Eurasia Journal of Mathematics, Science & Technology Education, 2016, 12(2), 303-320

Attitude toward Science

Teaching of Spanish and
Turkish In-service Elementary
Teachers: Multi-group
Confirmatory Factor Analysis
Fikret Korur
Mehmet Akif Ersoy University, TURKEY
Rocío Vargas Vargas & Noemí Torres Serrano
University of Almería, SPAIN

Received 20 April 2015 Accepted 24 August 2015  Published online 28 Oct 2015

Elementary school teachers’ having a positive attitude toward science teaching might
encourage students to develop positive attitudes toward science learning. This cross-
cultural study aimed to validate the seven-factor structure of the Dimensions of Attitude
toward Science (DAS) scale by applying it in two countries. Moreover, it aimed to
determine the percentage distributions of in-service elementary school teachers’
attitudes on the confirmed seven categories. The DAS scale was translated into Turkish
and Spanish and administered to 185 teachers in Burdur and 202 teachers in Almeria.
Multi-group confirmatory factor analysis confirmed that the factor structures of the
translated scales for the two countries correspond to the same seven-factor structure by
fitting the theoretical model. The DAS scale was compatible for both countries at an
adequate level. In both samples, most of the teachers were against gender stereotyping in
the science teaching/learning process, and they indicated they did not have any anxiety,
fear, or stress toward teaching science. The crucial difference between the two samples
was seen in the teachers’ responses to items measuring “Enjoyment” and “Perceived
Dependency Context Factors.”

Keywords: attitudes toward science teaching, in-service elementary teachers, multi-group

confirmatory factor analysis, Turkish attitude scale, Spanish attitude scale

INTRODUCTION

Attitude is defined as “choosing personal actions based on internal states of

understanding and feeling” (Driscoll, 2000, p. 350). To improve the quality of
education, it is not sufficient to specify teaching methods with certain activities; it is
also necessary to consider teachers’ feelings, motivations, and, especially, attitudes.
Teachers with negative or less positive attitudes toward science teaching are less
likely encourage their students to develop positive attitudes toward science (Jarvis

Correspondence: Fikret Korur,

Department of Elementary Education, Faculty of Education, Mehmet Akif Ersoy
University, , 15030, Burdur, Turkey.
E-mail: fikretkorur@mehmetakif.edu.tr
doi: 10.12973/eurasia.2016.1215a

Copyright © 2016 by iSER, International Society of Educational Research

ISSN: 1305-8223
F. Korur et. al

& Pell, 2004; Osborne, Simon, & Collins 2003; van State of the literature
Aalderen-Smeets & Walma van der Molen, 2013).
Because they usually have low level of self-efficacy  Teachers with negative or less positive
(Tosun, 2000) or they could not use time effectively attitudes toward science teaching are less
to teach and discuss science in the class (Harlen & likely encourages their students to develop
Holroyd, 1997; Jarvis & Pell, 2004). On the other positive attitudes toward science.
hand, teachers with positive attitudes toward  It is crucial to examine science teachers’
science have greater self-efficacy and self- attitudes across various countries to
confidence to improve their students’ attitudes determine whether local needs and demands
(Bursal, 2010; Osborne et al., 2003). Therefore, are the same in the larger population.
elementary school teachers’ attitude toward science  Several scales developed previously included
teaching should be carefully identified, since the and measured several attitude structures
progress in science lesson mostly belong teachers’ together. It is better to measure them on a
values and attitudes toward teaching science single scale. Further, many existing scales
(Blalock et al., 2008; Haney, Czerniak, & Lumpe, designed to measure attitude toward science
1996; Osborne et al., 2003; Saad & BouJaoude, 2012; teaching lack of statistical (e.g., reliability of
Yilmaz-Tuzun, 2008). The definitions and results) and theoretical standard (e.g.,
attributions of subject of attitude toward teaching construct validity).
science were not explicitly analyzed in many studies,
Contribution of this paper to the literature
as they are far from distinguishing the subject from
other measures like motivation (Blalock et al., 2008;  In both countries, teachers possess high self-
van Aalderen-Smeets & Walma van der Molen, 2013; efficacy and this could be the reasons for the
van Aalderen Smeets, Walma van der Molen, & Asma, low rate of anxiety.
2012). For this purpose, Dimensions of Attitude  The DAS scale is a new, coherent, and effective
Toward Science (DAS) scale was revised by van in assessing teachers’ attitudes toward
Aalderen-Smeets and Walma van der Molen (2013) science teaching since items measuring
to measure pre-service and in-service teachers’ “attitude toward science” and “science
attitude toward science teaching. In this study, at attitude,” were removed. The Turkish and
first the DAS was translated into Spanish and Spanish versions of the DAS scale appear to
Turkish and structure of translated versions was represent valid and reliable measures in both
validated with respect to original structure. Second, samples.
the variance in the factor structure across both  The cross-cultural adaptation of the DAS scale
samples was evaluated. Finally, the data collected to might have led the further studies to combine
confirm the scale was assessed to determine teachers’ attitude toward science teaching
elementary in-service teachers’ perceptions for with the findings of other international
attitudes toward science teaching in both samples. studies such as TIMSS and PISA-Science.
Conceptual framework
In this particular study, the new theoretical framework for primary teachers’
attitude toward science, which corresponds personal attitude dimension, and the
teaching of science, which corresponds professional attitude dimensions was adopted
from the study of van Aalderen-Smeets et al. (2012, pp. 154-172). The general schema
of the framework is given in Figure 1.
In general, both dimensions of attitude include the same sub-dimensions as
cognition, affect, self-efficacy and behavioral intention. At first, the cognitive beliefs
included three cognitive attributes, as: a) ‘perceived relevance’ that is related to
importance and relevance of science and science teaching in primary schools. b)
‘perceived difficulty’ that is related to general beliefs about difficulty of science
relative to other disciplines and difficulty of teaching science. The items correspond
to beliefs of a teacher that she/he think ‘average’ teacher has regarding teaching
science (van Aalderen-Smeets & Walma van der Molen, 2013). c) ‘gender beliefs’ that
is related to various abilities of boys and girls in science and potential ability
differences between male and female teachers with respect to teaching science and

304 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

Figure 1. Theoretical framework for the seven-factor structure of the DAS Scale
(adapted from van Aalderen-Smeets et al., 2012).
their enjoyment in doing so. Indeed, some researchers claimed that developing an
attitude is directly related to developing cognitive beliefs (Tosun, 2000; Türkmen,
2013).
Secondly, affective states included ‘enjoyment’ and ‘anxiety’. Enjoyment is related
to positive emotions regarding science in real life, science related activities, and
science teaching. Enjoyment mainly depends on self-efficacy, which is linked with job
satisfaction and job stress (Klassen & Chiu, 2010). Anxiety, corresponds to negative
and unfavorable feelings regarding not only science related topics and activities but
also science teaching. To reduce anxiety in teaching science courses, and to make
students develop positive attitudes toward science, teachers should take courses
including true science applications, such as activities and experiments (Martin-
Dunlop & Fraser, 2008; Lin, Cheng, Chang, Li, Chang & Lin, 2014).
The third component of attitude toward teaching science was perceived control
that included ‘self-efficacy’ and ‘context dependency’. Self-efficacy includes both
views and feelings of personal capability to exhibit specific behavior (Bandura, 1997).
Elementary school teachers have higher levels of self-efficacy because of the contexts
in which they work (Klassen & Chiu, 2010). In this study, it is used to express in-
service teachers’ beliefs in their ability to succeed in science and their capability to
teach science. The next attribute, context dependency, is mostly related to context
factors of professional attitude toward science (van Aaalderen-Smeets et al., 2012).
They indicated four main context factors that are collegial support, lack of resources,
time allocated for science in the curriculum, and the time and effort needed to prepare
science lessons. The perception of these factors by the in-service teachers reflects
their beliefs that several external factors make their science teaching hard or easy.
Context factors affect attitudes toward science / science teaching (Ergül, 2009;
Martin-Dunlop & Fraser, 2008; van Aalderen-Smeets & Walma van der Molen, 2013).
The final component of attitude toward science is ‘behavioral intention’ that
measures someone’s behavior related to science mostly in daily life like reading a
science book and watching scientific programs. In terms of attitude toward science
teaching, behavioral intention includes teachers’ behaviors either finding extra hours
to teach science or willing to do science lessons more than other lessons like
mathematics (Van Aaalderen-Smeets et al., 2012). The analysis of gender stereotype
approaches along with self-efficacy approaches presents “Percevied Difficulty” and
“Gender Beliefs” were misevaluated or not sufficiently understood by the participant
teachers (van Aalderen-Smeets & Walma van der Molen, 2013). Therefore, they

