See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281929895

Structural Equation Model on Learners' Conceptions

of Learning and Approaches to Learning as Predictors
of Physics Self-Efficacy

Article · April 2014

DOI: 10.7718/iamure.ije.v10i1.885

CITATIONS READS
0 262

1 author:

Ray Butch D. Mahinay

Department of Education of the Philippines
13 PUBLICATIONS   9 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Project CARSEL (Comprehensive Alternative Resources for Self-directed Learning) View project

All content following this page was uploaded by Ray Butch D. Mahinay on 20 September 2015.

The user has requested enhancement of the downloaded file.

Vol. 10 March 2014 Vol. 10 · March 2014
Print ISSN 2244-1476 • Online ISSN 2244-1484
International Peer Reviewed Journal
doi: http://dx.doi.org/10.7718/iamure.ije.v10i1.885
This Journal is produced by IAMURE Multidisciplinary Research,
an ISO 9001:2008 certified by the AJA Registrars Inc.

Structural Equation Model on Learners’

Conceptions of Learning and
Approaches to Learning as Predictors
of Physics Self-Efficacy

RAY BUTCH D. MAHINAY

http://orcid.org/000-0001-8506-0712
raybutch.m@gmail.com
Mindanao University of Science and Technology
Cagayan de Oro City, Philippines

ABSTRACT

Many studies in the past have established the positive association of self-
efficacy to academic motivation and achievement. To substantively improve
the understanding of learner’s self-efficacy, education researchers must need to
explore how it is structured. This study proposed a structural model of Physics
self-efficacy (PSE) and outlined its relationships with conceptions of learning
Physics (CLP) and approaches to learning Physics (ALP) through structural
equation modeling. Three questionnaires from Tsai (2008) were adapted
for a field survey that was carried out to 317 fourth year students randomly
chosen among seven national high schools in the east district of the division of
Cagayan de Oro City, Philippines to measure the scales on PSE, CLP and ALP.
Correlation and path analyses showed that CLP has significant relationship with
ALP. Consequently, the ALP also has significant relationship to PSE. Specifically,
both the students’ low-level and high-level conceptions of learning physics exerted
positive influence to surface and deep motives and strategies in their approaches
in learning the subject. Further, the students’ ALP inferred direct contribution

111
IAMURE International Journal of Education

to PSE than the CLP. These results provide valuable information in this area
to effusively understand the nature and process of learning physics pertinent to
other metacognitive variables.

KEYWORDS

Science education, learning conception, learning approaches, physics

learning, self-efficacy, structural equation modelling, Philippines

INTRODUCTION

In the aim to raise learning performances in Philippine high schools, it has

been a constant effort in research to investigate on students’ aspects, dynamics
and influences to explain causations and effects of poor learning achievement.
The contention of these research are enthused by the results of the National
Achievement Test which is locally prepared and administered by the Department
of Education (DepEd) and the Trends in Mathematics and Science Study (TIMSS)
which is an international standardized testing on students’ achievement. For the
past years, results for both tests inferred poor condition of math and science
education.
This researcher being a Physics teacher for almost 10 years in DepEd realized
that it is motivating to study on how high school students view and learn
physics. In one institutional pre-test performance report conducted by Center
for Educational Measurement (2011) in Cagayan de Oro City public schools, it
was revealed that 68% of the students have a quality index of very poor and only
2% reached with a quality index of below average. This dismal situation needs
attention on strengthening Physics learning in high schools especially that it is
foundational to most college STEM degrees.
There are myriad of ways on how to explore learners’ factors toward learning
achievement as dealt with many researchers over the past millennia. One
remarkable study is on self-efficacy by Bandura (1977) which is about learner’s
beliefs in one’s capabilities to organize and execute course of action required to
produce given attainments. Self-efficacy provided strong foundational concept
to education researchers such that it has been generally correlated to academic
motivation and achievement (Schunk & Pajares, 2009). However, Usher and
Pajares (2008) claimed that in order to find out ways and enhance learner’s self-
efficacy, education researchers must need to know how it is formed and structured.

112
 Vol. 10 · March 2014

To explore the structural relationships within learner’s belief system of self-

efficacy, two learning constructs, namely, conceptions of learning and approaches
to learning can be adopted (Chiou & Liang, 2012). These two constructs are
like input-output variables in a student’s learning continuum. It befits this
researcher’s intention to ascertain how high school students view and learn
physics. Accordingly the learning constructs were specified to conceptions of
learning Physics and approaches to learning Physics. A structural equation model
is proposed to explain the interaction of learner’s conceptions of learning Physics,
approaches to learning Physics, and Physics self-efficacy.
This researcher contends that through this model, science educators can be
guided on designing more appropriate learning pedagogies to improve learner’s
self-efficacy and learning experience towards Physics. This will be most useful
especially in public high schools where the learning situation is most challenging.

