Aydin 2009
Aydin 2009
Aydin 2009
Measurement
http://epm.sagepub.com/
Published by:
http://www.sagepublications.com
Additional services and information for Educational and Psychological Measurement can be found
at:
Subscriptions: http://epm.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations: http://epm.sagepub.com/content/69/5/868.refs.html
What is This?
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Validity Studies Educational and
Psychological Measurement
Volume 69 Number 5
October 2009 868-880
Development and Psychometric Ó 2009 SAGE Publications
10.1177/0013164409332213
The aim of this study was to develop a scale assessing high school students’ self-
efficacy beliefs in chemistry-related tasks and to assess psychometric properties of
scores on this scale. A pilot study with a sample of 150 high school students provided
initial evidence for two-factor structure of 16-item scale, named High School Chemis-
try Self-Efficacy Scale (HCSS). The final form of the HCSS was administered to 362
high school students in Turkey. Confirmatory factor analysis fit indices and factor
pattern coefficients supported the proposed structure, with the two factors directly
corresponding to the hypothesized dimensions. These dimensions were chemistry
self-efficacy for cognitive skills (10 items, a reliability = .90) and self-efficacy for
chemistry laboratory (6 items, a reliability = .92). In addition, the two-factor model
for the HCSS was invariant across school types. Analyses of latent mean structure
indicated that private school students were statistically more efficacious than public
school students in chemistry laboratory skills. The HCSS is a promising tool to iden-
tify chemistry self-efficacy beliefs in high school students.
O ver the past 2 decades, research studies indicated that students find most
chemistry concepts difficult to understand. Because of the abstract nature of
chemistry, students are not able to explain chemical principles at the molecular level
or apply them to real life (Carter & Brickhouse, 1989; Dalgety & Coll, 2006; Gopal,
Kleinsmidt, Case, & Musonge, 2004; Kennedy, 1996; Tsaparlis, 1997; Watson &
Authors’ Note: We would like to acknowledge the help of Ceyda Yamudi and Cansel Kadioglu for their
assistance with the collection of the data and thank to Dr. Anita Woolfolk Hoy and Dr. William E. Load-
man for their valuable comments on earlier versions of this manuscript. Please address correspondence
to Yeşim Çapa Aydın, Department of Educational Sciences, Middle East Technical University, Ankara,
Turkey 06531; e-mail: capa@metu.edu.tr
868
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Çapa Aydın, Uzuntiryaki / High School Chemistry Self-Efficacy 869
McEwen, 1994). Consequently, most students perform poorly in chemistry and pre-
fer not to enroll chemistry classes (Britner & Pajares, 2006). In addition, students’
beliefs about their ability affect their achievement. Social cognitive theory (Bandura,
1986) proposed that people’s self-beliefs have a great influence on their thoughts,
behaviors, and performance. Among these beliefs, perceived self-efficacy, which is
defined as ‘‘people’s judgments of their capabilities to organize and execute courses
of action required to attain designated types of performances’’ (Bandura, 1986,
p. 391), constitutes the central part of the theory. Students’ choices of science-related
activities, their efforts to perform them, and their persistence and resilience when
faced with obstacles are all affected by their self-efficacy beliefs. Students who
believe in their ability to be successful in science tend to select science-related activ-
ities, study hard to accomplish these activities, and do not give up in the face of diffi-
culties, whereas students who do not believe in their ability in science will be more
likely to avoid science activities (Bandura, 1997; Britner & Pajares, 2001).
A bidirectional relationship exists between self-efficacy and student achievement.
There is theoretical and empirical support for this assertion. Bandura (1997) stated that
students interpret the results of their previous experience with the task and develop
beliefs about their capability. In particular, students who perform successfully in execut-
ing a task will believe in their capability, thus their sense of efficacy will be enhanced.
Likewise, highly efficacious students will perform better. Accordingly, in the related lit-
erature, there are numerous studies providing evidence for the relationship between
science self-efficacy and science achievement (Andrew, 1998; Britner & Pajares, 2001;
Kupermintz, 2002; Lau & Roeser, 2002; Pajares, Britner, & Valiante, 2000). These stu-
dies revealed that science self-efficacy predicts science achievement better than vari-
ables like gender or parental background. In chemistry, Dalgety and Coll (2006)
reported that students did not have high self-efficacy in all areas of chemistry; students
believe less in their ability in advanced skills such as tutoring peers and designing experi-
ments. To enhance student achievement and engagement in chemistry, the self-efficacy
construct could be considered by researchers as a viable predictor of student achieve-
ment. The purpose of this study was to develop and validate scores on a measure of
chemistry self-efficacy for high school students.
