct effect + indirect effect.
The construct performance was calculated with:
(2)
where stands for the i-th number (e.g. i=5 represents the latent variable fraction of the fifth observed value
in the database) of the j-th latent variable (e.g. j=2 represents the second latent variable in the structural model)
(Anderson & Fornell, 2000; Hock et al., 2010). The IPMA in Table 4 was plotted to show the correlation between the
importance and performance distribution of the intention to use GAR-STEM apps shown in Figure 3. Furthermore,
the management meaning could be explained with the acquired regional quadrant.
Figure 3. Importance and performance distribution of the intention to use GAR-STEM apps.
Quadrant I showed high importance and high performance and represented the “keep up the good work”
zone, revealing that players place more stress on entertainment with high satisfaction. In this case, designers must
maintain the original advantage and enhance the entertainment of digital cultural and creative games. This zone
therefore was the core competitiveness in the design. Quadrant II showed low importance but high performance and
represented the possible overkill zone, showing that players do not focus much on attitude towards use but present
high satisfaction. Design resources could be adjusted and transferred to other variables with higher importance
to reduce the cost. Quadrant III showed low importance and low performance and represented the low priority
zone, revealing that app users do not place high emphasis or satisfaction on practicability. Designers therefore do
not need to spend too much time on such items. Quadrant IV showed high importance but low performance and
represented the “concentrate here” zone, presenting the high emphasis but low satisfaction of GAR-STEM app users
459
(P. 450-465) ISSN 2538–7138 /Online/
on media interactivity. In this case, designers should consider improvements and adopt relative design strategies
and thinking to promote design standards for app media interactivity. Figure 3 reveals that media interactivity was
located in Quadrant IV (the “concentrate here” zone), with high importance but low performance. In other words,
the interface and user experience design need to be processed and improved, and further design strategies need
to be adopted to enhance players’ satisfaction with the media interactivity and encourage players to use the app
through a favorable interactive interface design.
The research results revealed that the eight hypotheses were supported and that the path coefficient of each
hypothesis had high significance (Table 6). After verifying the hypotheses with relevant research, the conclusion
and suggestions were proposed as follows.
Table 6. Hypothesis test result.
Path
H Path Description t-Value Results
Coefficient
H1 MI->EN Media interactivity presents a positive effect on entertainment. 0.43 4.164*** Supported
H2 MI-> ATT Media interactivity shows a positive effect on attitude towards use. 0.28 2.961*** Supported
H3 MI-> PR Media interactivity reveals a positive effect on practicability. 0.64 9.045*** Supported
H4 EN-> ATT Entertainment reveals a positive effect on attitude. 0.29 2.024** Supported
H5 EN-> BI Entertainment presents a positive effect on behavioral intention. 0.38 2.949*** Supported
H6 PR-> ATT Practicability presents a positive effect on attitude. 0.36 3.374*** Supported
H7 PR-> BI Practicability shows a positive effect on behavioral intention. 0.28 2.459** Supported
Attitude towards use reveals a positive effect on behavioral
H8 ATT-> BI 0.23 1.692* Supported
intention.
Discussion
Higher Media Interactivity, Expectation of Entertainment, Attitude towards Use, and Practicability
The results of this research reflected the research viewpoint of Wang, Hsieh, and Song (2012), who indicated
that users’ perceived media richness of information systems will positively affect the perceived entertainment.
Sullivan (1999) pointed out interactivity as the dialogue between information systems and users. In addition to
the interaction, information system users can respond to the information and content offered by information
systems in the process. Coyle & Thorson (2001) regarded the positive effect of interactivity on the practicability
of information systems and the assistance in forming consistency between better attitudes towards information
systems and use behavior. This result was also consistent with the studies of Wang, Hsieh, and Song (2012), Sullivan
(1999), and Coyle and Thorson (2001).
Higher Entertainment, Expectation of Attitude towards Use, and Behavioral Intention
The results reflected the research viewpoint of Hirschman & Holbrook (1982), who pointed out high values
for entertainment, expectation of attitude towards use, and behavioral intention. Ajzen & Fishbein (1975) claimed
that attitude and subjective norms are the primary factors in behavioral intention. Elements related to imagination,
feeling, fun, and symbolic meanings (Hirschman & Holbrook, 1982) or festive and ludic perception (Sherry, 1990)
often appear with hedonic sensations in the pleasure process. The characteristics of entertainment have often been
applied in research on behavior (Scarpi, 2005; Sherry, 1990) and used to present subjective and personal contents
((Babin, Darden, & Griffin, 1994). From the aspect of system use, users perceiving the higher hedonic sensation of
information systems will reveal higher intention of adoption (Gopal, Bostrom, & Chin, 1992). Such a result was also
consistent with the research of Scarpi (2005), Sherry (1990), and Gopal, Bostrom, and Chin (1992).
460
ISSN 2538–7138 /Online/ (P. 450-465)
Higher Practicability, Expectation of Attitude towards Use, and Behavioral Intention
Davis (1989) indicated that better system practicability will enhance the expectation of attitude towards use
and behavioral intention. Meanwhile, Davis (1989) proposed that individual behavioral performance in the Tech-
nology Acceptance Model is determined by behavioral intention, which can be affected by personal attitudes and
subjective norms. According to Davis’ Technology Acceptance Model, user attitude towards systems is influenced
by perceived ease of use and that perceived usefulness and users’ behavioral intention is affected by attitude and
perceived usefulness, which also affects perceived ease of use (Davis, 1989). Such a model has been broadly applied
to various studies on technology acceptance behavior, and the hypotheses in the model have been verified several
times (Ahn et al., 2007; Moon & Kim, 2001; Venkatesh & Bala, 2008). As a result, the research result was consistent
with Davis (1989), who indicated that system use is decided by behavioral intention, that behavioral intention is
determined by attitude toward use and perceived usefulness, and that attitude toward use is decided by perceived
usefulness and perceived ease of use.
Higher Attitude towards Use results in Higher Expectation of Behavioral Intention
The research result was consistent with the findings of Davis (1989), who mentioned that good attitude to-
wards use will result in high behavioral intention. The Technology Acceptance Model was extended from Theory
of Reasoned Action, which aims to predict and understand human behavior. It is considered that an individual’s
behavior is determined by behavioral intention, while behavioral intention might be influenced by personal attitudes
and subjective norms. According to Davis’ (1989) Technology Acceptance Model, user attitudes towards systems
will be affected by perceived ease of use and perceived usefulness, users’ behavioral intention will be influenced
by attitude and perceived usefulness, and perceived usefulness will affect perceived ease of use. Such a model has
been broadly applied to research on technology acceptance behavior and the hypotheses in the model have been
verified numerous times (Ahn et al., 2007; Moon & Kim, 2001; Venkatesh & Bala, 2008). The results were consistent
with the research of Davis (1989), Venkatesh & Bala (2008), and Moon & Kim (2001).
Importance-Performance Analysis (IPMA)
With IPMA, designers can realize the situations of players in the improvement of dimensions and coping
strategies. Improvement would be required when a factor of the intention to use GAR-STEM apps was emphasized
but presented low performance. On the contrary, when the importance of a dimension was low but showed high
performance, it might be overemphasized, and the invested resources could be reduced. Nonetheless, it would
be meaningless to focus on intention of use with low importance and consider the total effect of the path coef-
ficient of the structural model, as some the high-importance intention of use items presented good performance.
Through IPMA, it was found that players highly stress the media interactivity of GAR-STEM apps but do not show
high satisfaction. The analysis value of media interactivity appeared in Quadrant IV, revealing high importance but
low performance (the “concentrate here” zone), indicating that players highly emphasize the media interactivity
interface and user experience design of GAR-STEM apps but are not highly satisfied. In this case, designers should
take improvement into account and adopt relative design strategies and thinking to promote design standards
for app media interactivity, such as the experience design of the user interface and players’ experience design, to
allow players to invest in the situation through favorable interface design.
Conclusions
The model proposed in this research could predict the behavioral intention in the application of GAR-STEM.
In addition, the overall model evaluation scale had good research adaptability. The design of useful STEM teaching
materials could be used to present and explain concepts in an effective manner and design appropriate teaching
activities so as to enable all students to understand the contents they have learned. Good STEM teaching material
design should cover different teaching objectives for different grades, and furthermore should note the under-
standing level of students on the subject and various teaching strategies and presentation methods in teaching.
Therefore, this research put forth the following academic and practical implications:
461
(P. 450-465) ISSN 2538–7138 /Online/
Academic and Practical Meaning
(1) Academic implications: Research on cultural and creative issues is increasing. The prediction model for the
behavioral intention of GAR-STEM apps proposed in this research presents innovation and offers a reference for
successful app design. From the viewpoint of the behavioral intention to use GAR-STEM apps, app designs with
media interactivity, entertainment, and practicability are the key factors in players’ attitude and intention of use. The
research results therefore could be referenced in regard to GAR-STEM app design and behavioral intention theory.
(2) Practical implications: In respect of the teaching knowledge in STEM, teachers should be able to organize
and understand learning contents and present them to students by effective interaction means. Techniques in
presentation include the application of the most persuasive statements, examples, simulations, explanations, and
demonstrations using teaching aids to present teaching materials in a way that is acceptable to students. In addi-
tion, teaching material design through multimedia scenarios could contribute to understanding difficult subjects,
particularly science, technology, engineering and mathematics. However, teachers should also understand which
subjects students are interested in, their difficulties in learning, and how to further transfer the contents in teaching
materials into effective teaching, and they should be able to adjust teaching methods for students with different
abilities and backgrounds (for example, how to design the experience design and user experience design in rela-
tion to the media interactive user interface of GAR-STEM). It is important to take the entertainment aspect into
account. From the perspective of system application, the higher the “entertainment” of an information system a
user experiences, the stronger the adoption intention will be.
Research Restrictions and Suggestions for Successive Research
This research merely utilized scales from the literature for the measurement to understand the inadequacy
of the theoretical dimensions, and qualitative research based on expert interviews was not taken into consider-
ation. Meanwhile, the representativeness of the samples was a major restriction in the quantitative research. The
factors of the behavioral intention of GAR-STEM apps therefore were major concerns. Possible research directions
are listed below.
(1) Practical research
In addition to referring to the literatures, qualitative expert interviews can be included to discuss the key fac-
tors in the behavioral intention to use GAR-STEM apps so as to reinforce the model prediction.
(2) Extension of the research model
Theoretical points of view were adopted in this research; however, only subjective behavioral intention and
inferences were discussed. Regarding the factors of the behavioral intention to use GAR-STEM apps, it is suggested
to include expert interviews to make up for the inadequacy of the quantitative research. It is also suggested to
extend the research model by adding learning style and involvement as moderators in order to observe the mod-
eration effect of involvement on intention of use. In this case, limited resources could be invested in moderators
to achieve maximal efficiency.
References
Ahn, T., Ryu, S., & Han, I. (2007). The impact of Web quality and playfulness on user acceptance of online retailing. Information
& Management, 44(3), 263-275.
Ajzen, I., & Fishbein, M. (1975). Belief, attitude, intention and behavior: An introduction to theory and research: Reading, MA:
Addison-Wesley.
Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological
Bulletin, 84(5), 888.
Akçayır, M., & Akçayır, G. (2017). Advantages and challenges associated with augmented reality for education: A systematic
review of the literature. Educational Research Review, 20, 1–11.
Akçayır, M., Akçayır, G., Pektas, H. M., & Ocak, M. A. (2016). Augmented reality in science laboratories: The effects of augmented
reality on university students’ laboratory skills and attitudes toward science laboratories. Computers in Human Behavior,
57, 334-342.
Al-Gahtani, S. S. (2016). Empirical investigation of e-learning acceptance and assimilation: A structural equation model. Applied
Computing and Informatics, 12(1), 27-50.
Anderson, E. W., & Fornell, C. (2000). Foundations of the American customer satisfaction index. Total Quality Management, 11(7),
869-882.
462
ISSN 2538–7138 /Online/ (P. 450-465)
Armstrong, J. S., & Overton, T. S. (1977). Estimating nonresponse bias in mail surveys. Journal of Marketing Research, 14(3), 396-402.
Azuma, R., Baillot, Y., Behringer, R., Feiner, S., Julier, S., & MacIntyre, B. (2001). Recent advances in augmented reality. Computer
Graphics and Applications IEEE, 21(6), 34e47.
Babin, B. J., Darden, W. R., & Griffin, M. (1994). Work and/or fun: measuring hedonic and utilitarian shopping value. Journal of
Consumer Research, 20(4), 644-656.
Bell, D. (2016). The reality of STEM education, design and technology teachers’ perceptions: A phenomenographic study. Inter-
national Journal of Technology and Design Education, 26(1), 61–79.
Bybee, R. W. (2013). The case for STEM education: Challenges and Opportunities. NSTA press.
