MURAT BURSAL

TURKISH PRESERVICE ELEMENTARY TEACHERS’SELF-EFFICACY BELIEFS REGARDING MATHEMATICS

AND SCIENCE TEACHING

Received: 23 February 2008; Accepted: 17 August 2009

ABSTRACT. This study investigated Turkish preservice, elementary teachers’ personalmathematics teaching efficacy (PMTE), and science teaching efficacy (PSTE) beliefs atthe end of their teacher education program. A majority of the participants believed theywere well prepared to teach both elementary mathematics and science, but their PSTEscores were significantly lower than their PMTE scores. However, a significant correlationwas found between the PMTE and PSTE scores. No significant gender effect on PMTEand PSTE scores was observed, but unlike the results from other countries, Turkish femalepreservice elementary teachers were found to have slightly higher PMTE and PSTE scoresthan their male peers. High school major area was found to be a significant predictor ofparticipants’ PMTE and PSTE scores. Participants with mathematics/science high schoolmajors were found to have significantly higher PMTE and PSTE scores than those withother high school majors.

KEY WORDS: high school major area, personal mathematics teaching efficacy, personalscience teaching efficacy, preservice elementary teacher

INTRODUCTION

Previous research on student attitudes toward mathematics andscience indicates that teachers are the most important single influenceon students’ attitudes toward these areas, and one fourth of theseinfluential teachers are reported to be the elementary school teachers(Cox & Carpenter, 1989; Putney & Cass, 1998). Since many oftomorrow’s teachers are today’s preservice teachers, the beliefs theyhold should be of concern to teacher educators. The quality ofmathematics and science instruction at the elementary school leveldepends on the preparation of preservice elementary teachers withappropriate content knowledge and pedagogical content knowledgeand also positive beliefs about teaching these courses (Briscoe &Stout, 2001; Harper & Daane, 1998). Without sufficient knowledge,enthusiasm, and self-efficacy in these areas, it is unlikely that futureelementary teachers will be able to provide effective instruction(Battista, 1986; Stevens & Wenner, 1996; Tosun, 2000).

International Journal of Science and Mathematics Education (2010) 8: 649Y666# National Science Council, Taiwan 2009

There has been a great deal of concern expressed in the literature aboutpreservice elementary teachers’ low level of mathematical and scienceknowledge. For example, it was repeatedly reported that preserviceelementary teachers have problems in simple mathematical skills (e.g.,interpreting decimals) and operating with decimals and fractions (Battista,1986; Graeber, Tirosh, & Glover, 1989; Quinn, 1997; Thipkong & Davis,1991). Similarly, science education researchers noted that elementaryteacher candidates possess a low level of knowledge of the concepts,facts, and skills concerning science (Stevens & Wenner, 1996; Tekkaya,Cakiroglu, & Ozkan, 2004; Tosun, 2000; Wenner, 1993). Furthermore,preservice elementary teachers were found to have problems inconstructing and expressing mathematical relationships from scientificdata (Briscoe & Stout, 2001).

Numerous researchers have concluded that the weak mathematicsand science backgrounds of preservice elementary teachers signifi-cantly contributed to their lack of content knowledge in these areas.However, overcoming these problems does not merely depend onincreasing the content of coursework that teacher candidates take incollege; it is well documented that increasing the amount of contenthas little effect on preservice teachers’ confidence to teach(Hadfield, Littleton, Steiner, & Woods, 1998; Palmer, 2006; Stevens& Wenner, 1996; Young & Kellogg, 1993). In light of the previousstudies, teacher candidates should be taught in environments wherethey not only learn the mathematics and science content but alsoenhance their self-efficacy beliefs about teaching these disciplines(Cantrell, Young, & Moore, 2003; Harper & Daane, 1998; Putney &Cass, 1998; Stevens & Wenner; Swars, Daane, & Giesen, 2006;Tosun, 2000).

The term self-efficacy was introduced by Bandura (1977) as part of hissocial cognitive theory. He used the concept of reciprocal determinism toexplain that each behavioral, personal, and environmental factor influences—and is also influenced by—the other factors. Later, he argued that, “People’slevel of motivation, affective states, and actions are based more on what theybelieve than on what is objectively true.” (1997, p. 2). Self-efficacy representsthe belief that an individual possesses and is defined as, “the beliefs in one’scapability to organize and execute the courses of action required to producegiven attainments” (p. 3). Bandura emphasized that self-efficacy beliefs aresituation-specific; therefore, people undertake and perform the activities thatthey believe they have the capability of handling but tend to avoid situationsthat are believed to challenge their capabilities.

MURAT BURSAL650

Mathematics and Science Teaching Self-efficacy

Ashton and Webb (1986) adapted the concepts of personal teachingefficacy, or teaching self-efficacy, which are defined as a person’s beliefin his or her ability to teach effectively. Since the context and subjectmatter are likely to affect one’s self-efficacy beliefs, teaching efficacydefinitions have been extended to specific subject areas. In the areas ofmathematics and science teaching, teaching self-efficacy beliefs aredefined as personal mathematics teaching efficacy (PMTE) and personalscience teaching efficacy (PSTE). PMTE beliefs refer to the extent thatteachers believe they have the capacity to affect students’ mathematicsachievement positively, and PSTE refers to similar teacher beliefs thataffect students’ science achievement (Cantrell et al., 2003; Moore &Watson, 1999).