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 305
F. Korur et. al

suggested that in future studies, these two dimensions should be evaluated in greater
depth, along with their relationship with other dimensions.
Teachers’ attitudes toward science teaching in Turkey and Spain
In the report of Education, Audiovisual and Culture Executive Agency-EACEA
(2011) that revealed how the science education in Europe is organized, average
performances of participants from some countries such as Turkey and Spain were
found to be below the European Union average. In this report it is recommended that,
the reasons for decrease in achievement, interest and attitude toward science should
be researched. Moreover, Štefaniková & Prokop (2015) indicated that there is a
decreasing interest toward learning science and students usually have negative
attitudes toward science from various cultures. It is clear that the attitudes of teachers
toward science teaching influence students’ attitudes as well as their academic
achievement (Bilen & Köse, 2012; Türkmen, 2013; Yakar & Baykara, 2014; Yaşar &
Anagün, 2008). In this context, it is crucial to determine perceptions of elementary in-
service teachers for attitudes toward science teaching, specific to Spain and Turkey.
In this particular study, the Spanish and Turkish versions of the current DAS scale
contribute to analyzing elementary school teachers’ attitude toward science teaching
in detail.
In Turkey, teacher candidates had moderate-level attitude scores and did not
perceive themselves as enthusiastic about teaching science subjects (Bursal, 2010;
Duban & Gökçakan, 2012; Türkmen, 2013). In Spain, in-service science teachers’ held
moderate attitudes toward the teaching science profession (Guisasola, Robinson, &
Zuza, 2007) and prospective elementary school teachers on the other hand had a
rather positive attitude toward teaching science related to which self-efficacy beliefs
or attitudes toward science / science teaching (Brígido, Borrachero, Bermejo, &
Mellado, 2013).
Effective instruction mostly depends on self-efficacy, and enjoying the job (Azar,
2010; Bursal, 2010; Ergül, 2009; Guisasola et al., 2007; Tekkaya, Çakıroğlu, & Özkan,
2004). Klassen and Chiu (2010) found that teachers with high self-efficacy reported
high levels of job satisfaction. Further, teachers reported low job satisfaction because
they had high levels of stress. According to The Organization for Economic Co-
operation and Development (OECD) report, teachers from a few countries (including
Turkey and Spain) reported comparatively weak self-efficacy, and teachers’ job
satisfaction usually increases with tenure (OECD, 2009).
The two countries have similar elementary school teacher training programs as
pre-service teachers must take 60 credits and 25 hours per credit point. Further, both
countries participate in the ERASMUS student exchange program. In Turkey and
Spain, students begin elementary school in the calendar year of their sixth birthday,
and education is free of charge. In Turkey, science lessons start in third or fourth
grade (at the age of 10 to 11), which corresponds to the second cycle in Spain.
However, in Spain, science lesson start in the first cycle (at the age of 6 to 8).
In both countries, students take Program for International Student Assessment
(PISA) and Trends in International Mathematics and Science Study (TIMSS) exams.
More than 80% of the students who took the PISA in Turkey agreed that “Science and
Technology teaching is important for society” (International Council for Science
[ICSU], 2011, p. 13). In the PISA-Science section, although students in Turkey held
more positive attitudes than their counterparts in Spain, their achievement in PISA-
2012 was lower. These results raise the query whether the current variance among
students’ attitudes are similar to the elementary school teachers’ attitudes toward
science teaching in the two countries. Therefore, it is paramount to admit worldwide
standards for effective science teaching (PISA, 2012) and to analyze teachers’

306 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

attitudes toward teaching science that is one of the factors affecting students’
attitudes toward learning science.
Studies of pre-service elementary school teachers’ attitudes and self-efficacy in
Turkey and Spain indicated the teachers did not enjoy studying science, but they felt
competent to teach it, meaning they had adequate self-efficacy (Brígido et al., 2013;
Bursal, 2008; 2010; Ergül, 2009; Tosun, 2000). The review of the studies in Turkey
and Spain that measure the attitudes of in-service elementary teachers toward
science teaching generally focus on self-efficacy, job satisfaction and enjoyment. The
aim of the current study is to analyze the attitudes even for the other sub-dimensions
discussed in the framework.
The scales measuring attitudes toward science
Researchers that measured attitudes toward science teaching used various scales
in previous studies. Türkmen (2013), who also conducted some other longitudinal
studies of attitudes toward science, adapted The Science Teaching Attitude Scale II
from Moore and Foy (1997) with 95 in-service elementary and 38 in-service science
teachers. The scale included items measuring general attitude toward science lessons
rather than attitude toward science teaching.
The mostly used instrument was the Attitude Scale Toward Science Teaching,
which was adapted into Turkish by Özkan, Çakıroğlu, and Tekkaya (2002) based on
Thompson and Shrigley (1986). Thompson and Shrigley’s original scale includes 22
items and it divides attitude into four sub-dimensions. It is adapted into various
cultures and used in many studies. However, it was prepared for pre-service teachers
and lacks items that directly measure elementary school teachers’ attitudes toward
science teaching (Tosa, 2009). In addition, van Aalderen-Smeets and Walma van der
Molen (2013) indicated that it included several attitude structures like self-efficacy
related to science, feeling toward science, and self-efficacy of science teaching
components. They recommended that it is better to measure them with different
subscales than to represent them on a single scale. Osborne et al. (2003), Blalock et
al. (2008) and Gardner (1995) emphasized that many existing scales designed to
measure attitude toward science teaching lack of statistical (e.g., reliability of results)
and theoretical standard (e.g., construct validity). In this sense, as the result of need
for a new and up-to-date scale that would measure the attitude toward science
teaching, van Aalderen-Smeets and Walma van der Molen (2013) revised the DAS
scale, which was first developed by van Aalderen-Smeets et al. (2012). They offered
to adapt the scale to the other cultures. Thus, this particular study provides Spanish
and Turkish adapted versions of the DAS scale.
In their study, van Aalderen-Smeets and Walma van der Molen (2013) gave the
Dutch version of the DAS scale to 158 in-service and 398 pre-service teachers and
found inter-item correlations from .39 to .59. Their explanatory factor analysis with
direct oblimin rotation revealed that the items were loaded on seven factors with
loadings between .37 and .90.
Van Aalderen-Smeets and Walma van der Molen (2013) also applied a multifaceted
and atypical confirmatory factor analysis (CFA) including various steps beginning
with principal axis factoring to confirm the seven-factor structure. The reliability
coefficients (Cronbach’s alpha) for the seven categories ranged from .74 to .93. The
inter-item correlation coefficients among the sub-categories ranged from .44 to .85.
Sixty nine percent of total variance was explained with the seven-factor structure. Van
Aalderen-Smeets and Walma van der Molen also validated the scale’s construct and
face validity. They proposed that with 28 items and its seven-factor structure, the DAS
scale can thoroughly measure elementary school teachers’ attitudes toward science
teaching.