To understand self-efficacy beliefs better, it is essential to understand how students
develop these beliefs. Bandura (1997) proposed that self-efficacy beliefs are shaped by
the cognitive processing and integration of four main sources of information. The most
significant source of efficacy information is mastery experience; students’ previous per-
formances influence their self-efficacy. In addition to mastery experience, students can
observe others perform the task and develop self-efficacy beliefs. This is another source
of self-efficacy beliefs, namely, vicarious experience. It has the greatest influence when
there are similarities between the model and observer. For example, models of students
with the same gender are more credible and have more power to affect self-efficacy
beliefs. Social persuasion—verbal and nonverbal judgments that students receive from
others—is also a significant source of efficacy information. These judgments encourage
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
870 Educational and Psychological Measurement
students to put forth more effort to succeed, leading to further development of their sense
of efficacy. Last, physiological states like anxiety, stress, and arousal shape self-efficacy
beliefs. Bandura (1997) hypothesized that students form their self-efficacy beliefs
through the interpretation of information from these four sources. In other words, ‘‘effi-
cacy beliefs are the product of cognitive processing of diverse sources of efficacy infor-
mation conveyed enactively, vicariously, socially, and physiologically’’ (Bandura,
1997, p. 115). In this cognitive process, students weigh and integrate the different infor-
mation from these sources and make their efficacy judgments accordingly. Self-efficacy
beliefs are different from outcome expectancy and self-concept (Bandura, 1997).
Although they are related and may provide a type of cue used by individuals to assess
their efficacy beliefs (Schunk, 1989, 1991), each reflects different views of oneself. Out-
come expectancy is the expected outcome of one’s actions. On the other hand, self-
efficacy beliefs include judgments about one’s ability to perform a task and they are task
and situation specific. As Bandura (1986) clarified, ‘‘The belief that one can high jump
six feet is an efficacy judgment; the anticipated social recognition, applause, trophies,
and self-satisfactions for such a performance constitute outcome expectations’’ (p. 391).
Self-concept is defined as individuals’ beliefs and opinions about themselves in general
and includes judgment of self-worth (Pajares & Schunk, 2001). The sentence ‘‘I am
good at chemistry’’ represents a self-concept belief, whereas the sentence ‘‘I can explain
the fundamental definitions in chemistry’’ is a self-efficacy belief because it is more spe-
cific and includes situational judgment of capabilities. Therefore, in measuring self-
efficacy beliefs, researchers should avoid using omnibus-type items that provide global
scores because they decontextualize the self-efficacy behavior into a generalized person-
ality trait (Bandura, 1986, 1997).
This study aimed to design and validate scores from a scale, namely, High
School Chemistry Self-Efficacy Scale (HCSS), to assess high school students’
beliefs in their ability to accomplish tasks related to chemistry. Specifically, the
aims of the current study were as follows: (a) to describe the development of the
HCSS, (b) to assess psychometric characteristics of scores from the HCSS, (c) to
determine whether the HCSS is factorially invariant with public and private school
students, and (d) to test the latent mean structure of the HCSS across school types.
In Turkey, it is observed that students in public and private schools have different
experiences in chemistry classes. Private schools are reported to have better facil-
ities such as fully equipped chemistry laboratories and resources for students and
teachers (Erdogan, 2002; Ministry of Education, 2007). Considering that mastery
experience is believed to be essential for cultivating self-efficacy (Bandura, 1997),
the present study also examined the factorial invariance of the scale with public
and private school students.
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Çapa Aydın, Uzuntiryaki / High School Chemistry Self-Efficacy 871
Method
Samples
In this study, two independent samples were used; Sample 1 was used for the
pilot study, and Sample 2 was used for the main study.
Instrumentation
Considering primarily the guide by Bandura (2006) about constructing self-
efficacy scales, an initial item pool of 22 items was generated based on related litera-
ture, experience, and review of high school chemistry curriculum. A 9-point scale
was chosen for the response format, where 1 corresponded to very poorly and 9 cor-
responded to very well. A higher score on this scale indicated higher self-efficacy in
chemistry. Items were developed in Turkish, and a specialist in Turkish language
examined grammar and clarity of the items. Necessary revisions were made.