FRAMEWORK

This study explored on the extent of relationships among the three variables,
namely: a) conceptions of learning Physics (CLP); b) approaches to learning
Physics (ALP); and c) Physics self-efficacy (PSE).

Conceptions of Learning Physics

The construct on conceptions of learning is treated as one’s beliefs and
understanding of the nature of learning (Chiou & Liang, 2012). It is formed from
a learner’s personal experience and is therefore a task- and domain-dependent.
Many researches into students’ conceptions of learning have indicated that
students conceive learning in qualitatively different ways (Purdie & Hattie,
2002). It is a strengthened notion that learners have varying conceptions of
learning towards different subject domains. Among the science areas, students’
conceptions of learning Biology may reveal different conceptions of learning in
Physics.
This study adopted the concept of Tsai (2004) on CLP to which he
dichotomized it into two broad categories: high level and low level. For the low
level category, the students’ conception of learning is represented by the following
subcategories: a) memorizing; b) testing; and c) calculating and practicing. It
denotes that these are passive and transmissive view of science learning. On the
other hand, the higher level category encompasses the following: a) increasing
in knowledge; b) applying; and c) understanding. It suggests that students view
learning science in an active and constructive view.
113
IAMURE International Journal of Education

Approaches to Learning Physics

A generic way of describing “what the student does” is precisely in terms
of their ongoing approaches to learning (Biggs, Kember & Leung, 2001). An
approach to learning describes the nature of the relationship between student,
context and task. Students may utilize different ways to study different academic
subjects; therefore the nature of student’s approaches to learning is also domain-
specific. This holds true on the science areas that student’s approaches to learning
Biology, Chemistry and Physics vary.
Based on studies on approaches to learning, researchers have identified two
approaches to learning science (Chiou & Liang, 2012) which are congruent to
the original concept of Biggs, Kember and Leung’s (2001) and Tsai’s (2008)
student learning processes, the deep and surface approaches. Deep approach is
characterized as an intrinsic motivation (deep motive) to actively comprehend
and integrate the new learning materials with existing ideas (deep strategy);
on the other hand, the surface approach is featured by its external motivation
(surface motive) to solely memorize or reproduce the learning materials (surface
strategy).
Specifically, deep motive is when students express intrinsic motivation in
learning as triggered by intense curiosity and interest; deep strategy is when
students utilize a more meaningful way to learn such as making connections
and extracting key points. Conversely, surface motive is when students possess
extrinsic motivation to learn such as passing an exam and pursuing a high grade;
surface strategy is when students use more rote-like strategies to learn such as
unreflective memorization.

Physics Self-Efficacy
Self-efficacy plays a crucial role in determining various aspects of human
behaviors, such as the choice of goals and strategies, the effort needed to achieve
the goals, and the persistence when confronting obstacles (Bandura, 1997).
It is rooted on Social Cognitive Theory (Bandura, 1997) that conjectured on
individual’s beliefs as agent to one’s behavior.
In educational settings, studies have shown that self-efficacy exerts a positive
effect on their learning goals, academic achievements and career choices (Chiou
& Liang, 2012). As already mentioned, self-efficacy is domain-specific and task-
dependent. So like with the two aforementioned constructs, self-efficacy for this
study is accordingly pointed to one content domain which is Physics.

114
 Vol. 10 · March 2014

Interaction of CLP, ALP and PSE

Bandura (1997) and Schunk and Pajares (2008) purported that science self-
efficacy is affected by both a learner’s pre-task personal characteristics and during-
task learning behaviors that conclusively determine his academic performance.
Conceptions of learning refer to an individual’s belief in learning (Chiou & Liang,
2012) and can stand as a pre-task factor. Whereas approaches to learning denote
how a learner performs learning tasks (Biggs, Kember & Leung, 2001) and this
can represent during-task behavior. With this in a Physics learning continuum,
the CLP is a pre-task construct and in the same manner that the ALP a during-
task construct.
Furthermore, the relationships between self-efficacy and variables of interest
are reciprocal in nature (Bandura, 1997). The students’ pre-existing self-efficacy
may influence their learning motivations, behaviors and performances, their
interpretation of the resultant learning experiences and outcomes will in turn
affect their self-efficacy, and it creates a cyclical pattern in the process.
Implicitly both CLP and ALP are strong variables that can influence the
formation of PSE. It has been elaborated richly in literature that studies on self-
efficacy are suggested to focus on a specific domain, to explore the criteria that
students use to interpret their learning experiences, and to find out the essential
predictive variables that may involve in the causal, reciprocal structure of self-
efficacy (Chiou & Liang, 2012). Hence, Physics is consistently used as a content
domain with the two important academic constructs, CLP and ALP, which are
supposed to serve as measures in the formation of PSE.