Chen, Q., & Wells, W. D. (1999). Attitude toward the site. Journal of Advertising Research, 39(5), 27-38.
Chen, Q., Clifford, S. J., & Wells, W. D. (2002). Attitude toward the site II: New information. Journal of Advertising Research, 42(2), 33-45.
Cheng, K. H., & Tsai, C. C. (2013). Affordances of augmented reality in science learning: Suggestions for future research. Journal
of Science Education and Technology, 22(4), 449-462.
Chin, W. W. (1998). The partial least squares approach to structural equation modeling. Modern Methods for Business Research,
295(2), 295-336.
Chin, W. W., & Newsted, P. R. (1999). Structural equation modeling analysis with small samples using partial least squares. Statisti-
cal Strategies for Small Sample Research, 1(1), 307-341.
Chow, I., & Huang, L. (2017). A software gamification model for cross-cultural software development teams. In Proceedings of
the 2017 International Conference on Management Engineering, Software Engineering and Service Sciences (pp. 1-8). ACM.
Cohen, J. (1988). Statistical power analysis for the behavior science. (2nd). New Jersey: Laurence Erlbaum Associates, Publishers,
Hillsdale.
Coyle, J. R., & Thorson, E. (2001). The effects of progressive levels of interactivity and vividness in web marketing sites. Journal
of Advertising, 30(3), 65-77.
Csikszentmihalyi, M., & Csikszentmihalyi, I. (1975). Beyond boredom and anxiety (Vol. 721). San Francisco: Jossey-Bass.
Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management
Science, 32(5), 554-571.
Daft, R. L., Lengel, R. H., & Trevino, L. K. (1987). Message equivocality, media selection, and manager performance: Implications
for information systems. MIS Quarterly: Management Information Systems, 11(3), 355-366.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly:
Management Information Systems, 13(3), 319-340.
Dede, C. (2009). Immersive interfaces for engagement and learning. Science, 323(5910), 66-69.
Dennis, A. R., & Kinney, S. T. (1998). Testing media richness theory in the new media: The effects of cues, feedback, and task
equivocality. Information Systems Research, 9(3), 256-274.
Deterding, S., Khaled, R., Nacke, L. E., & Dixon, D. (2011). Gamification: Toward a definition. In CHI 2011 Gamification Workshop
Proceedings (Vol. 12). Vancouver BC, Canada.
Ducoffe, R. H. (1996). Advertising value and advertising on the web. Journal of Advertising Research, 36(5), 21-21.
English, L. (2017). STEM education K-12: Perspectives on integration. International Journal of STEM Education, 3(1), 3.
Fleming, T. M., Bavin, L., Stasiak, K., Hermansson-Webb, E., Merry, S. N., Cheek, C., et al. (2017). Serious games and gamification
for mental health: Current status and promising directions. Frontiers in Psychiatry, 7, 215.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error.
Journal of Marketing Research, 18(1), 39-50.
Gefen, D., Straub, D., & Boudreau, M.-C. (2000). Structural equation modeling and regression: Guidelines for research practice.
Communications of the Association for Information Systems, 4(1), 7.
Geisser, S. (1974). A predictive approach to the random effect model. Biometrika, 61(1), 101-107.
Gopal, A., Bostrom, R. P., & Chin, W. W. (1992). Applying adaptive structuration theory to investigate the process of group support
systems use. Journal of Management Information Systems, 9(3), 45-69.
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing theory and Practice, 19(2), 139-152.
Hamari, J., & Koivisto, J. (2015). “Working out for likes”: An empirical study on social influence in exercise gamification. Computers
in Human Behavior, 50, 333-347.
Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares path modeling in international marketing.
Advances in International Marketing, 20(1), 277-319.
Hirschman, E. C., & Holbrook, M. B. (1982). Consumption: Emerging Concepts. Journal of Marketing, 46(3), 92-101.
Hock, C., Ringle, C. M., & Sarstedt, M. (2010). Management of multi-purpose stadiums: Importance and performance measure-
ment of service interfaces. International Journal of Services Technology and Management, 14(2-3), 188-207.
Hsu, S.-H., Chen, W.-H., & Hsueh, J.-T. (2006). Application of customer satisfaction study to derive customer knowledge. Total
Quality Management and Business Excellence, 17(04), 439-454.
Hulland, J., & Business, R. I. S. o. (1999). Use of partial least squares (PLS) in strategic management research: A review of four
recent studies. Strategic Management Journal, 20(2), 195-204.
Huotari, K., & Hamari, J. (2017). A definition for gamification: Anchoring gamification in the service marketing literature. Electronic
Markets, 27(1), 21–31.
Jin, C.-H. (2014). Adoption of e-book among college students: The perspective of an integrated TAM. Computers in Human
Behavior, 41, 471-477.
Katehi, L., Pearson, G., & Feder, M. (2009). The status and nature of K-12 engineering education in the United States. The Bridge,
39(3), 5–10.
463
(P. 450-465) ISSN 2538–7138 /Online/
Kim, C., Kim, D., Yuan, J., Hill, R. B., Doshi, P., & Thai, C. N. (2015). Robotics to promote elementary education pre-service teachers’
STEM engagement, learning, and teaching. Computers & Education, 91, 14–31.
Kim, S., Song, K., Lockee, B., & Burton, J. (2018). Students’ perception of gamification in learning and education. In Gamification
in Learning and Education (pp. 49–57). Springer, Cham.
Kristensen, K., Martensen, A., & Gronholdt, L. (2000). Customer satisfaction measurement at post Denmark: results of application
of the European customer satisfaction index methodology. Total Quality Management, 11(7), 1007-1015.
Lan, Y.-F., & Sie, Y.-S. (2010). Using RSS to support mobile learning based on media richness theory. Computers & Education, 55(2),
723-732.
Lee, K. (2012). Augmented reality in education and training. TechTrends, 56(2), 13-21.
Lee, M.-C. (2009). Factors influencing the adoption of internet banking: An integration of TAM and TPB with perceived risk and
perceived benefit. Electronic Commerce Research and Applications, 8(3), 130-141.
Lin, T., Duh, H., Wang, H., & Tsai, C. (2013). An investigation of learners’ collaborative knowledge construction performances and
behavior patterns in an augmented reality simulation system. Computers and Education, 68, 314-321.
Liu, D., Santhanam, R., & Webster, J. (2017). Toward Meaningful Engagement: A Framework for Design and Research of Gamified
Information Systems. MIS Quarterly, 41(4).
Liu, P. H. E., & Tsai, M. K. (2013). Using augmented-reality-based mobile learning material in EFL English composition: An explora-
tory case study. British Journal of Educational Technology, 44(1), E1-E4.
Lu, Y., Kim, Y., Dou, X. Y., & Kumar, S. (2014). Promote physical activity among college students: using media richness and inter-
activity in web design. Computers in Human Behavior, 41, 40-50.
Maes, J., Leroy, H., & Sels, L. (2014). Gender differences in entrepreneurial intentions: A TPB multi-group analysis at factor and
indicator level. European Management Journal, 32(5), 784-794.
Martín-Gutiérrez, J., Saorín, J. L., Contero, M., Alcañiz, M., Pérez-López, D. C., & Ortega, M. (2010). Design and validation of an
augmented book for spatial abilities development in engineering students. Computers & Graphics, 34(1), 77–91.
Mohammadi, H. (2015). RETRACTED: Factors affecting the e-learning outcomes: An integration of TAM and IS success model.
Telematics and Informatics, 32(4), 701-719.
Moon, J.-W., & Kim, Y.-G. (2001). Extending the TAM for a World-Wide-Web context. Information & Management, 38(4), 217-230.
Morschheuser, B., Hassan, L., Werder, K., & Hamari, J. (2018). How to design gamification? A method for engineering gamified
software. Information and Software Technology, 95, 219-237.
Morschheuser, B., Riar, M., Hamari, J., & Maedche, A. (2017). How games induce cooperation?A study on the relationship between
game features and we-intentions in an augmented reality game. Computers in Human Behavior, 77, 169–183
Nicholson, S. (2012). A user-Centered theoretical framework for meaningful gamification, paper presented at the Games+ Learn-
ing+ Society 8.0. 8.0, Madison, USA.
Peltokorpi, V. (2015). Corporate language proficiency and reverse knowledge transfer in multinational corporations: interactive ef-
fects of communication media richness and commitment to headquarters. Journal of International Management, 21(1), 49-62.
Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended
remedies. Journal of Applied Psychology, 88(5), 879-903.
Prensky, M. (2001). Fun, play and games: What makes games engaging. Digital game-based learning, 5(1), 5-31.
Prestopnik, N. R., & Tang, J. (2015). Points, stories, worlds, and diegesis: Comparing player experiences in two citizen science
games. Computers in Human Behavior, 52,492–506.
Rajanen, M., & Rajanen, D. (2017). Usability benefits in gamification. In Proceedings of the 1st GamiFin Conference (pp. 87-95).
Reinartz, W., Haenlein, M., & Henseler, J. (2009). An empirical comparison of the efficacy of covariance-based and variance-based
SEM. International Journal of Research in Marketing, 26(4), 332-344.
Rice, R. E. (1993). Media appropriateness: Using social presence theory to compare traditional and new organizational media. Hu-
man communication research, 19(4), 451-484.
Rigdon, E. E. (2012). Rethinking partial least squares path modeling: In praise of simple methods. Long Range Planning, 45(5),
341-358.
Rigdon, E. E., Ringle, C. M., & Sarstedt, M. (2010). Structural modeling of heterogeneous data with partial least squares. Review
of Marketing Research, 7(7), 255-296.
Rodrigues, L. F., Costa, C. J., & Oliveira, A. (2014, May). How gamification can influence the web design and the customer to use
the e-banking systems. In Proceedings of the International Conference on Information Systems and Design of Communica-
tion (pp. 35-44). ACM.
Roth, S., Schneckenberg, D., & Tsai, C. W. (2015). The ludic drive as innovation driver:Introduction to the gamification of innova-
tion. Creativity and Innovation Management, 24(2), 300–306.
Saha, R., Manna, R., & Geetha, G. (2012, September). Captchino-a gamification of image-based captchas to evaluate usability
issues. In 2012 International Conference on Computing Sciences (pp. 95-99). Washington: IEEE.
Sánchez, R. A., & Hueros, A. D. (2010). Motivational factors that influence the acceptance of Moodle using TAM. Computers in
Human Behavior, 26(6), 1632-1640.
dos Santos, A. C., Zambalde, A. L., Veroneze, R. B., Botelho, G. A., & de Souza Bermejo, P. H. (2015, August). Open innovation and
social participation: a case study in public security in Brazil. In international conference on electronic government and the
information systems perspective (pp. 163-176). Springer, Cham.
Saoud, J., & Jung, T. (2018). An Ethical Perspective of the use of AR Technology in the Tourism Industry. In Augmented Reality and
Virtual Reality (pp. 33-46). Springer, Cham.
464
ISSN 2538–7138 /Online/ (P. 450-465)
Saraf, N., Langdon, C. S., & Gosain, S. (2007). IS application capabilities and relational value in interfirm partnerships. Information
Systems Research, 18(3), 320-339.
Scarpi, D. (2005). Hedonic and utilitarian behaviour in specialty shops. The Marketing Review, 5(1), 31-44.
Segars, A. H., & Grover, V. (1998). Strategic information systems planning success: an investigation of the construct and its mea-
surement. MIS Quarterly, 22(2), 139-163.
Sherry Jr, J. F. (1990). A sociocultural analysis of a Midwestern American flea market. Journal of Consumer Research, 17(1), 13-30.
Simões, J., Redondo, R. D., & Vilas, A. F. (2013). A social gamification framework for a K-6 learning platform. Computers in Human
Behavior, 29(2), 345-353.
Slack, N. (1994). The importance-performance matrix as a determinant of improvement priority. International Journal of Opera-
tions & Production Management, 14(5), 59-75.
Sotiriou, S., & Bogner, F. X. (2008). Visualizing the invisible: Augmented reality as an innovative science education scheme. Ad-
vanced Science Letters, 1(1), 114-122.
Squire, K. D., & Jan, M. (2007). Mad City Mystery: Developing scientific argumentation skills with a place-based augmented reality
game on handheld computers. Journal of Science Education and Technology, 16(1), 5–29.
Stone, M. (1974). Cross‐validatory choice and assessment of statistical predictions. Journal of the Royal Statistical Society: Series
B (Methodological), 36(2), 111-133.
Stull, A. T., Fiorella, L., Gainer, M. J., & Mayer, R. E. (2018). Using transparent whiteboards to boost learning from online STEM
lectures. Computers & Education, 120,146–159.
Su, C. H. (2018). Exploring sustainability environment educational design and learning effect evaluation through migration
theory: An example of environment educational serious games. Sustainability, 10(10), 3363.
Su, C. H. (2017). Designing and Developing a Novel Hybrid Adaptive Learning Path Recommendation System (ALPRS) for Gamifi-
cation Mathematics Geometry Course. Eurasia Journal of Mathematics, Science and Technology Education, 13(6), 2275-2298.
Su, C. H., & Cheng, C. H. (2015). A mobile gamification learning system for improving the learning motivation and achievements.
Journal of Computer Assisted Learning, 31(3), 268-286.
Su, C. H., & Cheng, T. W. (2019). A Sustainability Innovation Experiential Learning Model for Virtual Reality Chemistry Laboratory:
An Empirical Study with PLS-SEM and IPMA. Sustainability, 11(4), 1027.
Sullivan, J. (1999). What are the functions of corporate home pages? Journal of World Business, 34(2), 193-210.
Talley, T. (2016). The STEM coaching handbook: Working with teachers to improve instruction. Routledge.
Tenenhaus, M., Amato, S., & Esposito Vinzi, V. (2004, June). A global goodness-of-fit index for PLS structural equation modelling.
In Proceedings of the XLII SIS scientific meeting (Vol. 1, pp. 739-742).
Tenenhaus, M., Vinzi, V. E., Chatelin, Y.-M., & Lauro, C. (2005). PLS path modeling. Computational Statistics & Data Analysis, 48(1),
159-205.
Webb, D. (2013). STEM Lesson Essentials, Grades 3-8: integrating Science, Technology, Engineering, and Mathematics. Teacher
Education and Practice, 26(2), 358-364.
Venkatesh, V., & Bala, H. (2008). Technology acceptance model 3 and a research agenda on interventions. Decision Sciences,
39(2), 273-315.
Wang, W., Hsieh, J. P.-A., & Song, B. (2012). Understanding user satisfaction with instant messaging: An empirical survey study.
International Journal of Human-Computer Interaction, 28(3), 153-162.
Wetzels, M., Odekerken-Schröder, G., & van Oppen, C. (2009). Using PLS path modeling for assessing hierarchial construct models:
Guidelines and empirical illustration. MIS Quarterly, 33(1), 177-195.
Wu, H. K., Lee, S. W. Y., Chang, H. Y., & Liang, J. C. (2013). Current status, opportunities and challenges of augmented reality in
education. Computers & education, 62, 41-49.
Wu, L., & Chen, J.-L. (2005). An extension of trust and TAM model with TPB in the initial adoption of on-line tax: an empirical
study. International Journal of Human-Computer Studies, 62(6), 784-808.
Yang, H.-H., & Su, C.-H. (2017). Learner Behaviour in a MOOC Practice-oriented Course: In Empirical Study Integrating TAM and
TPB. The International Review of Research in Open and Distributed Learning, 18(5), 35-66.
Zhao, Q., Chen, C.-D., & Wang, J.-L. (2016). The effects of psychological ownership and TAM on social media loyalty: An integrated
model. Telematics and Informatics, 33(4), 959-972.
Chung-Ho Su PhD in Design, Management, Associate Professor, Department of