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 307
F. Korur et. al

Based on the findings of the detailed review, in-service teachers’ attitudes toward
science teaching should be identified in Spain and Turkey, since the comparison is
valuable to support various related findings in global reports (e.g., PISA, ICSU).
Moreover, cross-cultural studies, such as Campbell, Medina-Jerez, Erdogan, and
Zhang (2010), suggest that it is crucial to examine science teachers’ attitudes across
various countries to determine whether local needs and demands are the same in the
larger population. Further, van Aalderen-Smeets and Walma van der Molen (2013)
recommended comparing the findings of applications of the DAS scale in different
languages and countries with the results of their study. Therefore, this study adapted
the scales into Turkish and Spanish. The multi-group CFA is applied in order to
validate the DAS scale with proposed seven-factor structures for two countries and to
confirm that the model is invariant across groups. Bryne (2010) stated that, this
purpose was one of the five purposes defined in several studies that aimed to seek
evidence of multi-group equivalence. The two main research questions of this study
are as follows:
1) Do the adapted scales confirm the originally implemented seven-factor
structure of the DAS scale through their application in two groups, Burdur,
Turkey and Almeria, Spain? More specifically, the answer to this question was
analyzed through two hypotheses:
Hypothesis 1: The multi-dimensional structure of the DAS comprising
self-efficacy (SE), perceived relevance (PR), gender stereotyping (GS),
anxiety (A), difficulty of science teaching (DST), perceived
dependency on context factors (PDC), and enjoyment (E) is valid
across both groups.
Hypothesis 2: This seven-factor structure of the DAS is invariant
across the two groups.
2) What are the percentage distributions of in-service elementary school
teachers’ attitudes on the confirmed seven categories for both groups?

METHOD

This investigation is a comparative survey that uses a questionnaire implemented

in two countries (as discussed by Klandermans & Smith, 2002) to identify and
compare the interests, abilities, attitudes, or competencies of respondents (Fraenkel,
Wallen, & Hyun, 2012). The findings are mostly analyzed descriptively.
Participants and sampling procedure
The study population is about 1000 in-service elementary school teachers in
Burdur, Turkey and approximately 2900 in-service elementary school teachers in
Almeria, Spain. The sample consisted of 185 (83 female) teachers in Burdur and 202
(94 female) teachers in Almeria who voluntarily filled out the scale. The data
collection process took 4 weeks in Burdur and 8 weeks in Almeria. In Burdur, the
largest cohort of teachers had 21–30 years of teaching experience (117 teachers; 63.2
%); it was 1–10 years in Almeria (121 teachers; 59.9 %). In Burdur 11% and in
Almeria 12 % of the teachers had an MS degree. Moreover, 62% and 71% of the
Burdur and Almeria teachers, respectively, had graduated from the university in their
city.
Data collection instrument
The DAS scale is based on the theoretical framework of van Aalderen-Smeets et al.
(2012). The reliability and validity analyses were carried out by van Aalderen-Smeets
and Walma van der Molen (2013).
Given the DAS scale’s development procedure and unique structure, the
researchers of this particular study decided to translate it for the Turkish and Spanish

308 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

contexts after gathering written permission from the developers. It is a challenging

task to translate and to ask questions about attitudes (Behling & Law, 2000).
Therefore, the process was carried out thoroughly by analyzing the studies related to
translation of scales (Beaton, Bombardier, Guillemin, & Ferraz, 2000; Byrne, 2008;
Gierl, 2000; Küçüközer, Kırtak-Ad, Ayverdi, & Eğdir, 2012). In these studies, mostly a
five-step guideline including translation, review, back translation, pretesting, and
documentation was used. Therefore in this study a) the DAS scale was translated into
target languages by two independent bilingual translators. b) A third bilingual
language experts checked the translated scales and obtained one form scale. c) The
scales were subjected to a blind back translation. d) The back translated version was
analyzed and compared with the source scale. There was not any critical ambiguity in
terms of both understanding the language and methodology of the items for both
languages. In the original form of the scale two items include examples given in
parenthesis “Techniektorens” and “Natuiek, Leefwereld” were particular to the
source language and they were cultural terms. The experts concluded that the item
could be understood clearly without these terms, so they were removed from both
translations. Each version of the DAS scale also included 28 items. Specific to this
study, there were seven demographic questions about the participants’ age, gender,
and teaching experience. e) Then the scale in the source language was administered
to bilingual samples twice and almost one month apart. The correlation coefficients
among total scores of the source and target language applications was r=.817 and
r=.784, for Burdur and Almeria samples respectively. All of this five-step procedure
was used to verify language (semantic and conceptual) equivalence of the translated
scales. Two language experts—academicians and one MS student supported the final
scale’s face validity. All of the procedures to answer the first research questions of this
study including both reliability analysis and the CFA to evaluate the model fit can also
be an evidence to validate construct equivalence of the translated versions. The
Turkish and Spanish versions of the DAS scale are shown in Appendix a and b,
respectively.
Data analysis
The score for teachers’ attitudes toward science teaching were assigned the
following codes: 1 “Totally disagree,” 2 “Disagree,” 3 “Neutral,” 4 “Agree,” 5 “Totally
agree.” On the five-point Likert type scale, higher scores are interpreted as more a
positive attitude. The first research question of this study was analyzed with the CFA
and multi-group CFA with AMOS to determine the model fit of the scales. Before
conducting analysis, it is checked (measure for outliers is Mahalanobis distance at
p<.001, Tabachnick and Fidell, 2007, p.99) and validated that there is no outlier. The
linear regression-curve estimation was conducted for all relationships of the model
and determined that they were seems to be linear. The shape of the scatter-plot
diagrams was also analyzed for homoscedasticity and this assumption was also
validated. For the normality, as Byrne (2010) suggested at first univariate normality
(skewness and kurtosis values are within the range of -3 and +3) was validated. For
the multivariate normality, all of the critical ratio values for both samples were
analyzed and they are below 5.00, indicating that the data normally distributed. The
CFA process is a theory-testing statistical technique that begins with the hypothesis
that a relationship exists between the observed variables and their underlying latent
constructs (Child, 1990).
The model fit was tested for the theoretical structure included seven factors; as
Figure 1 presents. At first, the CFA was conducted to the answer to test the first
hypothesis by running the original seven-factor structure model namely Model7

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 309
F. Korur et. al

Table 1. Fit indices with suitable thresholds used in this study

Fit Index Critical Values Resource(s)
χ2 (CMIN) Low χ2 value and Brown (2012)
p > .05 Tabachnick and Fidell (2007)
Bryne (1989)
If p < .05 Hu and Bentler (1999)
χ2/df Good Fit χ2/df < 1 Byrne (1989)
Acceptable Fit χ2/df < 2
RMSEA RMSEA < .05 Brown (2012)
(good fit) Hu and Bentler (1999)
RMSEA < .08 Jöreskog and Sörbom (1993)
(fair fit) Tabachnick and Fidell (2007)
RMR RMR≤.05 Jöreskog and Sörbom (1993)
(good fit)
SRMR SRMR≤.08 Hu and Bentler (1999)
(reasonably good fit)
GFI .95 ≤ GFI ≤ 1 Tabachnick and Fidell (2007)
AGFI .85 ≤ AGFI ≤ 1 Tabachnick and Fidell (2007)
CFI .90 ≤ CFI ≤ .95 Brown (2012)
(adequate fit)
IFI .90 ≤ CFI ≤ .95 Brown (2012)
(adequate fit)
Note: χ2 = chi-square, RMSEA = root mean square error of approximation, RMR = the root mean square residual, SRMR = the
standardized root mean square residual, GFI = goodness of fit, AGFI = adjusted goodness of fit, CFI = comparative fit index, and IFI =
incremental fit index.