For content validation purposes, a panel of 12 chemistry high school teachers
and researchers working in the field of chemistry, chemistry education, educational
psychology, and measurement reviewed the items in terms of content representa-
tiveness. The panel suggested including an item assessing the ability to explain the
particulate nature of matter because it is one of the essential topics covered in high
schools. Moreover, an item related to self-efficacy in identifying measurement
units was removed due to its mathematical content.
Pilot Study
The initial scale consisting of 22 items was piloted with 150 high school chemis-
try students (Sample 1). In an effort to explore the factorial structure of the scale, an
exploratory factor analysis (with principal axis factoring and direct oblimin rotation
with delta set at 0) was conducted through the SPSS. The Kaiser-Meyer-Olkin mea-
sure of sampling adequacy was .89, which suggested that data were appropriate for
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
872 Educational and Psychological Measurement
• Chemistry self-efficacy for cognitive skills (CSCS; 10 items): This factor refers to
students’ beliefs in their ability to use intellectual skills in chemistry. Sample items
in this factor read, ‘‘How much can you describe the structure of an atom?’’ and
‘‘To what extent can you explain chemical laws and theories?’’
• Self-efficacy for chemistry laboratory (SCL; 6 items): This factor refers to students’
beliefs in their ability to accomplish laboratory tasks including skills in both cogni-
tive and psychomotor domain. Sample items read, ‘‘How well can you use the
equipment in the chemistry laboratory?’’ and ‘‘How well can you write a laboratory
report summarizing main findings?’’
The correlation between CSCS and SCL was found to be .55. The internal consis-
tency of the two scale scores was estimated by the Cronbach alpha coefficient,
yielding a coefficient of .84 for chemistry self-efficacy for cognitive skills and .94
for self-efficacy for chemistry laboratory.
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Çapa Aydın, Uzuntiryaki / High School Chemistry Self-Efficacy 873
Procedure
The HCSS was administered to the students in class at the end of school year by
a graduate student, who was knowledgeable about the research. Participation was
voluntary, and no incentive was used for completing the scale. Administration
required 10 to 15 min.
Data Analysis
Confirmatory factor analysis (CFA) was conducted to test a two-factor model
of the HCSS as revealed in the pilot study. Next, the factorial invariance of the
scale with public and private school students was examined. These analyses
were performed using Analysis of Moment Structures 4.0 (Arbuckle & Wothke,
1999). While evaluating model fit, Hoyle (1995) suggested examining several fit
indices rather than using only chi-square statistics, which are sensitive to sample
size. Therefore, in this study, multiple indices of fit were examined to supple-
ment chi-square statistics: the comparative fit index (CFI), the non-normed fit
index (NNFI), and root mean square error of approximation (RMSEA) along
with its 90% confidence intervals (CIs). The CFI and NNFI are classified as
incremental fit indices as they assess the amount of improvement in fit by com-
paring the target model to a more restricted baseline model. Values greater than
.90 indicate an adequate model fit (Bentler, 1990; Hu & Bentler, 1995; Kline,
1998), although values of .95 or greater are preferable (Hu & Bentler, 1999).
The RMSEA, which is an example of absolute fit index, asks the question,
‘‘How well would the model, with unknown but optimally chosen parameter
values, fit the population covariance matrix if it were available?’’ (Browne &
Cudeck, 1993, pp. 137–138). MacCallum and Austin (2000) strongly recom-
mended reporting RMSEA because of sensitivity to model misspecification,
availability of effective guidelines for interpretation, and availability of a confi-
dence interval. Values of RMSEA lower than .05 indicate a close fit, values
between .05 and .08 indicate mediocre fit, and values greater than .10 indicate
poor fit (Browne & Cudeck, 1993). MacCallum, Browne, and Sugawara (1996)
consider values in the range of .08 and .10 as mediocre fit.
In addition, while testing factorial invariance, a chi-square difference test was
performed between models with different restrictions. Nonsignificant change in
chi-square statistic indicates invariance of the factorial structure (Byrne, 2001).
However, several researchers (Cheung & Rensvold, 2002; M. S. Thompson &
Green, 2006) proposed that w2 is strongly influenced by sample size and non-
normality like the w2 statistic itself and recommended using other fit indices while
evaluating differential fit. CFI is the suggested index as it provides the best
information in providing evidence of invariance. A change of CFI equal to or less
than .01 would indicate invariance (Chen, 2007; Cheung & Rensvold, 2002).