Animation and Game Design, Shu-Te University, No. 59, Hengshan Rd.,
Yanchao Dist., Kaohsiung City 824, Taiwan (R.O.C.)
E-mail: mic6033@stu.edu.tw
Website: http://www.dgd.stu.edu.tw/team_member/%E8%98%87%E4%
B8%AD%E5%92%8C/
465
EXPLORING CHANGES
IN PRIMARY STUDENTS’
ATTITUDES TOWARDS SCIENCE,
TECHNOLOGY, ENGINEERING
AND MATHEMATICS (STEM)
ACROSS GENDERS AND GRADE
LEVELS
Abstract. Primary education is an essential Shao-Na Zhou,

stage and has an important impact on Hui Zeng,
students’ learning attitudes throughout the Shao-Rui Xu,
coming school years. The research explored Lu-Chang Chen,
the attitudes towards science, technology, Hua Xiao
engineering and mathematics among
students through all grade levels in primary
school. The Project-based Integrated STEM
Program was proposed to study the chang-
es of primary students’ attitudes towards
Introduction
STEM. An assessment of S-STEM which
consists of the STEM subscale and the 21st Nowadays, the development of science and technology greatly boost
century skills subscale was utilized for both the economy of many industrialized nations and developing countries.
pre-test and post-test. The results showed In the 21st century, there are increasing demands of jobs including posi-
that primary students exhibited little differ-
tions such as electrical and electronics engineer, computer programmer,
and construction manager, especially with the development of artificial
ent attitudes on the S-STEM in the pre-test,
intelligence. The labor market asks for the 21st century skills of knowledge,
regardless of gender and grade level. As critical thinking and collaboration in science, technology, engineering
evident from the comparison between the and mathematics (STEM) (Partnership for 21st Century Skills, 2004). The
pre-test and post-test, the Project-based demand reflects the importance of STEM education and motivates K-12
Integrated STEM Program had a positive ef- schools ( K-12 education refers to education for children from kindergar-
fect on student attitudes towards STEM. It is
ten through twelfth grade) and universities to pay more attention to the
teaching and training of 21st century skills. Attitudes towards STEM-related
encouraging if there are more STEM-related
subjects from kindergarten kids through K-12 school students could have
programs implemented at all educational an impact on the possibility of the participation in the future STEM careers
stages covering the primary level. (Business-Higher Education Forum, 2010; President’s Committee of Advi-
Keywords: gender differences, grade sors on Science and Technology [PCAST], 2010). Therefore, the improve-
levels, project-based integrated STEM pro- ment of students’ attitudes towards STEM requires increasing efforts in
gram, school students, STEM education.
K-12 school systems.
However, many countries including United States and Countries in
Europe were currently facing the challenge of decreasing numbers of
students who were interested in the STEM-related subjects (National Sci-
ence Board [NSB], 2012; Osborne & Dillon, 2008). Many companies were
experiencing the shortage of qualified candidates with STEM-related skills
Shao-Na Zhou, Hui Zeng, Shao-Rui Xu,
Lu-Chang Chen, Hua Xiao and many job opportunities were unfilled (Pathways to Prosperity Project,
South China Normal University, China 2011). It was reported by University of California at Los Angeles (UCLA)
that K-12 and post-secondary students lost their interest in STEM and
40-60% of students in the university majoring in engineering and science
466
ISSN 1648–3898 /Print/ EXPLORING CHANGES IN PRIMARY STUDENTS’ ATTITUDES TOWARDS SCIENCE,
TECHNOLOGY, ENGINEERING AND MATHEMATICS (STEM) ACROSS GENDERS AND GRADE
ISSN 2538–7138 /Online/ LEVELS
(P. 466-480)
switched to other majors or failed to obtain a degree (Drew, 2011). Many students in secondary school and
post-secondary were switching from STEM majors to other fields. As well, most secondary school students
come out on top in math, but did not choose STEM majors in college. In addition, only half of those students
who originally studied STEM majors actually completed STEM degrees. (Carnevale, Smith, & Melton, 2011). A
number of reports have warned that the current education system did not provide students with the appropriate
science and technology training to prepare them for the skills and employment needs of the 21st century (EC,
2016; UNESCO, 2015). Holdren and Lander (2012) explained that the reasons for explaining the phenomenon
included boring introductory courses and the lack of teaching support after school in mathematics courses.
In 2011, educators were called on by the National Research Council (NRC) to increase the number of students
who pursued STEM-related careers after secondary school (National Research Council, 2011). Experts on the
President’s Committee of Advisors on Science and Technology stated that improving K-12 students’ interest
and attitudes towards STEM was as significant as augmenting the overall STEM skills (President’s Committee
of Advisors on Science and Technology [PCAST], 2010).
Literature Review
Student attitudes towards STEM across genders
In the science literature, researchers found some contradictory evidences regarding the difference of
student attitudes towards STEM across genders. On the one hand, some studies indicated that males had
more positive attitudes than females (Jarvis & Pell, 2005). Another study revealed that the likelihood of career
interest in a STEM-related field was 2.9 times higher for males than that for females by the end of secondary
school (Sadler, Sonnert, Hazari, & Tai, 2012). Some researchers reported that male students in the middle school
presented a more favorable attitude towards STEM than females (Catsambis, 1995; Jones, Howe, & Rua, 2000;
Piburn & Baker, 1993; Greenfield, 1996). A study, conducted by Catsambis (1995), indicated that male students
took more positive attitudes towards STEM than females. Simpson and Oliver (1985), and Hykle (1993) suggested
that males were more inspired to succeed in science than females. Many females lacked of positive attitudes
towards science to continue with scientific inquiries (Hacieminoglu, 2016). Sadler et al. (2012) revealed that
females had fewer desires to pursue STEM careers. In most cases, females removed the STEM-related opportuni-
ties out of their career goals (Lent et al., 2005). It seemed that females prefered to connect school assignments
from a more social context with the real world. Also, formal role models were also a significant factor that was
often missing for girls in STEM areas (McCrea, 2010). In some researches of primary school students’ attitudes
towards science, girls showed less liking for STEM subjects than boys. More specifically, girls preferred fewer
STEM lessons (Denessen, Vos, Hasselman, & Louws, 2015) and boys were eager to study in STEM areas (DeWitt
& Archer, 2015).
On the other hand, some researchers emphasized similar attitudes towards STEM among male and female
students (Dhindsa & Chung, 2003; Miller, Lietz, & Kotte, 2002; Smist, Archambault, & Owen, 1994; Akpinar, Yildiz,
Tatar, & Ergin, 2009). It was also pointed out that students’ attitudes towards STEM depended on the type of
science that they prefered. For example, some research results revealed that males had a more favorable at-
titude towards physical sciences while females showed a more favorable attitude towards biological sciences
(Schibeci & Riley, 1986; Weinburgh, 1995). Another similar research by Jones, Howe and Ria (2000) reported that
males were interested in physics sciences, while females preferred life sciences. Furthermore, as stated from a
previous study, girls were less positive than boys in their semantic perceptions of STEM, but girls were notably
more positive than boys in semantic perceptions of STEM as a career (Christensen, Knezek, & Tyler-Wood, 2014).
Student attitudes towards STEM across grade levels
Primary science is an essential part of the K-12 science education system. It is recommended that students
are taught to develop scientific attitudes towards STEM at an early age, starting with basic knowledge and skills
from infant school through the primary years. Disappointingly, the scientific lessons have been diminished
by the enactment of No Child Left Behind (NCLB) in many primary schools in the United States (Chonkaew,
Sukhummek, & Faikhamta, 2016). It was reported that students’ attitude towards STEM, such as enjoyment of
STEM lessons and career interest in engineering and science, could be influenced significantly by grade levels
467
EXPLORING CHANGES IN PRIMARY STUDENTS’ ATTITUDES TOWARDS SCIENCE,
LEVELS ISSN 2538–7138 /Online/
(P. 466-480)
(Toma & Greca, 2018). It seemed that as grade level increased, scientific attitudes towards STEM dropped con-
sistently (Ali et al., 2013; Said, Summers, Abd-El-Khalick, & Wang, 2016).
As reported, the interest from ten-year-old students in STEM-related subjects was high, regardless of gen-
der (Haworth, Dale, & Plomin, 2008), but their interest fell down considerably by the age of fourteen (Osborne,
Simon, & Collins, 2003). An updated research revealed that student attitudes towards STEM were gradually
improved from the age of ten and they were formed by the age of fourteen (Osborne, Simon, & Tytler, 2009). This
period mainly might influence students’ career interest and career choices. Lindahl (2007), after a longitudinal
research with 12-to-16 years old students, concluded that attitudes towards STEM and career interest became
evident at the age of thirteen and it became more difficult to engage students in STEM-related activities at later
ages. In Denessen, Vos, Hasselman, and Louws’ research (2015), first-year students in grade one in a secondary
school generally had more positive STEM attitudes than second-year students in grade two in secondary school.
Research Focus
Students’ choices of majors in college and their career interest after graduation are affected by their at-
titudes towards STEM at the primary school, middle school, or secondary school levels (Astin & Astin, 1993;
Maltese & Tai, 2011). However, in the literature reviews, most researches were inclined to explore student atti-
tudes towards STEM among students groups from grade four in upper primary school to postsecondary school
(Denessen, Vos, Hasselman, & Louws, 2015; Haworth, Dale, & Plomin, 2008; Lindahl, 2007; Osborne, Simon, &
Collins, 2003). There is an absence of related research about STEM attitudes involving primary students from
grade one to grade three. Undoubtedly, primary education including grade one to grade three is an essential
stage and has an important impact on student attitudes and scientific views throughout the coming school
years. It suggests there is a need to study the attitudes towards STEM among students through all grade levels
in primary school, involving the differences of attitudes across genders and grade levels. Furthermore, it is a
crucial issue about how to improve student attitudes towards STEM at primary level (Kahveci, 2015). Therefore,
a proposal for Project-based Integrated STEM Program was presented to gain a more particular knowledge
about its impact on primary students’ attitudes towards STEM.
The present research proposed the following questions:
1. Do primary students exhibit different attitudes towards STEM according to gender?
2. Do primary students exhibit different attitudes towards STEM according to grade level?
3. What change could the proposed Project-based Integrated STEM Program bring about on primary
students’ attitudes towards STEM?
The Project-based Integrated STEM Program
In order to improve student attitudes towards Science, Technology, Engineering, & Mathematics (STEM),
different proposals have been put forward in the previous researches or national reports. On one hand, some
researches concentrated on the renovation of each subject related to STEM (Bybee, 2013). On the other hand,
some educators and researchers focused on the integrative STEM education. For instance, SEEA group in
Scotland published national reports which recommend the implementation of STEM education (Science and
Engineering Education Advisory group, 2012). Australian Government and American Society for Engineering
Education put forward an integrated STEM curriculum in Secondary Education (Australian Government, 2013;
ASEE, 2011; Ritz & Fan, 2014; Pitt, 2009). These governments and organizations mainly highlight the signifi-
cance of developing specialized STEM programs with inquiry teaching method at all educational stages. In the
research conducted by Ritz and Fan (2014), the researcher suggested a multidisciplinary STEM education with
the treatment of integrative STEM disciplines in real life. Also, Heil, Pearson, and Burger (2013) emphasized that
there was a need to develop STEM programs under sufficient theoretical frameworks and empirical studies.
However, the researcher pointed out that STEM programs were generally implemented in secondary school
and there were limited related programs conducted at the primary level (Heil, Pearson, & Burger, 2013). In this
research, the aim was to attempt at proposing a Project-based Integrated STEM Program to improve student
attitudes towards STEM in primary school.
468
(P. 466-480)
The Project-based Integrated STEM Program has been served as a support for STEM teaching activities
with twelve projects through grade one to grade six at the primary level. Within the Project-based Integrated
STEM Program, students’ learning process contains three phases for each project lasting for about 1 hour.
The first phase: It is a 5 minutes’ teacher-oriented activity with problem-based learning method. In this
phase, the teacher provides several questions on a special STEM theme with a real-life context. Students are
encouraged to ponder over the strategies of the experimental design and alternative solution of the problems
that they may encounter when they focus on a special STEM theme.
The second phase: It is a student-oriented section, in which students have their own creative ideas and
design their personal projects. In this phase, students are randomly divided into small teams with two students
in one team. In each team, students share their ideas with each other and make a decision to choose a better
experimental design. Students are encouraged to connect experimental design with scientific and mathemat-
ics knowledge. They construct their designs and explore new ideas. Sometimes they need to redesign their
experiments when the previous one is invalid or without any convincing results. Students are asked to draw a
conclusion from experimental results and explain the experimental phenomenon with the integrative knowl-
edge of science and mathematics. It lasts for about 40 minutes in this phase.
The third phase: In the final phase of about 15 minutes, presentations are carried out by each team. Stu-
dents present the experimental phenomena and results of their projects, and try to address the key concepts
and principles of the scientific and mathematics knowledge related to the STEM theme. In this phase, the role
of the teacher is to evaluate students’ statements and give supplementary explanations.
In the Project-based Integrated STEM Program, the same set of 12 STEM projects, which are described in
Table 1, were provided for students in grade one, grade two, and grade three. In China, the science course started
in the third year in the primary school according to the curriculum provision before the year of 2017. At the end
of 2017, the Chinese Ministry of Education published the latest science curriculum standards determining that
students should learn science from the first year of primary education. Teaching lower primary students content
knowledge related to science is a greenfield and tough work in the current educational background in China.
Therefore, the same STEM projects were designed for students from grade one to grade three. Table 2, table
3, and table 4 illustrate twelve STEM projects for students in grade four, grade five, and grade six, respectively.
This program is one of various school-based courses, which students are free to choose as an optional course.
The course is carried out once per week, and it does not occupy the normal school time.
Table 1. Twelve STEM projects for students in grade one, grade two, and grade three.
Week STEM Projects The descriptions of each project
1 Purple flour Color change by reaction with the iodine
2 Egg-beater The operation and function of the gear
3 Tumbler The knowledge of the gravity
4 Gyroscope How does a gyroscope spin on a plane
5 Yo-yo How does a yo-yo rotate in a vertical direction
6 Trailer The mechanical structure of the automobile
7 Needle emitting pellets The air pressure
8 Sandball How to produce sound by the vibration
9 Pirate ship The curvilinear motion
10 Beads floating on the water The buoyancy
11 Crane The lever rule
12 Robot arm The elasticity
469
(P. 466-480)
Table 2. Twelve STEM projects for students in grade four.
1 Magic Balloon The citric acid reacts with the soda