(default model in AMOS) only the factor loadings are constrained equal to the data of
each group separately. Secondly, by satisfying the conditions of measurement
invariance a multi-group CFA was conducted by including factor-loading parameters
remained constrained and equality constraints were then placed on the factor
variances and covariance for both groups simultaneously to test the second
hypothesis. This process is specifically named as “the manual multiple-group
approach” (Byrne, 2010, p.214). The common fit indices are given in Table 1 with
their critical value ranges. In this particular study, the similarity between two groups
was controlled by ensuing a similar analysis procedure carried out by Sanchez,
Gallego, Soria, and Aborg (2008) was closely followed.
In addition to the critical values in Table 1, Hu and Bentler (1999), for example,
suggested a combination of rules for a reasonably good fit between the baseline model
and the source data. a) SRMR values are close to or below .08; b) RMSEA values are
close to or below .06; and c) CFI and TLI values are close to or greater .95. Hu and
Bentler indicated that these are necessary to improve the acceptability of Type I and
Type II error rates. There is no strict difference between their combined values and
the critical limits of this study, given in Table 1. In Table 1, χ2 (CMIN) value is not a
unique index of good model fit, and it is sensitive to sample size (Brown, 2012;
Tabachnick & Fidell, 2007). Statistically significant χ2 is not enough to imply a poor-
data-to-model fit. In such a case, the model fit could be evaluated by comparing the
value of χ2 with two times of degrees of freedom (Byrne, 1989), by analyzing the
critical limits for χ2/df value, and by analyzing the other fit indexes. Further, analysis
was conducted for the comparisons of the original seven-factor structure model
(Model7) with the model in which only the factor loadings are constrained equal
among groups (Model7MW); all factor loadings, variances and covariance are
constrained equal among groups (Model7SC); and all factor loadings, variances,
covariance and error variances are constrained equal among groups (Model7MR). As
Byrne (2010, p.221) indicated, evidence of invariance between Model7 and the other

310 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

Table 2. Goodness-of-fit indices from cfa conducted separately on the data for the Almeria and Burdur
samples
Sample Model χ2 df p χ2/df SRMR GFI AGFI RMSEA RMR IFI CFI
202 Model7 643.984 329 <.001 1.957 0,071 .922 .880 .070 .053 .919 .917
Spanish
185 Model7 611.143 329 <.001 1.858 0,056 .925 .884 .062 .047 .917 .916
Turkish
Note: χ2: chi square; df: degrees of freedom.

three models were based on the differences in CFI values presenting a probability
<.001. The second research question was evaluated through a detailed descriptive
analysis for both samples.

RESULTS

Findings for the first research question

Confirmatory factor analysis
The evaluation of the first research question is analyzed under two main
hypothesis first of which was assessed by CFA to validate the original Model7
separately for the Almeria and Burdur participants. Table 2 presents the Model7 fit
indices for both groups.
Table 2 presents the related fit indexes (CFI, GFI, AGFI, and IFI) that are all within
the accepted thresholds given in Table 1. The RMSEA values are smaller than .08,
which corresponds to a fair fit. Moreover, the rejection criteria for the fit formed by
various combinations of SRMR, RMR, RMSEA, and CFI values were also discarded with
respect to the values in Table 2, especially given SRMR <.08. Therefore, in both
countries, the DAS scale corresponds to the same seven-factor structure model
(Model7) as it adequately fits the theoretical model. The path diagram of the original
seven-factor structure (Model7) is presented in Figure 2. The respective factor
loadings and variance estimation values derived from both Almeria and Burdur
samples were combined on the same figure.
Multi-group analysis
After satisfying the conditions of group equivalence, the Model 7 was performed
through both groups simultaneously through multi-group CFA to test the second
hypothesis. At first, as suggested by Byrne (2010) multi group analysis was
performed by including both groups simultaneously and running Model7
(unconstrained model in AMOS) in which the factor-loadings are constrained equal.
The Model7 tested the factorial structure through two groups and the fit indices for
the Model7 were 2=1255.132, 2/df = 1.907, df=658, p < .001, RMSEA = .049, CFI =
.917, SRMR=0.070, RMR = .060, IFI = .918, GFI = .923, and AGFI = .882. The common
modification indices for the two groups with respect to the original Model7 imply that
the model fairly fit the data with all parameters freely estimated in the two groups.
The results of the multi-group CFA indicated that in terms of factor loadings seven
latent factors of attitude toward teaching, the adapted DAS scale was compatible for
both countries at an adequate level.
For further analysis, the baseline comparisons of the original Model7 with the
other models, Model7MW, Model7SC, and Model7MR is investigated. The related fit
indices for the multi-group CFA are presented in Table 3.
The differences between CFI values for the Model7 and the second and third
models met the cutoff criteria of below .01 indicating that it is an evidence of
invariance across models. The model in which only the factor loadings are
constrained equal among groups (Model7MW) as well as the model in which all factor

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 311
F. Korur et. al

Figure 2. The path diagram of the original seven-factor structure and respective factor loadings derived
from the two sample (Almeria, N=202/ Burdur, N=185)

Table 3. Multi-group CFA conducted simultaneously for both groups

Model χ2 df p SRMR GFI RMSEA IFI CFI Δχ2
Model 7 1255.132 658 <.001 .070 .923 .049 .918 .917
Model 7MW 1313.757 679 <.001 .071 .916 .049 .912 .911 Model 7MW-Model7=58.625
Model 7SC 1348.270 707 <.001 .074 .912 .049 .911 .910 Model 7SC-Model 7=93.138
Model 7MR 1623.382 735 <.001 .081 .878 .056 .876 .876 Model 7MR-Model 7=368.25
Note: χ2:chi square; df=degrees of freedom.

loadings, variances and covariance are constrained equal among groups (Model7SC)
were successfully fitted to the data of two groups. Therefore, it can be concluded that
all factor loadings constrained to be equal across two groups, Spanish and Turkish in-
service elementary school teachers. The confirmed factors correspond to categories
of attitude toward science teaching. The category names given by van Aalderen-
Smeets and Walma van der Molen (2013) were also used in this study.
The construct validity was also controlled by convergent and discriminant validity.
The convergent validity, which estimates the degree to which items of a theoretical
construct relate to each other, was tested. If the items of a construct load strongly
together, this is an indicator of high convergent validity (Ullman, 2007). The factor
loadings, most of which are above the critical value of .50, found at the end of the
multi-group CFA, ranged from .43 to .99 within the seven-factor structure. Moreover,

312 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

Table 4. The Cronbach’s Alpha values with respect to the seven sub-dimensions of the DAS scale
α α α
Item Numbers
(Almeria data) (Burdur data) (Both)
SE 1-4 .83 .85 .84
PR 5-9 .76 .71 .74
GS 10-14 .90 .84 .87
A 15-18 .95 .95 .95
DST 19-21 .91 .83 .87
PDC 22-24 .56 .60 .57
E 25-28 .95 .97 .96
OVERALL -- .82 .76 .80
Note. SE: self-efficacy; PR: perceived relevance; GS: gender stereotyping; A: anxiety; DST: difficulty of science teaching; PDC:
perceived dependency on context factors; E: enjoyment.