Since the introduction of the concept of teaching self-efficacy to theliterature, there has been a growing interest in discovering the impact ofthese beliefs in mathematics and science education. This concern is ofutmost importance in elementary education where elementary teachers areexpected to teach all subjects in their classrooms, but it is highly unlikelythat they are equally well prepared to teach or feel positive about teachingall subjects (Bauer & Toms, 1990; Ramey-Gassert & Shroyer, 1992;Sherwood & Westerback, 1983).

Research on Mathematics and Science Teaching Self-efficacy Beliefs

It has been repeatedly cited that preservice and inservice elementaryteachers lack mathematics and science knowledge and hold negativebeliefs about mathematics and science, which results in anxiety, poorattitudes toward these areas, and an unwillingness or hesitancy to teachthese disciplines (Feistritzer & Boyer, 1983; Hadfield, Littleton, Steiner,& Woods, 1998; Harper & Daane, 1998; Ramey-Gassert & Shroyer,1992). These findings lead to the conclusion that elementary teachers’negative attitudes toward mathematics and science adversely affect theirteaching self-efficacy beliefs about these disciplines, which in turn,eventually leads to ineffective instruction. Low self-efficacious teachersare cited as relying heavily on the use of teacher-directed instruction, suchas text-based instruction or lecturing, which is characterized by anauthoritative, teacher-centered role. Due to their low levels of confidencein their effectiveness, these teachers may avoid inquiry experiences anduse of manipulatives to prevent any challenging outcomes (Downing,

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 651

Filer, & Chamberlain, 1997; Putney & Cass, 1998; Rule & Harrell, 2006;Swars et al., 2006).

On the other hand, there is ample evidence that highly self-efficaciousteachers use more inquiry and student-centered teaching strategies(Marshall, Horton, Igo, & Switzer, 2009). These teachers feel confidentthat they have adequate training or experience to implement teachingstrategies for overcoming barriers to student learning. Highly self-efficacious teachers were found to be more effective and more likely tospend the time needed to develop concepts. In addition, their studentswere reported to have demonstrated more positive attitudes and achievedhigher performance levels on achievement tests (Cantrell et al., 2003;Riggs, 1991; Utley, Moseley, & Bryant, 2005).

An inadequate content course background was commonly given as themain reason for preservice elementary teachers’ hesitance or inability toteach mathematics and science (Bleicher & Lindgren, 2009; Tooke &Lindstrom, 1998), but the relationships between previous collegemathematics and science courses and preservice elementary teachers’attitudes toward these areas are not clearly established in the literature.Manning, Esler, and Baird (1982) found a significant positive relationshipbetween the number of college science courses taken and attitudes towardteaching science of prospective teachers; however, Stepans and McCormack(1985) reported a negative relationship, and Westerback (1982) andFeistritzer and Boyer (1983) found no relationship. Similarly, Wenner(1993) and Stevens and Wenner (1996) recorded nonsignificant correlationsbetween attitudes toward teaching mathematics and science and the numberof college mathematics and science courses taken.

Since females constitute the majority of elementary teachers in manycountries, the impact of preservice teachers’ gender on their attitudestoward mathematics and science is suggested as another reason for thenegative attitudes of elementary teachers towards teaching these topics. Ithas been reported that male elementary teachers express higher self-efficacy for teaching mathematics and science than female teachers inboth inservice and preservice situations (Brownlow, Jacobi, & Rogers,2000; Ho, Senturk, Lam, Zimmer, Hong, Okamoto, et al. 2000; Riggs,1991; Zettle & Raines, 2000). Brownlow et al. stated, “Regardless of theiractual science performance, women are much less self-confident in theircapabilities with science, and report much more science anxiety than domen.” (p. 120). Ho et al. conducted a meta-analysis of 151 studies andfound that female preservice elementary teachers possessed less favorableattitudes toward mathematics than did males. However, more studies thatare recent report nonsignificant differences between the attitudes of male

MURAT BURSAL652

and female teacher candidates in mathematics and science; this could be apossible result of the focus on improving female students’ attitudestoward these areas (Cantrell et al., 2003; Chin, 2005; Mulholland,Dorman, & Odgers, 2004).

Compared to the growing literature about preservice elementaryteachers’ beliefs and the effectiveness of teacher preparation programsto enhance teacher candidates’ teaching self-efficacy levels elsewhere, theresearch on future Turkish elementary teachers’ beliefs regardingmathematics and science teaching is relatively new and incomplete.There are only a few studies where the science teaching self-efficacybeliefs of Turkish preservice elementary teachers were investigated. Inthese studies, elementary teacher candidates were found to have moderatelypositive self-efficacy beliefs (Sarikaya, Cakiroglu, & Tekkaya, 2005). Onthe other hand, it was reported that Turkish preservice elementary teachershave lower science teaching self-efficacy than their American peers(Cakiroglu, Cakiroglu, & Boone, 2005).