Triangle stability and graphic
2 The application of triangle and quadrilateral in daily life
reinforcement
3 Electric Fan How does the circuit work
4 Equal-armed lever The equal-armed lever
5 Colorful table lamp How does the LED light work
6 Labor-saving lever Learn the labor-saving lever
7 Caged Birds The application of the parallax
8 Laborious lever Learn the laborious lever
9 Making a steelyard The application of the lever in daily life
10 Compound lever Learn the compound lever
11 Runaway spider The application of the eccentric wheel
12 Lever comprehensive application Learn the application of leverage principle
Table 3. Twelve STEM projects for students in grade five.
1 Rainbow windmill The integration of circuit and parallax

Fixed pulley and movable
2 Learn simple fixed pulley and movable pulley
pulley 1
3 Rain alarm The application of the humidity sensor
Fixed pulley and movable
4 How do fixed pulley and movable pulley work
pulley 2
5 Storm frog How does the motor drive gears
6 Wireless power supply How is the energy transmitted wirelessly
7 Rotation of the pulley Learn the rotation of the pulley
8 Dryers The application of the centrifugal force

Pulley acceleration and decel-
9 Learn acceleration and deceleration of the pulley
eration
10 Four-wheel drive (4WD) How to make an electric trolley
11 Composite pulley drive 1 Learn the pulley drive
12 Composite pulley drive 2 The application of the pulley drive
Table 4. Twelve STEM projects for students in grade six.
1 Laser alarm The principle of laser alarm
2 Idler The application of the idler in daily life
3 Track tank How to make a track tank
470
(P. 466-480)
4 Gear drive How does the gear drive work
5 Electric carbon swing The electromagnetic induction
6 Gear ratio Learn the transmission ratio of the gear
7 Energy conversion demonstrator Learn the energy conversion
8 Crown gears How do crown gears work
9 Speed regulator Learn to make a speed regulator
10 Helical gears How do helical gears work
11 Gravity trolley The application of the gravity
12 Crown gears and helical gears Learn the application of crown and helical gears
Participants
The research was conducted in a comprehensive school in Guangdong, a central province in southern China.
The school provides not only primary education, but also kindergarten education and middle school education.
The school is equipped with various laboratories, such as physics laboratory, biology laboratory, chemical experi-
ment and integrated science laboratory, which can provide with experimental sites for the present research. The
researcher only needs to supplement some necessary experimental materials when carrying out the program.
The school consisted of 877 students at the primary level from grade one to grade six. The size of the each of
the classes varied from 38 to 48. Students who chose the Project-based Integrated STEM Program as an optional
course constituted the sample of our research. The sample of participants included 242 students from grade one to
grade six (173 boys, 67 girls, and 2 students without gender markers), with 20 students in grade one (15 boys and
5 girls), 55 students in grade two (47 boys, 6 girls, and 2 students without gender markers), 48 students in grade
three (34 boys and 14 girls), 51 students in grade four (36 boys and 15 girls), 33 students in grade five (19 boys and
14 girls), and 35 students in grade six (22 boys and 13 girls). All of the students volunteered to join the research
program. Most of these students did not receive extra tutoring on STEM-related skills except of the traditional
instruction at school, because their parents who are general workers, farmers or self-employed labourers have
limited time and financial support for additional STEM-related education outside school.
Procedures
In this research, the Project-based Integrated STEM Program was carried out in the spring semester of 2018.
Each STEM project was conducted by a science teacher for each week, and the entire program with twelve STEM
projects for each grade lasted for twelve weeks.
In order to explore how primary students exhibit different attitudes towards STEM according to gender and
grade level, and what change the proposed Project-based Integrated STEM Program could bring about on primary
student attitudes towards STEM, the 15-minute pre-test and post-test on student attitudes towards STEM were
administered. The pre-test was carried out among students in the first week before the implementation of the first
STEM project, and the post-test was carried out after the completion of the last STEM project in the 12th week. It
is necessary to note that, when the pre-test was carried out, it happened to conflict with the schedule of physical
examination arranged by the school. There was a small amount of participants who were unable to take the pre-
test, while they took part in both the entire Project-based Integrated STEM Program and the post-test. Therefore,
the participants in the pre-test were 181 students in total (127 boys and 54 girls), including 12 students in grade
one (7 boys and 5 girls), 21 students in grade two (17 boys and 4 girls), 48 students in grade three (34 boys and
14 girls), 51 students in grade four (36 boys and 15 girls), 21 students in grade five (11 boys and 10 girls), and 28
students in grade six (22 boys and 6 girls).
471
(P. 466-480)
Assessment
A questionnaire on student attitudes towards science, technology, engineering, and math named S-STEM
developed by a research team at North Carolina State University (Unfried, Faber, Stanhope, & Wiebe, 2015) was
utilized for both pre-test and post-test. The S-STEM questionnaire was constructed based on the items of Erkut
and Marx’s (2005) STEM attitudes questionnaire and the items on the attitudes towards 21st century skills from the
Student Learning Conditions questionnaire (Friday Institute, 2010). The S-STEM questionnaire which was produced
for upper primary school students was translated into Chinese as simple and suitable as possible for lower graders
by two of researchers. In particular, for the students in grade one and grade two, teachers would read the sentences
if necessary to ensure that students understand the meaning of all items. The S-STEM questionnaire consisted of
37 5-point (Strongly agree to strongly disagree) Likert-type items (see Aappendix). Within the entire questionnaire,
there are 26 items assessing students’ STEM attitudes, with 8 items (Q1-Q8) towards Mathematics, 9 items (Q9-Q17)
towards Science, and 9 items (Q18-Q26) towards Engineering/Technology. The remaining 11 items (Q27-37) in the
questionnaire are set as the 21st century skills subscale.
Data Analysis
The obtained data of student performances on S-STEM were described with regard to mean scores and
standard deviations. The reliabilities of S-STEM for both pre-test and post-test were estimated using Cronbach’s
alpha coefficients. The Cronbach’s alpha coefficients were 0.896 and 0.907 for the pre-test and post-test, indicating
sufficient consistencies in the outcomes of two tests in this research. The data supported the use of the S-STEM
questionnaire for the research. A t-test statistic was applied to test whether there were significant differences
between boy students and girl students, or among lower primary students and upper primary students. Also, a
t-test statistic was used to test the significant differences between pre-test and post-test.
Research Results
Primary Students’ Performances on S-STEM in the Pre-test
The overall mean score of students on the whole S-STEM in the pre-test was higher than 3 (M=3.63, SD=.61),
that indicates a positive attitude on the whole S-STEM test for all participants. Students achieved a little higher
mean score in the 21st century skills subscale (M=3.70, SD=.83) than that in the STEM subscale (M=3.61, SD=.60).
One of the research aims was to explore whether primary students exhibited different attitudes towards sci-
ence, technology, engineering and mathematics. Student performances were compared on the whole S-STEM in
the pre-test according to gender and grade level. In terms of grade levels, all of the primary students were divided
into two groups: lower primary group and upper primary group. The lower primary group included students from
grade one to grade three. And students from grade four to grade six constituted the upper primary group. Table 5
presents the mean score for each group of students and the results of t-test statistic. For the whole S-STEM, the data
showed that there was no significant difference either between boy students and girl students (t=.25, p=.804; boys:
M=3.64, SD=.64; girls: M=3.62, SD=.55) or between lower primary students and upper primary students (t=1.58,
p=.117; lower: M=3.71, SD=.58; upper: M=3.57, SD=.63).
Table 5. Primary Students’ Performances on S-STEM in the pre-test and the t-test results on the differences
between boy students and girl students, and between lower primary students and upper primary
students.
Subscales Groups M (SD) t p Effect Size
Boys 3.64(.64)
.25 .804 .040
Girls 3.62(.55)
The whole S-STEM
Lower primary students 3.71 (.58)
1.58 .117 .235
Upper primary students 3.57(.63)
472
(P. 466-480)
Subscales Groups M (SD) t p Effect Size
Boys 3.64(.63)
1. 25 .211 .204
Girls 3.52(.53)
STEM subscale
Lower primary students 3.71(.59)
2.07 .040 .310
Boys 3.64(.85)
-1.51 .134 -.245
Girls 3.84(.77)
21st century skills subscale
Lower primary students 3.73(.80)
.34 .732 .051
To further study students’ different performances on both the STEM subscale and the 21st century skills sub-
scale of S-STEM, the data in each subscale were categorized into four groups in terms of the variables of gender
and grade level.
For the STEM subscale, the statistical value (t=1.25, p=.211) indicated that there was no significant difference
between boy students and girl students, with the mean score of 3.64 (SD=.63) for boy students and the mean score
of 3.52 (SD=.53) for girl students. However, a significant difference (t=2.07, p=.040) was detected in terms of grade
levels. From the data, the mean score of lower primary students (M=3.71, SD=.59) was statistically significantly
higher than that of upper primary students (M=3.52, SD=.60).
For the 21st century skills subscale of S-STEM, no significant differences were detected either between boy
students and girl students (t=-1.51, p=.134; boys: M=3.64, SD=.85; girls: mean score=3.84, SD=.77) or between
lower primary students and upper primary students (t=.34, p=.732; lower: M=3.73, SD=.80; upper: M=3.68, SD=.86).
It’s worth noting that boy students pursued higher scores than girl students in the whole S-STEM and the STEM
subscale, but girl students performed better on the 21st century skills subscale, even though all these differences
were not significant.
Primary Students’ Differences of Performances on S-STEM between the Pre-test and Post-test
Another aim of the study was to investigate what change the proposed Project-based Integrated STEM Pro-
gram would bring about on primary students’ attitudes towards STEM. The post-test for all students was carried
out after the completion of twelve STEM projects of the Project-based Integrated STEM Program in the 12th week.
The mean score with standard deviation of each group in the post-test is shown in table 6. Roughly comparing the
data between the pre-test and post-test, it could be found that students performed better in the post-test than in
the pre-test. Obviously, boy students had the same mean score of 3.64 in the whole S-STEM, the STEM subscale and
the 21st century skills subscale in the pre-test, but the mean scores in the post-test were much higher, with 3.82,
3,76, 3.97 for the whole S-STEM, the STEM subscale and the 21st century skills subscale, respectively. With respect to
the 21st century skills subscale, the mean score of each group was very close to 4, with 3.97, 3.93, 3.95, and 3.98 for
boy students, girl students, lower primary students, and upper primary students, respectively. However, all mean
scores of the four groups were less than 3.90 for the 21st century skills subscale in the pre-test.
Table 6. Primary Students’ Performances on S-STEM in the post-test.
M (SD)
Groups The whole S-STEM STEM subscale 21st century skills subscale
All participants 3.82 (.57) 3.76 (.62) 3.97 (.69)
Boys 3.90 (.56) 3.86(.62) 3.97(.68)
Girls 3.62 (.54) 3.49(.56) 3.93(.70)
473
(P. 466-480)
M (SD)
Groups The whole S-STEM STEM subscale 21st century skills subscale
Lower primary students 3.87 (.62) 3.83(.68) 3.95(.73)
Upper primary students 3.78 (.51) 3.69(.55) 3.98(.64)
A t-test statistic was used to test the significant differences between pre-test and post-test. The data are shown
in table 7. Overall, there was a significant difference on the whole S-STEM between pre-test and post-test (t=-3.27,
p=.001), indicating that primary students performed much better in the post-test than in the pre-test. For gender,
the difference (t=-3.64, p=.000) was detected significantly between pre-test and post-test for boy students on the
whole S-STEM, but the performance of girl students on the whole S-STEM in the post-test was equivalent to that in
the pre-test (t=-.02, p=.982). For grade level, students in both lower primary group and upper primary group had
better performances on the whole S-STEM in the post-test than that in the pre-test (lower: t=-1.78, p=.077; upper:
t=-2.60, p=.010 ). The results indicate that except for girl students, students in other groups showed much better
performances on the whole S-STEM in the post-test.
Table 7. The t-test results of the differences on S-STEM and its subscales between pre-test and post-test for
four Groups of students: boy students and girl students, lower primary students, and upper primary
students.
t-test (pre-post)
Subscales Groups
t p Effect Size
All participants -3.27 .001 -.321