Table 5. Percentages for the seven sub-categories by country

Almeria/Spain Burdur/Turkey

Positive Negative Positive Negative

Neutral Neutral
Responses Response Response Response
5 4 3 2 1 5 4 3 2 1
% %
1. SE 43.7 37.0 16.5 2.5 0.4 44.1 41.4 11.0 3.5 0.0
2. PR 42.4 36.4 17.9 2.7 0.6 41.7 39.8 13.1 4.7 0.7
3. GS 2.9 6.5 15.7 27.8 47.0 8.0 9.5 16.2 44.5 21.8
4. A 0.7 2.7 10.8 35.4 50.4 0.6 4.5 5.0 45.7 44.2
5. DST 1.8 13.5 27.1 33.0 24.6 3.3 19.8 17.1 33.3 26.5
6. PDC 16.0 33.2 35.8 12.4 2.6 15.0 43.2 16.0 17.8 8.0
7. E 26.6 38.2 31.7 3.0 0.5 37.0 52.7 6.8 2.2 1.3

the factors measured distinctly different concepts, supporting discriminant validity

with inter-correlation coefficients that ranged from .04 to .63. They were all below
the critical limit of .80 (Brown, 2012).
The reliability coefficient (Cronbach’s alpha) of the scale with respect to the whole
was calculated as  = .80. It was .82 and .76 in the Almeria and Burdur samples,
respectively. Table 4 shows the reliability coefficients corresponding to seven factors
for both the Almeria and Burdur data. The values indicate high reliability of the test
results (Fraenkel et al., 2012). Therefore, the results collected from both samples with
the DAS scale were reliable and valid.
Findings for the second research question
The latent structure of the DAS is multifactorial (i.e., seven factors) and since the
factor loadings are supported by CFA, the subscales might also be scored, since the
number of factors is indicative of the number of subscales, and the pattern of items
loading on factors reveals how the subscales should be scored (Brown, 2012). For this
reason, the participant teachers’ frequencies in the items of a sub-category (see Figure
2) were added up, and its percentages was calculated. The same calculation was
performed for seven-subcategories of both samples. The percentage distribution of
teachers’ responses to the items with respect to confirmed categories of the DAS scale
is presented in Table 5.

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 313
F. Korur et. al

In both samples, teachers’ negative responses constituted more than 80% in the
“Anxiety” dimension. This indicates that they had little to no anxiety, fear, or stress in
science teaching. In Burdur sample around 66% of teachers and in Almeria sample
almost 76% of teachers (shown by their negative responses; Table 5) were against
gender-stereotyping in the science teaching/learning process. In both samples,
agreement was strongly in the sub-dimensions (shown by their positive responses;
Table 5) of “Self-efficacy” and “Perceived Relevance” as the percentages was above 80
for each of them. The other two common positive categories are “Perceived
Dependency Context Factors” and “Enjoyment.” The teachers in Burdur perceived
that their attitudes toward science teaching were affected by contextual
environmental factors, e.g diversity of students more than teachers in Almeria. In
Burdur, almost 90% of the teachers indicated that they enjoyed science teaching and
liked their jobs. In Almeria, 65% of the teachers said the same, whereas 32% of them
said they had neutral feelings on the matter.

DISCUSSION

The DAS scale was developed by van Aalderen-Smeets and Walma van der Molen
(2013) was translated into Spanish and Turkish. The validity and reliability of the
results of both versions were also examined. The results confirmed that the seven-
factor structure of the translated scales is valid for the two samples. Further, the
multi-group CFA verifies the theoretical model proposed by van Aalderen-Smeets et
al. (2012). The items intended to measure a specific sub-dimension loaded onto that
specific factor without cross-loadings. Therefore, the seven factor structure of DAS is
invariant across the samples.
With respect to two sub-dimensions of cognitive states, “Gender Stereotyping” and
“Difficulty of Science Teaching”, van Aalderen-Smeets and Walma van der Molen
(2013) recommended that they should be analyzed thoroughly. The two sub-
dimensions, as they have crucial factor loadings, were analyzed in both samples. It
was found that the teachers were against gender stereotyping in science teaching.
Further, almost 60% of the teachers in the sample of this study perceived that
‘average’ elementary teacher did not find difficult to teach science. The other sub-
dimension is ‘perceived relevance’ which has high percentage scores related to
teachers’ perceptions indicating their strong cognitive beliefs. It is possibly explained
by the teachers’ understanding the importance of science teaching and knowing how
to adapt science to technological and social developments. The results of this study
(almost 80 % of the teachers) and related studies (Harlen & Holroyd, 1997; van
Aalderen-Smeets & Walma van der Molen, 2013) indicate that elementary teachers
identified that science teaching is important and they expressed it is relevant to
primary school level.
The percentage score for self-efficacy that is a sub-dimension of perceived control
is high in both countries. The reason might be the teachers’ high self-confidence as
well as their low rate of anxiety. Around 70% of the teachers graduated from the
university in the city where they lived, so they knew the culture, level of education,
and socio-economic status in these cities, which allowed them to develop self-
confidence. Further, the teachers in this study were mostly experienced and tended
to lack anxiety. For example, in Burdur the large percentage (70%) of participants had
more than 21 years of teaching experience. This result contradicts the report (OECD,
2009) indicating teachers’ self-efficacy beliefs in both countries are weak. However,
the results of this study are consistent with corresponding findings in related
literature (Brígido et al., 2013; Bursal, 2008, 2010; Ergül, 2009; Osborne et al., 2003;
Tekkaya et al., 2004; Tosun, 2000; Türkmen, 2013).
“Anxiety” together with “Enjoyment” are the sub-dimensions of affective states. In
both samples most of the teachers were tenured, and therefore, they may have felt

314 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

more relaxed and confident, as indicated by the OECD report (OECD, 2009). In terms
of “Enjoyment”, on the other hand, teachers in Burdur had higher scores. One might
expect the scores to be similar since both cities are in the Mediterranean region of
respective countries and Mediterranean people are known for their warm and
moderate job-related attitudes. However, there may be differences in job satisfaction
between the two countries, because of the perceived job stress (Bursal, 2010; Klassen
& Chiu, 2010). There was also a crucial difference between the two countries in terms
of “Perceived Dependency Context Factors”. The teachers, in Burdur, thought that
they might feel relax when they had more support, such as collegial support and time
allocation for science in the curriculum, as these factors decrease anxiety (Ergül,
2009; Martin-Dunlop & Fraser, 2008; van Aalderen-Smeets & Walma van der Molen,
2013). The behavioral intention of the teachers is not directly measured as the results
were limited how elementary teachers perceived the items in the DAS scale. However,
as indicated in related literature teachers’ perceptions of attitudes may affect their
behavioral intention and cause a change in their behaviors (Haney et al., 1996; van
Aalderen-Smeets et al., 2012; Yilmaz-Tuzun, 2008).