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
874 Educational and Psychological Measurement
Results
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Çapa Aydın, Uzuntiryaki / High School Chemistry Self-Efficacy 875
Table 1
Factor Pattern (and Structure) Coefficients for the Two-Factor
Model of the High School Chemistry Self-Efficacy Scale
Items Factor 1: CSCS Factor 2: SCL
Note: Italicized numbers are the factor pattern coefficients for each item with its designated factor. Noni-
talicized numbers are the factor structure coefficients of each item with its nondesignated factors. For
the present model, the factor pattern and factor structure coefficients are equal for the items with their
designated factors. All non-zero pattern coefficients were found to be statistically significant (p < .01).
CSCS = chemistry self-efficacy for cognitive skills; SCL = self-efficacy for chemistry laboratory.
more restrictive model in which the factor pattern coefficients were constrained to
equality across groups. Comparison of this model with the null model yielded a
nonsignificant difference in chi-square, w2 (14) = 11.15, and .001 change in CFI,
which indicated that the hypothesis of an invariant pattern of factor coefficients
was considered acceptable. Considering the differences in the learning environ-
ments in private and public schools, two additional runs were generated. In one of
them, only factor pattern coefficients associated with ‘‘self-efficacy for chemistry
laboratory’’ were constrained to be equal, while in the other only factor pattern
coefficients for ‘‘chemistry self-efficacy for cognitive skills’’ were restricted. Both
produced nonsignificant chi-square and less than .01 CFI change as shown in Table
2. The final CFA was used to test the invariance of factor variance and covariance
across groups. Comparison of this model, in which all factor pattern coefficients and
factor variance/covariance were restricted to be equal, with the null model supported
the equivalence of parameters across school types as well. Please note that consider-
ing the fact that invariance of error variances across groups is highly restrictive
(Byrne, 1998; MacCallum et al., 1994), a model with restricted error variances was
not tested. Overall, these findings suggested that the measurement model for the
HCSS was invariant across school types.
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
876 Educational and Psychological Measurement
Table 2
Simultaneous Tests for Factorial Invariance of the
High School Chemistry Self-Efficacy Scale
Models w2 df w2 df CFI CFI
Note: None of the w2 was significant at .05. CSCS = chemistry self-efficacy for cognitive skills;
SCL = self-efficacy for chemistry laboratory; CFI = comparative fit index.
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Çapa Aydın, Uzuntiryaki / High School Chemistry Self-Efficacy 877
obtained in this study were quite high for both of the factors: a = .90 (CI = .89,
.92) for the CSCS factor and a = .92 (CI = .91, .93) for the SCL factor (see Henson,
2001). Item-total correlations ranged from .71 to .80 for SCL and ranged from .54
to .72 for CSCS, which implied that all items in the instrument contributed to the
consistency of scores.
Discussion
The purpose of the present study was to develop and score validate scores on a
scale (HCSS) measuring high school students’ chemistry self-efficacy. After a thor-
ough literature search, content examination by experts, and a pilot study with 150
students to examine the underlying factor structure and to refine the scale, the final
form of HCSS comprising 16 items with two dimensions was administered to 362
high school students for cross-validation purposes. Confirmatory factor analysis
was used to determine whether the data confirmed the model. Interpretation of var-
ious fit indices and the factor pattern coefficients supported the proposed structure,
with the two factors directly corresponding to the hypothesized dimensions. These
dimensions were chemistry self-efficacy for cognitive skills and self-efficacy for
chemistry laboratory. In terms of internal consistency, reliability coefficients
reported for scores on the dimensions (0.90 for chemistry self-efficacy for cognitive
skills; 0.92 for self-efficacy for chemistry laboratory) were deemed acceptable. In
addition, considering different classroom environments in public and private
schools, the factorial invariance of the scale was also investigated. Results indi-
cated that the two-factor model for the HCSS was invariant across school types. In
other words, the scale can be used in both settings without any change.
Analysis of the latent mean structure was conducted to examine whether public
and private school students differed in terms of the two factors of the HCSS. Pri-
vate school students were statistically significantly more efficacious than public
school students in chemistry laboratory skills, at least for this sample of high school
students in Turkey. This finding was consistent with the prediction that private
schools provide better laboratory facilities. Thus students in private schools might
have more experience in chemistry laboratory tasks such as conducting an experi-
ment, writing a laboratory report, and using laboratory equipment. This would cul-
tivate their self-efficacy because mastery experience is the most influential source,
as postulated by Bandura (1997).