OBJECTIVES OF THE STUDY

This study aims to investigate the structure of Physics self-efficacy (PSE)

and delineate its relationships with conceptions of learning Physics (CLP) and
approaches to learning Physics (ALP) among the senior public high school
students through structural equation modelling.

METHODOLOGY

Research Design
This study employed quantitative methods of research. It is descriptive-
correlation in design which involved a survey in the public secondary schools in
Cagayan de Oro City for the first semester of the school year, 2013-2014.

115
IAMURE International Journal of Education

As a theory-generating study, structural equation modeling (SEM) was also

done. The SEM as an approach was utilized to examine the patterns of correlation
and/or covariance of CLP, ALP and PSE.

Respondents
The respondents of the study were 317 fourth year public high school students
from the East II district schools in the DepEd division of Cagayan de Oro City.
As prescribed by the Basic Education Curriculum of DepEd, these students focus
on Physics as their science subject. Among the respondents, 141 were males and
176 were females. Their ages ranged from 16-20 years old. These respondents
came from different socio-economic classes and academic achievement groups.

Instruments
This study used the following survey questionnaires of Tsai (2008) to gather
the necessary data: a) Self-Efficacy in Learning Science Survey; b) Conceptions
of Learning Questionnaire; and c) Questionnaire on Approaches to Learning.
Substantial revisions were made to suit the instruments to local conditions
such as the use of language, grammar and syntax, and the curricular set-up of
DepEd. All the tools which used a 5-point Likert scale ranging from strongly
agree (5 points) to strongly disagree (1 point) demonstrated statistical validity
and reliability. The values of the Cronbach’s alpha for the said instruments are as
follows: 0.84, 0.87 and 0.76.

Statistical Tools
Descriptive and inferential statistics were done using various statistical software
(i.e. MS Excel, Minitab 16 and IBM SPSS Amos 19) with the confidence level
at 95%. SEM techniques like the regression and path analyses were both used to
test the relationships of the measured variables. Model evaluation through several
measures of goodness-of-fit is the final step in structural equation modeling
(Arbuckle, 2010). Chi-square (ratio of χ2/df that is less than 3.0 indicates good
fit), normed fit index (NFI, 1.0 indicates good fit), comparative fit index (CFI,
1.0 indicates good fit), goodness-of-fit index (GFI, 1.0 indicates good fit), and
root mean square residual (RMR, 0 indicates good fit) were used to assess overall
fit of the model produced by the SEM tools.

116
 Vol. 10 · March 2014

RESULTS AND DISCUSSION

Conceptions of learning physics (CLP) represent student’s beliefs and

understanding of the nature of learning Physics. In physics learning continuum,
CLP is regarded as the pre-task factor. Students have varying level of CLP for
every subscale as shown in Figure 1.
Among the CLP subscales, calculating and practicing has the highest
representation (=4.01) and testing has the least (=3.46). It can be inferred that
most senior students conceive physics learning as practicing calculation and
solving problems. It is a common scenario for students to relate that physics
always entail numbers. Given such circumstance, they may have perceived
anxiety on the subject hence the low representation of the testing subscale.
It should be noted that Tsai (2004) classified CLP it into two broad categories:
high level and low level. Low-level CLP is characterized by the following subscales:
a) memorizing; b) testing; and c) calculating and practicing. Collectively, low-
level CLP has a weighted mean of 3.72 which is described as high. Whereas
high-level CLP is represented by the following subscales: a) increasing one’s
knowledge; b) applying; and c) understanding. If taken together, high-level CLP
has a weighted mean of 3.77 which is also described as high.

Figure 1. Students’ Levels on Conceptions of Learning Physics

117
IAMURE International Journal of Education

To complete the Physics learning continuum is the during-task factor:

the students’ approaches to learning Physics (ALP). It denotes how a student
performs learning tasks in Physics. It is shown in Figure 2 that students have
varying level of ALP for every subscale.