CONCLUSION AND RECOMMENDATIONS

This study presents Spanish and Turkish translated versions of the new and the
current DAS scale that provide broader perspective than previous scales’ in terms of
teachers’ attitude toward science teaching. In the translation and adaptation process,
highly suggested approaches reviewed and a well-planned five-step guideline was
conducted. The guideline includes some structural, semantic and conceptual
equivalence of the translations. Translation equivalence procedures were rigorously
explained, as they might lead up future studies aimed to adapt it into other languages.
The model fit, examined with CFA, of seven-factor structure of the DAS scale for both
versions also supports construct equivalence of the translated scales. Therefore, this
study in general contained clear directions for further studies aimed to adapt and
validate the DAS scale that can be adapted to other cultures.
The multi-group CFA confirmed that the factor structures of the adapted scales are
invariant across the Spanish and Turkish samples. The two versions of the DAS scale
appear to represent valid and reliable measures of attitudes toward science teaching
of in-service elementary teachers in both samples. The reason for this fit is likely that
the original scale was applied again in a European country (Netherlands), where the
education system, science curriculum, teacher training program, or expectations of
teachers may yield almost the same effects on teachers’ attitudes toward science
teaching. As stated by van Aalderen-Smeets and Walma van der Molen (2013), the
DAS scale is coherent and effective in assessing elementary teachers’ attitudes toward
science teaching because items measuring “attitude toward science” and “science
attitude,” which could be confused, were removed. The most crucial products of this
study are not only the Turkish and Spanish versions of the scale but also confirmation
of the cross-cultural structure of the DAS scale that would be useful in further studies
on attitudes toward science teaching. In fact the cross-cultural adaptation of the DAS
scale might have led the further studies to combine teachers’ attitude toward science
teaching with the findings of other international cross-cultural comparison studies
such as TIMSS and PISA-Science. The confirmed seven factors of the DAS are the sub-
categories of the attitude toward science teaching.
Future studies could analyze how in-service elementary school teachers’ attitudes
toward science teaching in each country affect students’ achievement related to the
seven categories confirmed in this study. Moreover, the model fit of the DAS scale can
be investigated in other cultural contexts as well. Finally, a subsequent investigation
could be organized with the teachers in the Almeria sample to explore the reasons
behind their moderate level of enjoyment in science teaching.

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 315
F. Korur et. al

REFERENCES

Azar, A. (2010). In-service and pre-service secondary science teachers’ self-efficacy beliefs
about science teaching. Educational Research and Reviews, 5(4), 175-188.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Beaton, D. E., Bombardier, C., Guillemin, F., & Ferraz, M. B. (2000). Guidelines for the process
of cross-cultural adaptation of self-report measures. Spine, 25 (24), 3186–3191.
Behling, O., & Law, K.S. (2000). Translating questionnaires and other research instruments:
Problems and Solutions. Sage Publications: London.
Bilen, K., & Köse, S., (2012). Yapılandırmacı öğrenme teorisine dayalı etkili bir strateji:
Tahmin-gözlem-açıklama (TGA) “bitkilerde büyüme ve gelişme” [An effective strategy
based on constructivist learning theory: Guess-observation-Interpretation “growth and
development in plants]. Pamukkale Üniversitesi Eğitim Fakültesi Dergisi, 31(1), 123-136.
Blalock, C. L., Lichtenstein, M. J., Owen, S., Pruski, L., Marshall, C., & Toepperwein, M., (2008).
In pursuit of validity: A comprehensive review of science attitude instruments 1935–
2005. International Journal of Science Education, 30(7), 961-977.
Brígido M., Borrachero, A. B., Bermejo, M. L., & Mellado V., (2013). Prospective primary
teachers’ self-efficacy and emotions in science teaching, European Journal of Teacher
Education, 36(2), 200-217. doi: 10.1080/02619768.2012.686993
Brown, T. A., (2012). Confirmatory factor analysis for applied research. New York, NY: The
Guilford Press.
Bursal, M., (2008). Changes in Turkish Pre-Service Elementary Teachers’ Personal Science
Teaching Efficacy Beliefs and Science Anxieties during a Science Method Course. Journal
of Turkish Science Education, 5(1), 99-112.
Bursal, M., (2010). Turkish, pre-service elementary teachers’ self-efficacy beliefs regarding
mathematics and science teaching. International Journal of Science and Mathematics
Education, 8(4), 649-666.
Byrne, B. M. (1989). A primer of LISREL: Basic applications and programming for confirmatory
factor analytic model. New York, NY: Springer-Verlag.
Byrne, B. M. (2008). Testing for multigroup equivalence of a measuring instrument: A walk
through the process, Psicothema, 20(4), 872-882.
Byrne, B. M. (2010). Structural equation modeling with Amos: Basic concepts, applications, and
programming (2nd ed.). New York, NY: Taylor and Francis Group.
Campbell, T., Medina-Jerez, W., Erdogan, I., & Zhang, D., (2010). Exploring science teachers'
attitudes and knowledge about environmental education in three international teaching
communities. International Journal of Environmental and Science Education, 5(1), 3-29.
Child, D., (1990). The essentials of factor analysis (2nd ed). London, England: Cassel Educational
Limited.
Driscoll, M. P., (2000). Pschology of learning for instruction (2nd ed.). Boston, MA: Allyn & Bacon.
Duban, N. Y., & Gökçakan, N., (2012), Sınıf öğretmeni adaylarının fen öğretimi öz-yeterlik
inançları ve fen öğretimine yönelik tutumları [Science teaching self-efficacy beliefs and
attitudes toward teaching science of preservice elementary teachers]. Çanakkale
Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 21(1), 267-280.
Education, Audiovisual and Culture Executive Agency (2011). Science Education in Europe:
National Policies, Practices and Research. Retrieved from
http://eacea.ec.europa.eu/education/eurydice/documents/thematic_reports/133EN.p
df
Ergül, N. R. (2009). Elementary pre-service teachers’ opinions on teaching science. Bulgarian
Journal of Science and Education Policy (BJSEP), 3(2), 153-172.
Fraenkel J. R., Wallen N. E., & Hyun H. H., (2012). How to Design and Evaluate Research in
Education (8th ed.). New York, NY: Mc Graw – Hill.
Gardner, P. L. (1995). Measuring attitudes to science: Unidimensionality and internal
consistency revisited. Research in Science Education, 25, 283–289, doi:
10.1007/BF02357402.
Gierl, M. J. (2000). Construct Equivalence on Translated Achievement Tests. Canadian Journal
of Education, 25(4), 280-296.
Guisasola, J., Robinson, M., Zuza, K. (2007). A Comparison of the Attitudes of Spanish and
American Secondary Science Teachers Toward Global Science and Technology Based
Problems/Threats. Journal of Environmental & Science Education, 2(1), 20-31.