Boys -3.64 .000 -.424
The whole S-STEM Girls -.02 .982 -.004
Lower primary students -1.78 .077 -.254
Upper primary students -2.64 .009 -.358
Boys -3.04 .003 -.354
STEM subscale Girls .33 .741 .061
Boys -3.71 .000 -.432
21st century skills subscale Girls -.67 .507 -.122
When it comes to the STEM subscale, overall, students in the post-test demonstrated a higher level of at-
titudes towards STEM than that in the pre-test (t=-2.64, p=.009). Besides, boy students performed significantly
better in the post-test than in the pre-test (t=-3.04, p=.003). Also, the difference between pre-test and post-test for
upper primary students was significant at the .05 level (t=-2.14, p=.034). While, the statistical value (t=.33, p=.741)
showed that girl students did not perform much differently on the attitude towards STEM between the pre-test
and the post-test. And the difference between pre-test and post-test for lower primary students was not detected
significantly (t=-1.35, p=.179).
474
(P. 466-480)
For the 21st century skills subscale, the significant differences resembled those of the whole S-STEM. Overall,
the difference between pre-test and post-test was significant (t=-3.47, p=.001), indicating that students in the
post-test achieved a noticeably higher performance on the attitude towards the 21st century skills than that in the
pre-test. Among four groups of students, there were significant differences between the pre-test and the post-test
among three groups of students: boy students (t=-3.71, p=.000), lower primary students (t=-2.07, p=.039), and up-
per primary students (t=-2.86, p=.005). However, no statistically significant difference was consistently detected
between pre-test and post-test for girl students (t=-0.67, p=.507).
Discussion
Primary Students’ Performances on S-STEM across Genders and Grade Levels
The above analysis in the pre-test provided us the interesting result that primary students exhibited little
different attitudes towards science, technology, engineering and mathematics on the S-STEM, regardless of gen-
der and grade level. No significant differences were detected either between boy students and girl students, or
between lower primary students and upper primary students. As evident from the results of different subscales of
the S-STEM, boy students showed a bit better attitude on the STEM subscale, but a little worse attitude on the 21st
century skills subscale than girl students. However, both differences were not significant. In the literature, there were
some findings regarding the difference of student attitudes towards STEM across genders. Most studies indicated
that boy students had more positive attitudes towards STEM than girl students (Catsambis, 1995; Greenfield, 1996;
Jarvis & Pell, 2005; Jones, Howe, & Rua 2000; Piburn & Baker, 1993) and girl students preferred fewer STEM lessons
(Denessen, Vos, Hasselman, & Louws, 2015). However, this research into primary students’ attitudes towards STEM
suggests that boy students and girl students do not exhibit differently on either the attitudes towards STEM, or the
attitudes towards the 21st century skills. There is the evidence supporting the previous researches which confirmed
similar attitudes towards STEM between boy students and girl students (Akpinar, Yildiz, Tatar, & Ergin, 2009; Dhindsa
& Chung, 2003; Miller, Lietz, & Kotte, 2002; Smist, Archambault, & Owen, 1994).
From the perspective of grade level, there was a disappointing result that upper primary students achieved
worse performance than lower primary students on the whole S-STEM, the STEM subscale, and the 21st century
skills subscale. A significant difference was detected in the STEM subscale between the two groups of students,
indicating that lower primary students performed much better on the attitude towards science, technology,
engineering and mathematics than upper primary students. Among the participants in this research, students
from grade one and grade two in the lower primary group rarely receive science course, as the science course in
China started from the third year in the primary school according to the curriculum provision before the year of
2017. All students in the upper primary group from grade four, grade five, and grade six have received different
years of science course according to the curriculum plan. The result suggests that the current science curriculum
and instruction fail to bring positive effect to students on their attitude towards science, technology, engineering
and mathematics at the primary level. The results strongly support the opinion in the solid body of literature that
students’ attitudes towards STEM dropped consistently as grade level increased (Ali et al., 2013; Osborne, Simon,
& Collins, 2003; Said, Summers, Abd-El-Khalick, & Wang, 2016). It reflects the failure in the traditional educational
system to improve primary students’ positive attitudes towards STEM. As reported, many countries were currently
facing the challenge of decreasing numbers of students who were interested in the STEM-related subjects (Na-
tional Science Board [NSB], 2012; Osborne & Dillon, 2008). Experts stated that improving K-12 students’ interest and
attitudes towards STEM was significant (President’s Committee of Advisors on Science and Technology [PCAST],
2010). The result gives full expression to the importance of STEM education and the training of 21st century skills.
It demands the educational system to provide students with the appropriate science and technology training to
prepare them for the skills and employment needs of the 21st century (EC, 2016; UNESCO, 2015).
The Effect of Project-based Integrated STEM Program on Primary Students’ Attitudes towards STEM
The Project-based Integrated STEM Program was a core strategy and implementation for supporting the
STEM teaching activities through grade one to grade six at the primary level in the research. As evident from the
comparison between pre-test and post-test, it could be found that the Project-based Integrated STEM Program
had a positive effect on student attitudes towards STEM. Overall, students improved their performances not only
475
(P. 466-480)
on the whole S-STEM, but also on the STEM subscale and the 21st century skills subscale in the post-test, compared
with the result in the pre-test. Although, some governments and organizations have addressed the significance
of STEM education and put forward different proposals in the national reports or previous researches, such as the
implementation of STEM education in Scotland (Science and Engineering Education Advisory group, 2012), and
the integrated STEM curriculum in Australian and American (ASEE, 2011; Australian Government, 2013; Pitt, 2009;
Ritz & Fan, 2014), most of these proposals were generally implemented in secondary school and there were limited
related projects conducted at the primary level (Heil, Pearson, & Burger, 2013). In this study, the Project-based
Integrated STEM Program was designed for students at the primary level and its instructional achievements were
remarkable and encouraging. It reveals that the implementation of the Project-based Integrated STEM Program is
effective for primary students. It is hopeful if there are more and more specialized STEM programs and integrated
STEM curricula propounded and implemented at all educational stages covering the primary level.
In the further analysis on the differences of student attitudes towards STEM between pre-test and post-test,
there were some special results from the aspects of gender and grade level. Regarding to gender, boy students
effectively enhanced their performances on the attitudes towards STEM in the post-test in contrast to the pre-test.
They showed significant differences on the whole S-STEM, the STEM subscale and the 21st century skills subscale
between the pre-test and post-test. However, the performances of girl students on either the whole S-STEM or
two subscales of the STEM and the 21st century skills in the post-test were equivalent to those in the pre-test. No
significant difference was consistently detected between pre-test and post-test for girl students. From the data,
even though there was not much difference between boy students and girl students in the pre-test, boy students
achieved much higher scores in the post-test than those of girl students, especially in the whole S-STEM (t=3.57,
p=.001), and the STEM subscale (t=4.45, p=.000). From the perspective of girl students who have participated in
twelve STEM projects identical to those of boy students at the same grade, the Project-based Integrated STEM
Program did not support the change of their attitudes towards STEM. It is in line with the previous research results
that boy students were more inspired to succeed in science than girl students (Hykle, 1993; Simpson & Oliver, 1985).
The reason for the lack of improvement of girl students may be consistent with previous studies in which girl stu-
dents preferred fewer STEM lessons and tended to report more unfavorable attitudes on STEM (Catsambis, 1995;
Denessen, Vos, Hasselman, & Louws, 2015; DeWitt & Archer, 2015; Greenfield, 1996; Jones, Howe, & Rua 2000; Piburn
& Baker, 1993). In addition, the majority of the proposed projects in the Project-based Integrated STEM Program
has been designed with the contexts of physical sciences, which failed to motivate girl students’ learning interest.
As revealed in previous researches, girl students might have had a more favorable attitude towards life sciences
(Howe and Ria, 2000) and biological sciences (Schibeci & Riley, 1986; Weinburgh, 1995).
For the aspect of grade level, both lower primary students and upper primary students had a positive change
on the whole S-STEM according to the data. It indicates that the proposed Project-based Integrated STEM Program
has achieved significant results to improve the attitudes towards STEM for students in both groups. It is seen that
the differences between pre-test and post-test on the whole S-STEM and the 21st century skills subscale for upper
primary students were significant at the 0.01 level, but the differences between pre-test and post-test on these two
aspects for lower primary students were significant at the .05 level. It is also observed that upper primary students
performed significantly better in the STEM subscale in the post-test, but lower primary students did not perform
much differently on the attitudes towards STEM between the pre-test and the post-test. Although, there was a
disappointing result that upper primary students performed worse than lower primary students on the whole
S-STEM, the STEM subscale, and the 21st century skills subscale in the pre-test, the Project-based Integrated STEM
Program has promoted more positive changes for upper primary students than lower primary students in the
post-test. The result is inspiring that although upper primary students currently may not show favorable attitudes
or interest towards STEM, they are able to develop more positive attitudes and interest by studying STEM-related
projects. From another perspective, there is a need to pay much attention to lower primary students who do not
perform much better on the STEM subscale in the post-test. It is worth pondering the reason that lower primary
students achieved significant improvement on the whole S-STEM and the 21st century skills subscale, while they
had no positive change on the STEM subscale. All details of the procedure including content settings, project
implementation approach and schedule arrangement of the Project-based Integrated STEM Program are needed
to reflect on. For lower primary students, developing suitable programs to improve their attitudes towards STEM
will be challenging and encouraging work in future research.
476
(P. 466-480)
Conclusions and Future Work
The present research explored the attitudes towards science, technology, engineering and mathematics among
students through all grade levels in primary school using the S-STEM questionnaire for the pre-test. The results
showed that primary students exhibited almost no difference of attitudes on the S-STEM, regardless of gender and
grade level. After the completion of twelve STEM projects of the Project-based Integrated STEM Program, students
had a positive change on their attitudes towards STEM in the post-test. However, boy students effectively enhanced
their performances on the attitudes towards STEM, while girl students did not. Also, the program has promoted