The HCSS developed in this study may enable researchers and teachers to identify
chemistry self-efficacy beliefs in high school students. As suggested by research stu-
dies (Andrew, 1998; Britner & Pajares, 2001; Kupermintz, 2002; Lau & Roeser,
2002; Pajares et al., 2000), there is a relationship between self-efficacy and science
achievement. However, because of task-specific nature of self-efficacy, these findings
cannot be generalized to other domains. It is also important to examine additional
criterion-related validity relationships. Researchers can examine the relationship
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
878 Educational and Psychological Measurement
between chemistry self-efficacy and chemistry achievement. In this sense, the HCSS
may be able to be used as a diagnostic tool. Then, using Bandura’s assertion related
to sources of self-efficacy, researchers can design instructional strategies and test
their effectiveness on chemistry self-efficacy utilizing experimental designs. Further
validation with different populations is also warranted to increase external validity.
Appendix
The High School Chemistry Self-Efficacy Scale (HCSS)
Directions: This questionnaire is designed to help
us gain a better understanding of the kinds of things
that create difficulties for students in chemistry.
very poorly
Please indicate your opinion about each of the
very well
statements below. Please do not skip any item.
average
poorly
Your answers are confidential:
well
THANKS FOR YOUR HELP J
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
Çapa Aydın, Uzuntiryaki / High School Chemistry Self-Efficacy 879
References
Andrew, S. (1998). Self-efficacy as a predictor of academic performance in science. Journal of
Advanced Nursing, 27, 596-603.
Arbuckle, J. L., & Wothke, W. (1999). Amos 4.0 User’s Guide. Chicago: SPSS.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood
Cliffs, NJ: Prentice Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares & T. Urdan (Eds.), Self-
efficacy beliefs of adolescents (pp. 307-337). Greenwich, CT: Information Age.
Bentler, P. M. (1990). Comparative fit indices in structural models. Psychological Bulletin, 107, 238-246.
Britner, S. L., & Pajares, F. (2001). Self-efficacy beliefs, motivation, race, and gender in middle school
science. Journal of Women and Minorities in Science and Engineering, 7, 271-285.
Britner, S. L., & Pajares, F. (2006). Sources of science self-efficacy beliefs of middle school students.
Journal of Research in Science Teaching, 43, 485-499.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S.
Long (Eds.), Testing structural equation models (pp. 136-162). Newbury Park, CA: Sage.
Byrne, B. M. (1998). Structural equation modeling with LISREL, PRELIS, and SIMPLIS: Basic con-
cepts, applications, and programming. Mahwah, NJ: Lawrence Erlbaum.
Byrne, B. M. (2001). Structural equation modeling with AMOS. Mahwah, NJ: Lawrence Erlbaum.
Carter, C. S., & Brickhouse, N.W. (1989). What makes chemistry difficult? Alternate perceptions. Jour-
nal of Chemical Education, 66, 223-225.
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural
Equation Modeling, 14, 464-504.
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement
invariance. Structural Equation Modeling, 9, 233-255.
Dalgety, J., & Coll, R. K. (2006). Exploring first-year science students’ chemistry self-efficacy. Interna-
tional Journal of Science and Mathematics Education, 4, 1-20.
Erdogan, I. (Ed.). (2002). Özel okullar ve e gitimde kalite [Private schools and quality in education].
_
Istanbul: Özel Okullar Dernegi.
Fan, X., & Thompson, B. (2001). Confidence intervals about score reliability coefficients, please: An
EPM guidelines editorial. Educational and Psychological Measurement, 61, 517-531.
Gopal, H., Kleinsmidt, J., Case, J., & Musonge, P. (2004). An investigation of tertiary students’ under-
standing of evaporation, condensation and vapour pressure. International Journal of Science Educa-
tion, 26, 1597-1620.
Graham, J. M., Guthrie, A. C., & Thompson, B. (2003). Consequences of not interpreting structure coef-
ficients in published CFA research: A reminder. Structural Equation Modeling, 10, 142-153.
Hancock, G. R. (1997). Structural equation modeling methods of hypothesis testing of latent variable
means. Measurement and Evaluation in Counseling and Development, 30, 91-105.