316 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

Haney, J. J., Czerniak, C. M., & Lumpe, A. T. (1996). Teacher beliefs and intentions regarding the
implementation of science education reform strands. Journal of Research in Science
Teaching, 33, 971 – 993.
Harlen, W. & Holroyd, C. (1997). Primary teachers’ understanding of concepts of science:
impact on confidence and teaching, International Journal of Science Education, 19(1), 93-
105. doi: 10.1080/0950069970190107
Hu, L.-T., & Bentler, P. M. (1999). Cut-off criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55.
International Council for Science (2011). Report of the ICSU Ad-hoc Review Panel on Science
Education. Retrieved from: http://www.icsu.org/publications/reports-and-
reviews/report-of-the-icsu-ad-hoc-review-panel-on-science-education/report-of-the-
icsu-ad-hoc-review-panel-on-science-education
Jarvis, T., & Pell, A. (2004). Primary teachers’ changing attitudes and cognition during a two-
year science in-service programme and their effect on pupils. International Journal of
Science Education, 26(14), 1787-1811.
Jöreskog, K. G., & Sörbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS
command language. Hillsdale, NJ: Lawrence Erlbaum Associates.
Klandermans, B., & Smith J. (2002). Survey Methods. A Case for Comparative Designs. In B.
Klandermans & S. Staggenborg (Eds.), Research Methods on Social Movements (pp. 3-32).
Minneapolis: University of Minnesota Press.
Klassen, R. M., & Chiu M. M. (2010). Effects on Teachers’ Self Efficacy and Job Satisfaction:
Teacher Gender, Years of Experience, and Job Stress. Journal of Educational Psychology,
102(3), 741-756.
Küçüközer, H., Kırtak Ad, V. N., Ayverdi, L., & Eğdir, S. (2012). Turkish Adaptation of
Constructivist Learning Environment Survey. Elementary Education Online, 11(3), 671-
688.
Lin J-L., Cheng M-F., Chang Y-C, Li H-W., Chang J-Y. & Lin D-M. (2014). Learning Activities That
Combine Science Magic Activities with the 5E Instructional Model to Influence
Secondary-School Students’ Attitudes to Science. Eurasia Journal of Mathematics, Science
& Technology Education, 10(5), 415-426. doi: 10.12973/eurasia.2014.1103a
Martin-Dunlop, C., & Fraser, B.J. (2008). Learning environment and attitudes associated with
an innovative science course designed for prospective elementary teachers.
International Journal of Science and Mathematics Education, 6(1), 163-190.
Moore, R. W. & Foy, R. L. H. (1997). The Scientific Attitude Inventory: A Revision (SAI II).
Journal of Research in Science Teaching, 34(4), 327–336.
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.
Özkan, Ö., Tekkaya, C. & Çakıroğlu, J. (2002). Fen Bilgisi Aday Öğretmenlerin Fen Kavramlarını
Anlama Düzeyleri, Fen Öğretimine Yönelik Tutum ve Öz yeterlik İnançları [Pre-service
science teachers’ level of understanding science concepts, attitudes toward teaching science
and self-efficacy beliefs]. V. Fen Bilimleri ve Matematik Eğitimi Kongresi, Ankara, Turkey.
Program for International Student Assessment (2012). PISA 2012 Results in focus. Retrieved
from http://www.oecd.org/pisa/keyfindings/pisa-2012-results.htm
Saad R. & BouJaoude S. (2012). The Relationship between Teachers’ Knowledge and Beliefs
about Science and Inquiry and Their Classroom Practices. Eurasia Journal of
Mathematics, Science & Technology Education, 8(2), 113-128.
Sanchez, A. M. R, Gallego, E. C., Soria, M. S., & Aborg, C. (2008). Technoflow among Spanish and
Swedish students: A Confirmatory Factor Multigroup Analysis , anales de psicologia,
24(1), 42-48.
Štefaniková S. & Prokop P. (2015). Do We Believe Pictures More or Spoken Words? How
Specific Information Affects How Students Learn about Animals. Eurasia Journal of
Mathematics, Science & Technology Education, 11(4), 725-733. doi:
10.12973/eurasia.2015.1380a
Tabachnick, B.G., & Fidell, L.S. (2007). Using Multivariate Statistics (5th ed.). Boston: Pearson
Education, 54-55.
Tekkaya, C., Çakıroğlu, J., & Özkan, Ö. (2004). Turkish pre-service science teachers’
understanding of science and their confidence in teaching it. Journal of Education for
Teaching : International research and pedagogy, 30, 57–66.

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 317
F. Korur et. al

The Organization for Economic Co-operation and Development-OECD (2009). Annual Report -
Teaching Practices, Teachers’ Beliefs and Attitudes (Ch.4). Retrieved from:
http://www.oecd.org/education/school/43023606.pdf
Thompson, C. L., & Shrigley, R. L., (1986). What Research Says. School Science and
Mathematics, 86(4), 331- 343.
Tosa, S. (2009). Teaching Science as Inquiry in US and in Japan: A Cross-Cultural Comparison of
Science Teachers’ Understanding of, And Attitudes Toward Inquiry-Based Teaching. UMI
No. AAI3361230. University of Massachusetts Lowell, Massachusetts.
Tosun, T. (2000). The beliefs of preservice elementary teachers toward science and science
teaching. School Science and Mathematics, 100(7), 376-382.
Türkmen, L. (2013). In-service Turkish elementary and science teachers’ attitudes toward
science and science teaching: A sample from Uşak province, Science Education
International, 24(4), 437-459.
Ullman, J. B. (2007). Structural Equation Modelling. In B. G. Tabachnick & L. S. Fidell (Eds.),
Using Multivariate Statistics (pp. 676-780). (5th ed.). Boston, US: Allyn & Bacon.
Van Aalderen‐Smeets, S. I., Walma van der Molen, J. H., & Asma, L. J. (2012). Primary teachers'
attitudes toward science: A new theoretical framework. Science Education, 96(1), 158-
182.
Van Aalderen-Smeets, S. I., & Walma van der Molen, J. H. (2013). Measuring primary teachers'
attitudes toward teaching science: Development of the dimensions of attitude toward
science (DAS) instrument. International Journal of Science Education, 35(4), 577-600.
Yakar Z. & Baykara H. (2014). Inquiry-Based Laboratory Practices in a Science Teacher
Training Program. Eurasia Journal of Mathematics, Science & Technology Education,
10(2), 173-183. doi: 10.12973/eurasia.2014.1058a
Yaşar, Ş., & Anagün, S. Ş. (2008). İlköğretim beşinci sınıf fen ve teknoloji dersi tutum ölçeğinin
geçerlik ve güvenirlik çalışmaları. Anadolu Üniversitesi Sosyal Bilimler Dergisi, 8(2), 223-
236.
Yilmaz-Tuzun, O. (2008). Preservice elementary teachers’ beliefs about science teaching,
Journal of Science Teacher Education, 19, 183-204. doi :10.1007/s10972-007-9084-1



318 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320
 Confirmatory analysis of teachers’ attitude

APPENDIX

a. Turkish form of the DAS scale

FEN ÖĞRETİMİNE YÖNELİK TUTUM ÖLÇEĞİ
AÇIKLAMA: Bu çalışma İspanya ve Türkiye’deki örneklemlerinde sınıf öğretmenlerinin fen öğretimine karşı
tutumuna yönelik genel eğilimi belirlemeyi amaçlamaktadır. Araştırma sonuçları tek bir kişi ile ilgili değil,
istatistiksel olarak var olan durumu ortaya koymak adına olduğu için isim yazmak zorunda değilsiniz.
Lütfen uygun kutulara X koyarak cevaplayınız. İşbirliğiniz için çok teşekkür ederiz.
Yrd. Doç. Dr. Fikret KORUR
Noemí Torres Serrano
Rocío Vargas Vargas
ÖLÇEK
Uygun Bölüme ‘X’
Koyunuz.