more positive changes for upper primary students than lower primary students.
Based on the current research results, future work maybe of the following aspects: (1)how to improve the
attitudes towards STEM for girl students at the primary level? From the data of the research, girl students have
participated in twelve STEM projects which were identical to those of boy students at the same grade, but the
Project-based Integrated STEM Program did not support the change of their attitudes towards STEM. Therefore,
designing suitable projects or curricula to develop the attitudes towards STEM for girl students at the primary level
is one of the challenging work in the later research. (2)how to develop lower primary students’ attitudes towards
STEM? In the present results, the Project-based Integrated STEM Program has promoted many positive changes
on the attitudes towards STEM for upper primary students, but lower primary students did not perform much
better on the STEM subscale of the S-STEM. It may make sense to ponder the reason that there was no significant
progress on the attitudes towards STEM for lower primary students. As well, it is worth designing suitable STEM-
related projects to improve lower primary students’ attitudes towards STEM in future work. (3)do primary students
exhibit different attitudes on subscales of science, technology, engineering and mathematics? In the research, the
assessment of S-STEM consists of 9 items, 8 items, and 9 items towards STEM respectively. It’s worthwhile to explore
primary students’ various performances on the subscales of the STEM, and it may show great importance on the
future research of the correlation of these subscales of STEM with each other.
Acknowledgements
The authors would like to acknowledge the help of the anonymous reviewers. The research is supported in
part by the Humanities and Social Sciences on Young Fund of the Ministry of Education of P.R. China under the
Grant No. 17YJC880140, and the Humanities and Social Sciences of the Ministry of Education of P.R. China under
the Grant No. 18YJA880096.
References
Akpinar, E., Yildiz, E., Tatar, N., & Ergin, Ö. (2009). Students’ attitudes toward science and technology: An investigation of gender,
grade level, and academic achievement. Procedia Social and Behavioral Sciences, 1(1), 2804-2808.
Ali, M. M., Yager, R. E., Hacieminoglu, E., & Caliskan, I. (2013). Changes in student attitudes regarding science when taught by teach-
ers without experiences with a model professional development program. School Science and Mathematics, 113(3), 109-119.
ASEE (American Society for Engineering Education). (2011). Preparing future engineers around the world. PRISM, 21(5), 26-34.
Astin, A. W., & Astin, H. S. (1992). Undergraduate science education: The impact of different college environments on the educational
pipeline in the sciences. Final Report. Los Angeles: Higher Education Research Institute.
Australian Government. (2013). Science, Technology, Engineering and Mathematics in the National Interest: A Strategic Approach.
Retrieved from https://www.chiefscientist.gov.au/wp-content/uploads/STEMstrategy290713FINALweb.pdf.
Bybee, R. W. (2013). The case for STEM education: Challenges and opportunities. Arlington: NSTA Press.
Business-Higher Education Forum. (2010). Increasing the number of STEM graduates: Insights from the U.S. STEM education model-
ing project. Washington, DC: Author.
Carnevale, A. P., Smith, N., & Melton, M. (2011). STEM: Science, technology, engineering, mathematics. Georgetown University Center
on Education and the Workforce: Washington, DC.
Catsambis, S. (1995). Gender, race, ethnicity, and science education in the middle grades. Journal of Research in Science
Teaching, 32(3), 243-257.
Chonkaew, P., Sukhummek, B., & Faikhamta, C. (2016). Development of analytical thinking ability and attitudes towards science
learning of grade-11 students through science technology engineering and mathematics (stem education) in the study
of stoichiometry. Chemistry Education Research & Practice, 17(4), 842-861.
Christensen, R., Knezek, G., & Tyler-Wood, T. (2014). Student perceptions of science, technology, engineering and mathematics
(STEM) content and careers. Computers in Human Behavior, 34 (2014), 173–186.
Denessen, E., Vos, N., Hasselman, F., & Louws, M. (2015). The relationship between primary school teacher and student attitudes
towards science and technology. Education Research International, 2015, 1-7.
477
(P. 466-480)
DeWitt, J., & Archer, L. (2015). Who aspires to a science career? A comparison of survey responses from primary and secondary
school students. International Journal of Science Education, 37(13), 2170–2192.
Dhindsa, H. S., & Chung, G. (2003). Attitudes and achievement of Bruneian science students. International Journal of Science
Education, 25, 907-922.
Drew, C. (2011). Why science majors change their minds.The New York Times, November 4, 2011.
EC (European Commision). (2016). Horizon 2020. Monitoring Report 2014. Luxembourg. http://ec.europa.eu/research/evalua-
tions/pdf/archive/h2020_monitoring_reports/second_h2020_annual_monitoring_report.pdf.
Erkut, S., & Marx, F. (2005). 4 schools for WIE (Evaluation report). Wellesley, MA: Wellesley College, Center for Research on Women.
Retrieved from http://www.coe.neu.edu/Groups/stemteams/evaluation.pdf.
Friday Institute. (2010). Governor Perdue’s North Carolina Student Learning Conditions Survey (SLCS): Survey Implementation Study.
Raleigh, NC: Author.
Greenfield, T. A. (1996). Gender, etnicity, science achievement and attitudes. Journal of Research in Science Teaching, 33, 901-933.
Hacieminoglu, E. (2016). Primary school students’ attitude toward science and related variables. International Journal of Environ-
mental & Science Education, 11(2), 35-52.
Haworth, C. M., Dale, P., & Plomin, R. (2008). A twin study in the genetic and environmental influences on academic performance
in science in nine-year-old boys and girls. International Journal of Science Education, 30(8), 1003-1025.
Heil, D. R., Pearson, G., & Burger, S. E. (2013). Understanding integrated STEM education: Report on a National Study. Paper presented
at 2013 ASEE Annual Conference & Exposition, Atlanta, Georgia.
Holdren, J. P., & Lander, E. S. (2012).Engage to excel: Producing one million additional college graduates with degrees in science,
technology, engineering, and mathematics (Executive Report). Washington, D.C.: President’s Council of Advisors on Science
and Technology.
Hykle, J. A. (1993). Template for gender-equitable science program. Paper presented at the annual meeting of the National As-
sociation for Research in Science Teaching. Atlanta, GA.
Lent, R. W., Sheu, H., Schmidt, J., Brenner, B. R., Wilkins, G., Brown, S. D., et al. (2005). Social cognitive predictors of academic
interests and goals in engineering: Utility for women and students at historically black universities. Journal of Counseling
Psychology, 52(1), 84–92.
Lindahl, B. (2007). A longitudinal Study of students’ attitudes towards science and choice of career. 80th NARST International Con-
ference, New Orleans.
Jarvis, T., & Pell, A. (2005). Factors influencing primary school children’s attitudes toward science before, during, and after a visit
to the UK National Space Centre. Journal of Research in Science Teaching, 42(1), 53-83.
Jones, G. M., Howe, A., & Rua, M. J. (2000). Gender differences in students’ experiences, interests and attitude toward science and
scientists. Science Education, 84, 180-192.
Kahveci, A. (2015). Assessing high school students’ attitudes toward chemistry with a shortened semantic differential. Chemistry
Education Research & Practice, 16(2), 283-292.
Partnership for 21st Century Skills. (2004). Homepage. Retrieved from http://www.21stcenturyskills.org/index.php.
Pathways to Prosperity Project (2011). Pathways to Prosperity: Meeting the Challenge of Preparing Young Americans for the 21st
Century. Harvard Graduate School of Education: Cambridge, MA.
Maltese, A. V., & Tai, R. H. (2011). Pipeline persistence: Examining the association of educational experiences with earned
degrees in STEM among US students. Science Education, 95(5), 877-907.
McCrea, B. (2010). Engaging girls in STEM. THE Journal. Retrieved from https://thejournal.com/articles/2010/09/08/engaging-
girls-in-stem.aspx.
Miller, L., Lietz, P., & Kotte, D. (2002). On decreasing gender differences and attitudinal changes. Factors influencing Aus-
tralian and English pupils’ choice of a career in science. Psyhology, Evolution, and Gender, 4, 69- 92.
National Research Council. (2011). Successful K-12 STEM Education: Identifying effective approaches in science, technology, engineer-
ing, and mathematics. Board on Science Education and Board on Testing and Assessment, Division of Behavioral and Social
Sciences and Education. Washington, DC: The National Academies Press.
National Science Board., (2012). Science and engineering indicators 2012. Arlington, VA: National Science Board.
Osborne, J. F., Simon, S., & Collins, S. (2003). Attitudes towards Science: A review of the literature and its implications. International
Journal of Science Education, 25, 1049-1079.
Osborne J., & Dillon J., (2008), Science education in Europe: Critical reflections (A Report to the Nuffield Foundation), London: King’s
College.
Osborne, J. F., Simon, S., &Tytler, R. (2009). Attitudes towards science: An update. In Proceedings of the Annual Meeting of the
American Educational Research Association, San Diego, Calif, USA.
PCAST, President’s Committee of Advisors on Science and Technology. (2010). Prepare and Inspire: K-12 Education in Science,
Technology, Engineering, and Math (STEM) for America’s Future. Washington, DC: Executive Office of the President.
Piburn, M. D., & Baker, D. R. (1993). If I were the teacher…qualitative study of attitude toward science. Science Education 77, 393-406.
Pitt, J. (2009). Blurring the Boundaries-STEM education and education for sustainable development. Design and Technology
Education: An International Journal, 14(1), 37-47.
Ritz, J. M., & Fan, S. C. (2014). International view of STEM education. Proceedings PATT-28 conference, Florida.
Sadler, P. M., Sonnert, G., Hazari, Z., & Tai, R. (2012). Stability and volatility of STEM career interest in high school: A gender study.
Science Education, 96(3), 411–427.
478
(P. 466-480)
Schibeci, R. A. , & Riley, J. P. (1986). Influence of students’ background and perceptions on science attitudes and achievement.
Journal of Research in Science Teaching, 23(3), 177-187.
Science and Engineering Education Advisory group. (2012). Supporting Scotland’s STEM education and culture. Retrieved from
http://www.gov.scot/resource/0038/00388616.pdf.
Smist, J. M., Archambault, F. X., & Owens, S. V. (1994). Gender difference in attitude toward science. Paper presented at the
annual meeting of the American Educational Research Association. New Orleans, LA, USA.
Simpson, R. D., & Oliver, J. S. (1985). Attitude toward science and achievement motivation profiles of male and female
science students in grades 6 through 10. Science Education, 69(4), 511-526.
Toma, R. B., & Greca, I. M. (2018). The effect of integrative STEM instruction on primary students’ attitudes toward science. EURASIA
Journal of Mathematics, Science and Technology Education. 14(4), 1383-1395.
Said, Z., Summers, R., Abd-El-Khalick, F., & Wang, S. (2016). Attitudes toward science among grades 3 through 12 Arab students
in Qatar: Findings from a cross-sectional national study. International Journal of Science Education, 38(4), 621–643.
Unfried, A., Faber, M., Stanhope, D. S., & Wiebe, E. (2015). The development and validation of a measure of student attitudes
toward science, technology, engineering, and math (s-stem). Journal of Psychoeducational Assessment, 33(7), 622-639.
UNESCO (United Nations Educational, Scientific and Cultural Organization), (2010). Education for all, Global monitoring report,
Retrived from: http://www.unesco.org/new/en/education/themes/leadingthe-international-agenda/efareport/reports/.
UNESCO (United Nations Educational, Scientific and Cultural Organization). (2015). Education for all 2000-2015. Achievements
and challenges, France. From https://doi.org/10.1126/science.1128690.
Weinburgh, M. (1995). Preparing gender inclusive science teachers: Suggestions from the literature. Journal of Science Teacher
Education, 6(2), 102-107.
Appendix
Items
1. Math has been my worst subject.