Figure 2. Students’ Levels on Approaches to Learning Physics

On the four ALP subscales, surface motive has the highest mean (=3.81). It can
be assumed that most of the students’ motivation to learn physics is reinforced by
external bases such as satisfying the teacher’s expectations or pleasing the family.
Anyhow, the students’ approaches to learn Physics is more brought upon by their
high deep strategy (=3.66). It meant that in the Physics subject, they actively
comprehend and integrate the new learning materials with existing ideas.
With regards to the dichotomy of Tsai (2008), ALP is manifested by surface
and deep approaches. The weighted mean for surface ALP is 3.53 and for the
deep ALP is 3.57 to which both are interpreted as high.
There is constancy among the levels of the three variables as can be gleaned
in Table 1. The high CLP of the students led them to have high ALP which
consequently are factors for them to have high PSE. The interaction of CLP,
ALP and PSE is grounded to Bandura’s (1997) assumptions that the students’
interpretation of the resultant learning experiences and outcomes will affect their
self-efficacy.

118
 Vol. 10 · March 2014

Table 1. Descriptive values on the students’ CLP, ALP and PSE

Standard Qualitative
Variables Mean
Deviation Description
Conceptions of Learning Physics (CLP) 3.74 0.48 High
Approaches to Learning Physics
3.55 0.46 High
(ALP)
Physics Self-Efficacy
3.42 0.49 High
(PSE)

The correlation matrix in Table 2 gives basis that significant relationships exist
to most of the subscales on the students’ CLP, ALP and PSE at 0.05 alpha level.
In general, CLP is significantly correlated to ALP (r=0.56, p=0.00). Specifically,
low-level CLP is positively correlated to surface ALP except for the subscale on
calculating and practicing and to deep ALP except for the subscale on testing.
High-level CLP is also positively correlated to both surface and deep ALP.
Furthermore, CLP is significantly correlated to PSE (r=0.45, p=0.00) as
ALP is likewise significantly correlated to PSE (r=0.49, p=0.00). This reinforced
the concept that relationships between self-efficacy and variables of interest are
reciprocal in nature (Bandura, 1997) as documented in modern studies of Tsai
(2008), Usher and Pajares (2008), Lynch (2010), Chiou and Liang (2012) and
Chan (2012).

Table 2. Correlation matrix among the subscales on the students’ CLP, ALP and
PSE
Physics
Surface Surface Deep Deep
Self-
Motive Strategy Motive Strategy
Efficacy
Memorizing 0.415* 0.075 0.289* 0.457* 0.413*
Testing 0.143* 0.117* 0.057 0.176* 0.199*
Calculating & Practicing 0.378* 0.032 0.282* 0.514* 0.349*
Increasing One’s
0.390* 0.115* 0.244* 0.471* 0.269*
Knowledge
Applying 0.507* 0.159* 0.340* 0.563* 0.357*

Understanding 0.485* 0.093* 0.361* 0.613* 0.382*

Physics Self-Efficacy 0.419* .096 0.415* 0.508* ---

Note: *p<.05

119
IAMURE International Journal of Education

Figure 3 is the structural model of the interaction of CLP and ALP (the
exogenous or predictor variables) to PSE (the endogenous or response variable).
This hypothesized model has the best fit measures (=49.68, p=0.00; CFI=0.951;
GFI=0.959; NFI=0.937; RMR=0.17).
The parameter estimates between CLP and ALP to their indicators generally
revealed significant positive effects. Also, the estimate of covariance between CLP
and ALP is 0.08 which is significantly different from zero at the 0.001 level. This
firmed up the proposition that CLP and ALP are good pre-task and during-task
factors to constitute a physics learning continuum.

Figure 3. The structural model on the interaction of CLP, ALP and PSE

Furthermore, parameter estimates showed that CLP have negative effect to

PSE (= -0.02). This established that the students’ CLP may inhibit their PSE in
a way that these students are apprehensive about numbers. It can be recalled that
their highest perception of physics learning is about calculating and practicing.
Conversely, the students’ ALP manifested stronger role in the development of
PSE (=1.53). Students’ confidence in learning Physics is more likely to develop
with their approaches to learning the subject. This accords to Usher and Pajares
(2008) that mastery experiences had been repeatedly confirmed as the major
source of self-efficacy. In line to Pajares (2002), students interpret the results of
their actions and use these interpretations to develop their beliefs about their
capabilities and then act in concert with the beliefs they created.
This study covered one district area of the Division of Cagayan de Oro City
so therefore the results should be interpreted with caution when generalizability
120
 Vol. 10 · March 2014

is projected. Since this study concerns with students’ beliefs, mixed methods
research triangulation could strengthen the framework and direction of this
study. Anyhow, this is an initial effort in using structural equation modeling in
exploring students’ physics learning in Philippine setting.