Katılmıyorum

Katılmıyorum
Maddeler

Katılıyorum
Katılıyorum
Kararsızım
Kesinlikle

Hiç
No
1 Öğrencilerden gelen fenle ilgili soruların rahatlıkla üstesinden gelebiliyorum.
2 Fen içeriği ile ilgili konuları ilkokul öğrencilerine anlatabilecek düzeyde yeterli bilgiye sahibim.
3 Öğrencilerin sınıfta araştırma ve tasarım yapmalarını destekleyebilecek düzeyde iyi malzeme bilgisine sahibim.
Eğer ilkokul öğrencileri fen ile ilgili ödevlerinde bir çözüm üretemezse, bu ödevlerinde ilerleme göstermeleri
4
konusunda onlara yardım edebileceğimi düşünüyorum.
5 İlköğretimde fen bilimlerine mümkün olduğunca erken başlanması gerektiğini düşünüyorum.
6 Fen bilimleri eğitiminin ilkokul öğrencilerinin gelişimi için gerekli olduğunu düşünüyorum.
Fen bilimleri eğitiminin ilkokul öğrencilerinin toplumdaki teknolojik problemlerle ilgilenmelerini sağlaması için
7
gerekli olduğunu düşünüyorum.
İlkokulda fen bilimleri eğitimi öğrencilerin çalışmaları ile ilgili doğru tercihler (kariyer seçimi ve ders seçimi gibi)
8
yapabilmeleri için gerekli olduğunu düşünüyorum.
İlkokulda fen bilimleri eğitimi o kadar önemlidir ki deneyimsiz öğretmenler bu alanda fazladan eğitimler
9
almalıdır.
Erkek ilkokul öğretmenlerinin, bayan ilkokul öğretmenlerine göre fen bilimlerini öğretmekten daha fazla
10
hoşlandıklarını düşünüyorum.
İlkokuldaki erkek öğrencilerin kız öğrencilere göre daha büyük olasılıkla fen ile ilgili ödev seçeceklerini
11
düşünüyorum.
Öğrencilerle bir araştırma veya teknik ödevi erkek ilkokul öğretmenlerinin, bayan ilkokul öğretmenlerine göre
12
daha rahat yapabileceklerini düşünüyorum.
İlkokuldaki erkek öğrencilerin, kız öğrencilere göre malzemelerle ve kimyasal maddelerle deney yapma
13
konusunda daha fazla hevesli olduklarını düşünüyorum.
14 Bir fen gösterisi için bilinçli olmayarak bir kız öğrenciden daha çok erkek öğrenci seçeceğimi düşünüyorum.
15 Sınıfta fen öğretirken gergin hissederim.
16 Fen öğretirken sinirli hissederim.
17 Fen öğretmek beni sinirli yapıyor.
18 Sınıfımda fen öğretmem gerektiğinde stresli oluyorum.
19 Çoğu ilkokul öğretmeninin, fen ile ilgili konuları öğretmeyi zor bulduğunu düşünüyorum.
Çoğu ilkokul öğretmeninin, fen bilimleri dersinin içeriği açısından konuları öğretmeyi zor bulduğunu
20
düşünüyorum.
21 Öğretmenlerin fen bilimleri dersinde karşılaşılan konuları karışık bulduklarını düşünüyorum.
22 Bana göre, uygun kullanıma hazır paket malzemelerin kullanılması sınıfta fen öğretimi için gereklidir.

23 Bana göre, uygun bir fen öğretim yönteminin olması sınıfta fen öğretmek veya öğretmemek için belirleyicidir.

24 Bana göre, meslektaşlarımın ve okulun desteği sınıfta fen öğretmek veya öğretmemek için belirleyicidir.
25 Fen öğretirken kendimi mutlu hissediyorum.
26 Fen öğretmek beni neşeli yapıyor.
27 Fen öğretmekten çok zevk alıyorum.
28 Fen öğretmek beni heyecanlandırıyor.

© 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320 319
F. Korur et. al

b. Spanish form of the DAS scale

ESCALA DE ACTITUDES EN LA ENSEÑANZA DE LAS CIENCIAS

Estimado,
Este estudio tiene como objetivo identificar la tendencia o actitud general de los maestros de educación primaria
"hacia la enseñanza de la ciencia en España y en Turquía''. Usted no tiene que escribir su nombre ya que el estudio
no está relacionado con una persona concreta, sino que se trata de un estudio donde los resultados se analizarán
estadísticamente con el fin de aclarar la situación actual.
Por favor, responde poniendo una X en las casillas correspondientes. Muchas gracias por su colaboración.
Asst. Prof. Dr. Fikret KORUR
Noemí Torres Serrano
Rocío Vargas Vargas
Escala
pon una "X" en las
casillas
correspondientes

En desacuerdo
Totalmente de
Artículos

desacuerdo
Totalmente
De acuerdo
acuerdo

Neutral
No.
1 Soy perfectamente capaz de resolver las preguntas de mis alumnos/as sobre ciencias.
Tengo suficiente conocimiento sobre los contenidos de ciencias para enseñar adecuadamente dichas
2
asignaturas en la educación primaria.
3 Tengo un dominio suficiente del material para ayudar a mi alumnado en la investigación y diseño de clase.
Si los niños/as de educación primaria no alcanzan una solución durante las tareas de ciencias, considero que
4
podré tener éxito ayudándolos a solucionar dichas tareas.
5 Pienso que las ciencias deben enseñarse en la educación primaria tan pronto como sea posible.
6 Pienso que la enseñanza de las ciencias es esencial en educación primaria para el desarrollo del/a niño/a.
Pienso que la enseñanza de las ciencias es esencial para hacer que los/as niños/as de educación primaria se
7
involucren más en los problemas tecnológicos de la sociedad.
Pienso que la enseñanza de las ciencias en la educación primaria es esencial para que los/as alumnos/as sean
8 capaces de tomar buenas decisiones acerca de sus estudios (por ejemplo, elección de una carrera y elección de
un curso)
La enseñanza de las ciencias es tan importante en la educación primaria que maestros/as inexpertos/as
9
deberían recibir una formación adicional en este área.
10 Pienso que la enseñanza de las ciencias sociales es más divertida para los maestros que para las maestras.
Pienso que en educación primaria los niños son más propensos que las niñas a elegir tareas que tienen que ver
11
con la ciencia.
Pienso que los maestros de educación primaria pueden hacer una investigación técnica con mayor facilidad que
12
las maestras.
Pienso que los niños de educación primaria son más entusiastas acerca de lo experimentado con materiales y
13
sustancias químicas que las niñas.
Pienso que yo inconscientemente sería más propenso a elegir para una demostración científica a un niño que a
14
una niña.
15 Me siento tenso/a mientras enseño ciencias en clase.
16 Me siento nervioso/a mientras enseño ciencias en clase.
17 La enseñanza de las ciencias me pone nervioso/a.
18 Me siento estresado/a cuando tengo que enseñar ciencias en mi clase.
Pienso que la mayoría de los/as maestros/as de educación primaria tienen dificultades para enseñar temas
19
relacionados con la ciencia.
Pienso que la mayoria de los/as maestros/as de educación primaria encuentran la ciencia una materia difícil de
20
enseñar en términos de contenido.
21 Pienso que los/as profesores/as encuentran complicados los temas que surgen en la ciencia.
Para mí, la disponibilidad de un paquete de materiales listo para su uso es esencial para la enseñanza de la
22
ciencia en el aula.
Para mí, la disponibilidad de un método de enseñanza de las ciencias es decisivo independientemente de si voy
23
a enseñar o no ciencias en clase.
Para mi, el apoyo de mis compañeros/as y de la escuela es decisivo independientemente de si voy a enseñar o
24
no ciencias en clase.
25 Me siento feliz mientras enseño ciencias.
26 Enseñar ciencias me hace estar alegre.
27 Me divierto mucho enseñando ciencias.
28 La enseñanza de las ciencias me entusiasma.

320 © 2016 iSER, Eurasia J. Math. Sci. & Tech. Ed., 12(2), 303-320