2. When I am older, I might choose a job that uses math.
3. Math is hard for me.
4. I am the type of student to do well in math.
5. I can understand most subjects easily, but math is difficult for me.
6. In the future, I could do harder math problems.
7. I can get good grades in math.
8. I am good at math.
9. I feel good about myself when I do science.
10. I might choose a career in science.
11. After I finish high school, I will use science often.
12. When I am older, knowing science will help me earn money.
13. When I am older, I will need to understand science for my job.
14. I know I can do well in science.
15. Science will be important to me in my future career.
16. I can understand most subjects easily, but science is hard for me to understand.
17. In the future, I could do harder science work.
18. I like to imagine creating new products.
19. If I learn engineering, then I can improve things that people use every day.
20. I am good at building and fixing things.
21. I am interested in what makes machines work.
22. Designing products or structures will be important for my future work.
23. I am curious about how electronics work.
24. I want to be creative in my future jobs.
479
(P. 466-480)
Items
25. Knowing how to use math and science together will help me to invent useful things.
26. I believe I can be successful in engineering.
27. I can lead others to reach a goal.
28. I like to help others do their best.
29. In school and at home, I can do things well.
30. I respect all children my age even if they are different from me.
31. I try to help other children my age.
32. When I make decisions, I think about what is good for other people.
33. When things do not go how I want, I can change my actions for the better.
34. I can make my own goals for learning.
35. I can use time wisely when working on my own.
36. When I have a lot of homework, I can choose what needs to be done first.
37. I can work well with all students, even if they are different from me.
Received: February 01, 2019 Accepted: June 02, 2019
Shao-Na Zhou Associate Professor, School of Physics and Telecommunication

Engineering, South China Normal University, Guangzhou 510006,
China.
E-mail: zhou.shaona@m.scnu.edu.cn
Hui Zeng PhD, School of Physics and Telecommunication Engineering, South
China Normal University, Guangzhou 510006, China.
E-mail: zenghui011222@163.com
Shao-Rui Xu Researcher, School of Physics and Telecommunication Engineering,
South China Normal University, Guangzhou 510006, China.
E-mail: jayxee@hotmail.co.uk
Lu-Chang Chen M.Sc., School of Physics and Telecommunication Engineering, South
China Normal University, Guangzhou 510006, China.
E-mail: 244557813@qq.com
Hua Xiao Professor, School of Physics and Telecommunication Engineering,
(Corresponding author) South China Normal University, Guangzhou 510006, China.
E-mail: xiaoh@scnu.edu.cn
480
INFORMATION
FOR CONTRIBUTORS
EDITORIAL AND REVIEW PROCEDURES
Journal of Baltic Science Education (JBSE) publishes original scientific research articles in the field of Natural Science Education and
related areas for all educational levels in the Baltic countries. It is possible to publish special (thematic) issues of JBSE. The papers should
be submitted and will be published in English. JBSE will promote to establish contacts between researchers and practical educators
both in the Baltic countries and countries around.
The authors of the manuscripts are responsible for the scientific content and novelty of the research materials. Articles, published
before in other international journals or papers’ collections will not be accepted for publication in JBSE.
As a publication that represents a variety of cross-disciplinary interests, both theoretical and practical, the JBSE invites manuscripts on
a wide range of topics, especially in the following areas:
• Didactics of natural sciences. • Philosophical, political, economical and social aspects
• Theory and practice in natural science teacher of natural science education.
education. • The supplementary natural science education.
• Integrated natural science education. • ICT in natural science education.
• Natural science and technological literacy. • The standardisation of natural science education etc.
• General and professional natural science education.
MANUSCRIPTS GUIDELINES
The structure of the research paper presented to the Journal of Baltic Science Education should be as follows: abstract - short report of the
investigation; introduction inc. aim and subject of the research; research methodologies and methods; results of the research incl. discussion;
conclusions; list of references in APA style.
The papers should be submitted in English. If English is a second language for the author, please consider having the manuscript
proof read and edited before submitting. The preliminary text of the article can be sent as a.doc file in the attachment by e-mail: mail.
jbse@gmail.com
The text must be elaborated in Word for Windows, using 12 point Times New Roman letters. An article should not exceed 7-10 A4 pages,
included figures, tables and bibliography. Publishing of longer articles should be negotiated separately. Texts margins: top and bottom 20mm,
left - 25mm, right - 20mm. The title: capital letters, 14pt, bold; space between the title and the author’s name is one line interval. Author’s name
and surname: small letters, 12pt, bold. Under the name, institution: 11 pt, italics; space between the title and the text: 1 line interval. Abstract
– about 100-150 words - precedes the text. The text: 12pt Single or Auto spacing, in one column. Key words: no more than five words. The
language must be clear and accurate. The authors have to present the results, propositions and conclusions in a form that can suit scientists
from different countries.
Titles of the tables and figures: 11 pt, small letters. Space between figures or tables and the text: 1 line interval. Introduction, titles of
chapters and subchapters: 12pt, bold, small letters. Numbers: Arabic, subchapters numbered by two figures (1.1, 1.2, etc.). Figures, tables and
captions should be inserted within the manuscript at their appropriate locations. Diagrams and graphs should be provided as finished black
and white line artwork or electronic images. When there are a number of illustrations, the author should endeavour to reduce the amount of
text to accommodate the illustrations in the limited space available for any article.
References in the text should be presented in brackets (Knox, 1988; Martin, 1995). If necessary, the page can be indicated: (Martin,
1995, p.48). The list of references should be presented after the text. The Words List of References: 11pt, bold, small letters. The references
should be listed in full at the end of the paper in the following standard form:
For books: Saxe, G.B. (1991). Cultural and Cognitive Development: Studies in Mathematical Understanding. Hillsdale, NJ: Erlbaum.
For articles: Bekerian, D.A. (1993). In Search of the Typical Eyewitness. American Psychologist, 48, 574-576.
For chapters within books: Bjork, R.A. (1989). Retrieval Inhibition as an Adaptive Mechanism in Human Memory. In: H.L. Roediger III & F.I.M.
Craik (Eds.), Varieties of Memory & Consciousness (pp. 309-330). Hillsdale, NJ: Erlbaum.
On a separate page, author - related data should be presented in English: name, surname, degree and academic title, institution, full
correspondence address in the clearest and most complete form /ordinary post and e-mail addresses /, position (to ensure anonymity in the
review process). The author (authors) should confirm in writing, that the manuscript has not been published in other journal or
handed over (transferred) to other journal for publication.
EDITORIAL AND REVIEW PROCEDURES