CONCLUSIONS

Students held a) high beliefs and understanding in the nature of learning

physics; b) high beliefs on their ongoing approaches to learning physics; and d)
high level of confidence that that they can carry out tasks in learning physics.
It is established that the students’ conceptions of learning physics and their
approaches to learning physics are both elemental in physics learning continuum
just like a pre-task and during-task system. However, it was depicted through
structural equation modeling that the students’ approaches to learning physics
contribute significantly in the formation of their physics self-efficacy than their
conceptions.

RECOMMENDATIONS

The outcome of this study proposes concrete suggestions in the curricular and
instructional make-up in DepEd. A supportive physics learning environment
that encourages self-appraising mechanisms is most practical for students so that
they can assess their own learning. In the line of Pajares (2002), students interpret
the results of their actions and use these interpretations to develop their beliefs
about their capabilities and then act in concert with the beliefs they created.
Besides, it should be always remembered that one of the principles of learning is
that students should keep track of their own achievement.
This impels a message to the international physics teaching community that
a constructivist curriculum which encourages active assignation of the students
in the teaching-learning process must be in place. This is to develop students’
deep motives and approaches to learning physics. Like in a classroom setting, a
teacher can introduce problem-based or project-based activities so students can
enthusiastically engage in making meaning of the experience and in the process
fulfill their own interest and motives in learning physics. All of these are relevant
to the social cognitive theory of Bandura (1986) highlighting the notion that
humans learn from experience, then they self-reflect and self-regulate.

121
IAMURE International Journal of Education

LITERATURE CITED

Arbuckle, J. (2012). IBM SPSS Amos 19 User’s Guide. Illinois: Amos Development
Corporation. Retrieved on August 23, 2013 from http://goo.gl/ZzPb0P

Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: Freeman.
Retrieved on September 15, 2013 from http://goo.gl/pqo1Do

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.

Psychological Review, 84, 191-215. Retrieved on July 16, 2013 from http://
goo.gl/jeR3V6

Bandura, A. (1986). Social foundations of thought and action: A social cognitive

theory. Englewood Cliffs, NJ: Prentice Hall. Retrieved on July 20, 2013 from
http://goo.gl/cTSQ7G

Biggs, J.B., Kember, D., & Leung, D.Y.P. (2001). The Revised Two Factor Study
Process Questionnaire: R-SPQ-2F. British Journal of Educational Psychology,
71, 133-149. Retrieved on August 12, 2013 from http://goo.gl/T0Yr9R

Center for Educational Measurement (2011). Institutional Pre-Test Performance

Report for Science and Technology IV. Taken from the Department of
Education-Division of Cagayan de Oro City.

Chan, K. W. (2012). Hong Kong Teacher Education Students’ Epistemological

Beliefs and their Relations with Conceptions of Learning and Learning
Strategies. The Asia-Pacific Education Researcher, Vol 16:2, pp.200-214.
Retrieved on July 17, 2013 from http://goo.gl/4QLcW0

Chiou, G. L. & Liang, J. C. (2012). Exploring the Structure of Science Self-

efficacy: A Model Built on High School Students’ Conceptions of Learning
and Approaches to Learning in Science. The Asia-Pacific Education Researcher,
Vol 21:1, pp.83-91. Retrieved on July 23, 2013 from http://goo.gl/OGHs4E

Lynch, D. (2010). Motivational Beliefs and Learning Strategies as Predictors of

Academic Performance in College Physics. College Student Journal, 44.4: 920-
927. Retrieved on March 28, 2013 from http://goo.gl/WwG15s

122
 Vol. 10 · March 2014

Schunk, D. & Pajares, F. (2009). Self-efficacy theory. Handbook of Motivation at

School. New York, NY: Routledge. Retrieved on July 21, 2013 from http://
goo.gl/KW36xb

Tsai, C. (2004). Conceptions of Learning Science among High School Students

in Taiwan: a Phenomenographic Analysis. International Journal of Science
Education, 26, 1733-1750. Retrieved on July 29, 2013 from http://goo.
gl/0cstL1

Usher, E. L. & Pajares F. (2008). Sources of Self-efficacy in School: Critical review

of the literature and future directions. Review of Educational Research, Vol 78,
751-796. Retrieved on August 9, 2013 from http://goo.gl/cQzLzx

Pajares, F. (2002). Overview of social cognitive theory and of self-efficacy. Retrieved

on August 9, 2013 from http://goo.gl/RWuIYg

Purdie, N. & Hattie,J. (2002). Assessing Students’ Conceptions of Learning.

Australian Journal of Educational and Developmental Psychology, Vol 2, pp
17-32. Retrieved on July 11, 2013 from http://goo.gl/NyGUWQ

Indexed by:

123

View publication stats