Hancock, G. R. (2001). Effect size, power, and sample size determination for structured means modeling
and mimic approaches to between-groups hypothesis testing of means on a single latent construct.
Psychometrika, 66, 373-388.
Henson, R. K. (2001). Understanding internal consistency reliability estimates: A conceptual primer on
coefficient alpha. Measurement and Evaluation in Counseling and Development, 34, 177-189.
Henson, R. K., & Roberts, J. K. (2006). Use of exploratory factor analysis in published research: Com-
mon errors and some comment on improved practice. Educational and Psychological Measurement,
66, 393-416.
Horn, J. L. (1965). A rationale and test for the number of factors in factor analysis. Psychometrika, 30,
179-185.
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013
880 Educational and Psychological Measurement
Hoyle, R. H. (1995). Structural equation modeling: Concepts, issues, and applications. Thousands Oaks,
CA: Sage.
Hu, L. T., & Bentler, P. M. (1995). Evaluating model fit. In R. Hoyle (Ed.), Structural equation model-
ing: Concepts, issues, and applications (pp. 76-99). Thousands Oaks, CA: Sage.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indices in covariance structure analysis: conven-
tional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.
Kennedy, E. (1996). What do they think of chemistry? Australian Science Teachers Journal, 42(2), 53-59.
Kline, R. B. (1998). Principles and practice of structural equation modeling. New York: Guilford.
Kupermintz, H. (2002). Affective and conative factors as aptitude resources in high school science
achievement. Educational Assessment, 8, 123-137.
Lau, S., & Roeser, R.W. (2002). Cognitive abilities and motivational processes in high school students’
situational engagement and achievement in science. Educational Assessment, 8, 139–162.
MacCallum, R. C., & Austin, J. T. (2000). Applications of structural equation modeling in psychological
research. Annual Review of Psychology, 51, 201-226.
MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis and determination of
sample size for covariance structure modeling. Psychological Methods, 1, 130-149.
MacCallum, R. C., Roznowski, M., Mar, C. M., & Reith, J. V. (1994). Alternative strategies for cross-
validation of covariance structure models. Multivariate Behavioral Research, 29, 1-32.
Ministry of Education (MEB). (2007). PISA 2006 Ulusal ön rapor [National Preliminary Report].
Ankara: Author. Retrieved May 10, 2008, from http://earged.meb.gov.tr/pisa/dokuman/2006/rapor/
Pisa_2006_Ulusal_On_Rapor.pdf
O’Connor, B. P. (2000). SPSS and SAS programs for determining the number of components using par-
allel analysis and Velicer’s MAP test. Behavior Research Methods, Instruments, & Computers, 32,
396-402.
Pajares, F., Britner, S.L., & Valiante, G. (2000). Relation between achievement goals and self-beliefs of
middle school students in writing and science. Contemporary Educational Psychology, 25, 406-422.
Pajares, F., & Schunk, D.H. (2001). Self-beliefs and school success: Self-efficacy, self-concept, and
school achievement. In R. J. Riding & S. G. Rayner (Eds.), Perception (pp. 239–266). London:
Ablex.
Schunk, D. H. (1989). Self-efficacy and cognitive achievement: Implications for students with learning
problems. Journal of Learning Disabilities, 22, 14-22.
Schunk, D. H. (1991). Self-efficacy and academic motivation. Educational Psychologist, 26, 207-231.
Thompson, B. (1997). The importance of structure coefficients in structural equation modeling confirma-
tory factor analysis. Educational and Psychological Measurement, 57, 5-19.
Thompson, M. S., & Green, S. B. (2006). Evaluating between-group differences in latent variable means. In
G. R. Hancock & R. O. Mueller (Eds.), Structural equation modeling: A second course (pp. 119-169).
Greenwich, CT: Information Age.
Tsaparlis, G. (1997). Atomic orbitals, molecular orbitals, and related concepts: Conceptual difficulties
among chemistry students. Research in Science Education, 27, 271-287.
Watson, J., & McEwen, A. (1994). Sixth form A level students’ perceptions of the difficulty, intellectual
freedom, social benefit and interest of science and arts subjects. Research in Science & Technologi-
cal Education, 12, 43-49.
Zwick, W. R., & Velicer, W. F. (1986). Comparison of the rules for determining the number of compo-
nents to retain. Psychological Bulletin, 99, 432-442.
Downloaded from epm.sagepub.com at Scientific library of Moscow State University on December 27, 2013