Manuscripts will be sent anonymously to reviewers with expertise in the appropriate area. All manuscripts will be rewieved by two
experts before JBSE’s accept them for publication. This process usually takes about two months. The journal co-editors will make minor
editorial changes; major changes will be made by the author(s) prior to publication if necessary. JBSE’s redaction will sent to author(s)
only one correcture which must be sent back within 2 weeks. JBSE will not review submissions previously published elsewhere through
print or electronic medium.
Manuscripts submitted to the JBSE cannot be returned to authors. Authors should be sure to keep a copy for themselves. Authors’
signatures should be at the end of the paper and its second checked proofs.
Manuscripts, editorial correspondence (and other correspondence for subscription and exchange), and any questions should be
sent to editor-in-chief or to regional redactors.
Journal`s requirements for the authors are available online: http://www.scientiasocialis.lt/jbse/files/JBSE_requirements_2019.pdf
481
JBSE
Problems of Psychology in the 21st

Century is an international, periodical,
peer reviewed scientific journal, issued by
the Scientia Socialis, UAB in cooperation
with SMC “Scientia Educologica”.
Editor-in-Chief
Prof. dr. Vincentas Lamanauskas, Scientific

Methodical Centre „Scientia Educologica“,
Republic of Lithuania
Problems of Psychology in the 21st Century

(ISSN 2029-8587) is abstracted and/or indexed
in:
Index Copernicus -
http://journals.indexcopernicus.com
EBSCO - http://search.ebscohost.com
Editorial Board
Dr., prof. Ferda Aysan, Dokuz Eylul University, Turkey

Dr., prof. Serhiy Boltivets, Grigory Kostyuk Psychological Institute of the Ukrainian National
Academy of Pedagogical Sciences, Ukraine
Dr., assoc. prof. Irena Gailiene, SMC "Scientia Educologica", Republic of Lithuania
Dr., prof. Irakli Imedadze, Dimitri Uznadze Georgian Psychological National Society, Georgia
Dr. Julia Lakhvich, Belarusian State University, Republic of Belarus
Dr., prof. Vladimir S. Karapetyan, Armenian State Pedagogical University named after Kh.
Abovyan, Armenia
Dr., prof. Vincentas Lamanauskas, Scientific Methodical Centre „Scientia Educologica“, Republic of
Lithuania (Editor-in-Chief)
Dr., prof. Maria Ledzińska, University of Warsaw, Poland
Dr., prof. Aleksandr Lobanov, Belarusian State Pedagogical University, Republic of Belarus
Dr., prof. Guna Svence, Riga Teacher Training and Educational Management Academy, Latvia
Website: http://www.jbse.webinfo.lt/PPC/Problems_of_Psychology.htm
Website: http://www.scientiasocialis.lt/ppc/
482
Dear colleagues,
GAMTAMOKSLINIS UGDYMAS / NATURAL SCIENCE EDUCATION – is a
periodical, peer reviewed, scientific-methodical journal, issued by the SMC „Scientia
Educologica“ in cooperation with Scientia Socialis. It is an international journal, wherein
the scientific and methodical articles published in Lithuanian, English and Russian
languages. This journal is intendent for the teachers of general education schools, the
lecturers of higher educational institutions and all, who are interested in the problems of
natural science education.
The GU/NSE journal welcomes the submission of manuscripts that meet the general
criteria of scientific and methodical (practical) papers.
GAMTAMOKSLINIS UGDYMAS / NATURAL SCIENCE EDUCATION

ISSN 1648-939X
http://gu.puslapiai.lt/GUwww/indeks.htm
http://gu.puslapiai.lt/GUwww/Indeksavimas.htm
http://oaji.net/journal-detail.html?number=514
Responsible This journal is abstracted / listed / indexed / cited in:
COPERNICUS INDEX , LIST OF SCIENCE EDUCATION
Publication JOURNALS, DRJI, JOURNALS OF INTEREST TO CHEMICAL
Peer Reviewed EDUCATORS, SciEdu 4 U.com, SKYLIGHT, OAJI, WebQualis
(QUALIS/CAPES ), ERIH PLUS, MIAR.
Open Access
Applied research/practical/methodical work. This type of submission is best suited for
practical/didactical work and reports, as well as position papers raising original and
provocative theoretical or practical discourses and questions (small-scale research, applied
research, didactical/methodical papers, case studies, best educational practices etc.). Each
submission is carefully reviewed by two independent reviewers and ranked based on:
quality of preparation, relevance to the educational community, didactical quality,
originality, and importance of the contribution.
Instruction for authors and other details are available on the journal`s website at:
http://gu.puslapiai.lt/GUwww/Bendra_informacija.htm
GU/NSE is an Open Access journal accessible for free on the Internet. Partial article
processing charges are: 5-8 EUR per one A4 page. Papers must be submitted on the
understanding that they have not been published elsewhere and are not currently under
consideration by another publisher. Optimal paper`s size: 5-8/12 pages.
For contacts, questions and papers submission: gu@gu.puslapiai.lt
Sincerely yours, Editorial Board
483
ISSN 1648-3898 /Print/, ISSN 2538-7138 /Online/
Compiled by: Vincentas Lamanauskas

Linguistic Editor: Ilona Ratkevičienė
Cover design by: Jurgina Jankauskienė
Layout design by : Linas Janonis

15 June 2019. Publishing in Quires 10,5. Edition 200
Publisher Scientia Socialis Ltd.,

Donelaicio Street 29, LT-78115 Siauliai, Lithuania
E-mail: scientia@scientiasocialis.lt
Phone: +370 687 95668
http://www.jbse.webinfo.lt/centras.htm
http://www.scientiasocialis.lt

Printing Šiauliai printing house
9A P. Lukšio Street
LT-76207 Šiauliai, Lithuania
Phone: +370 41 500 333.
Fax: +370 41 500 336
E-mail: info@dailu.lt

Journal of Baltic Science Education, Vol. 18, No. 3, 2019

Uploaded by

Copyright:

Available Formats

Journal of Baltic Science Education, Vol. 18, No. 3, 2019

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Baltic Science Education, Vol. 18, No. 3, 2019

Uploaded by

Copyright:

Available Formats

Vol.18, No.

Published since 2002 Address:

ISSN 1648–3898 (Print) © Scientia Socialis Ltd., Lithuania, 2019

ISSN 2538–7138 (Online)

ISSN 2538–7138 /Online/

MENTORING ON SCIENCE TEACHER EDUCATION IN RUSSIA AND INTERNATIONAL PERSPECTIVES

THE EFFECT OF EARTHCOMM LEARNING MODEL AND SPATIAL THINKING ABILITY ON

THE EFFECT OF COMPUTER-AIDED 3D MODELING ACTIVITIES ON PRE-SERVICE TEACHERS’

FACTORS INFLUENCING THE FIRST YEAR STUDENTS’ ADAPTATION TO NATURAL

RESEARCH TRENDS AND ISSUES OF EDUCATION FOR SUSTAINABLE DEVELOPMENT-RELATED

LEARNING ENVIRONMENT, STUDENTS’ BELIEFS, AND SELF-REGULATION IN LEARNING PHYSICS:

A MODEL OF INTEREST IN STEM CAREERS AMONG SECONDARY SCHOOL STUDENTS

MONODISCIPLINARITY IN SCIENCE VERSUS TRANSDISCIPLINARITY IN STEM EDUCATION

THE EFFECT OF USERS’ BEHAVIORAL INTENTION ON GAMIFICATION AUGMENTED REALITY IN

EXPLORING CHANGES IN PRIMARY STUDENTS’ ATTITUDES TOWARDS SCIENCE, TECHNOLOGY,

INFORMATION FOR CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481

PROBLEMS OF PSYCHOLOGY IN THE 21ST CENTURY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482

NATURAL SCIENCE EDUCATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483

MENTORING ON SCIENCE TEACHER EDUCATION IN RUSSIA AND INTERNATIONAL

Muhammet Usak, Alfiya R. Masalimova

Nowadays, mentoring as an educational form of Russia’s occupational activity is undergoing fundamental

Received: March 20, 2019 Accepted: May 22, 2019

Muhammet Usak PhD, Associate Professor, Institute of Pedagogy and Psychology,

ON GEOGRAPHY LEARNING ISSN 2538-7138 /Online/

Abstract. Geography learning is

Table 1. Pattern of 2x2 factorial experimental.

High spatial Group 1 Group 2

Table 2. Selection of experimental classes and control classes.

Instrument and Procedures

Table 3. Distribution of research subject.

High spatial 64 57 121

Table 4. Normality tests.

Standardized Residual for Post .044 167 .200*

Table 5. Homogeneity test.

.881 3 163 .452

Table 6. Descriptive statistics.

High 20.14 3.01 64

Levene’s Test for

Figure 1. Geography learning outcomes based on spatial thinking ability.

Source Type III SS df MS F p

Corrected Model 851.946a 3 283.982 24.746 .0001

a. R Squared = .313 (Adjusted R Squared = .300)

Effects of Earthcomm Learning on Geography Learning Outcomes

Effect of Spatial Thinking Ability on Geography Learning Outcomes

Earthcomm Learning Interaction and Spatial Thinking Ability to

Received: February 04, 2019 Accepted: May 05, 2019

Muhammad Aliman PhD Student, Geography Education Department, State University

TEACHERS’ SPATIAL ABILITIES ISSN 2538-7138 /Online/

AND ATTITUDES TOWARDS 3D

Ali Ihsan Benzer, Abstract. The computer-aided 3D

Table 1. Course content.

Figure 1. The education framework.

ROT Pre-test 27 9.33 3.56

ROT Pre-test 28 9.14 3.52

AS3DM Pre-test 27 3.77 .48

Post Test-Pre Test N Mean Rank Sum of Ranks z p

Negative ranks 5 15.80 79.00

The Effect of 3D Modeling Activities on Attitude toward 3D Modeling

Received: February 07, 2019 Accepted: May 15, 2019

SCIENCE STUDIES IN HIGHER ISSN 2538-7138 /Online/