Similar to recent studies in other countries, researchers reportednonsignificant differences between Turkish male and female preserviceteachers’ beliefs about teaching mathematics and science (Gencer &Cakiroglu, 2007; Isiksal, 2005; Isiksal & Cakiroglu, 2005). Turkishteacher candidates were generally found to hold positive attitudes towardteaching these disciplines in the future; but at the same time, theyexpressed concern regarding their weak content backgrounds, especiallyin science (Sarikaya et al., 2005). Very similar elementary teacherpreparation programs occur across Turkey in which all teacher candidatesgenerally take the same number of mathematics and science courses;therefore, there is little or no variability in the number of collegemathematics or science credits among these students. However, Turkishteacher candidates would have completed different numbers of mathe-matics and science courses in their high school program, which requiresstudents to select disciplinary specialties called “majors.” Sarikaya et al.suggested that teacher candidates’ high school background (being amathematics/science major versus another major) would have a keyimpact on their attitudes toward mathematics and science.

The criterion for entering into different college programs in Turkeydepends on each student’s score in the Student Selection Examination[Öğrenci Seçme Sınavı, ÖSS], which is administered annually to highschool graduates. This 3.5-h test consists of 90 mathematics/science and90 Turkish/social sciences multiple-choice questions. The examinationscore is calculated by using the number of correct responses as well as thehigh school major grades and grade point average. Depending on the high

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 653

school type (general school, science school, vocational school, teacherschool, etc.) and the selected major in a high school (science/mathematics, social sciences, Turkish/mathematics), Turkish studentsare directed to apply for programs within their major. As for applyingto a program from another major, there is not any restriction, but asignificant portion of the ÖSS score is reduced when students apply for acollege program outside of their high school major or their school’sspecialty area (Yüksek Öğretim Kurulu, YÖK, 1998).

Turkish high school curricula for different majors are prepared by theMinistry of National Education [Milli Eğitim Bakanlığı, MEB] and mustbe implemented by all Turkish high schools. Based on these commoncurricula (MEB, 2006), a typical mathematics/science major takesmathematics and science courses every year in high school, while othermajors do not take any science course after Grade 9. Other major studentscontinue to take mathematics classes after Grade 9, but they usually takefewer credits than mathematics/science majors do. Therefore, one canconclude that the high school major determines future career options andthe content area backgrounds of the mathematics/science majors and othermajors differ drastically.

Turkish Elementary Teacher Preparation Programs

Elementary school teaching qualification in Turkish universities involves4-year undergraduate programs with a common curriculum designed bythe Yüksek Öğretim Kurulu (Higher Education Council). Teachercandidates take introductory pedagogical and content courses in theirfirst 2 years prior to taking methods courses. The participants in this studyhad taken two mathematics and five science courses: basic mathematics-I(two credits), basic mathematics-II (two credits), biology (three credits),chemistry (three credits), physics (three credits), environmental science(two credits), and science laboratory (two credits). These college-levelcontent courses covered the topics and fundamental principles addressedin the elementary mathematics and science curricula.

Teaching methods courses in elementary reading and writing, science,mathematics, and social studies are offered in the third year of theprogram as two-course (fall and spring) sequences. The year aftercompleting the methods courses, senior preservice teachers start studentteaching. A teaching practicum is a seminar course where preserviceteachers practice in schools and share their experiences in the collegeclassroom. Teacher candidates are required to attend elementary schoolsfor one full day or two half days for a minimum of 12 weeks. Upon

MURAT BURSAL654

successful completion of the undergraduate program within 8 years,teacher candidates receive an elementary school teaching license to teachGrades 1–5 in all public and private elementary schools (YÖK, 1998).

The research questions investigated in this study are:

1. What are Turkish preservice elementary teachers’ self-efficacy beliefsabout teaching mathematics and science at the time they graduate fromthe teacher education program?

2. Do female and male Turkish preservice elementary teachers differ intheir self-efficacy beliefs about teaching mathematics and/or science?

3. Does the high school major of Turkish preservice elementary teachersaffect their self-efficacy beliefs about teaching mathematics and/orscience?

METHODOLOGY

This study used a case approach to investigate preservice elementaryteachers’ beliefs about teaching mathematics and science. The studentscame from the same university and same teacher education program butentered the program with different high school majors.

Participants

The sample consisted of 127 Turkish preservice elementary teachers from acentral-Anatolian Turkish university. Sixty-five (51%) of the participants weremale, and 62 (49%) were female. Fifty-seven students (45%) were high schoolmathematics/science majors, whereas 70 (55%) were from other majors. Theparticipants were in their final semester of a 4-year undergraduate program.All participants had taken the same number of mathematics/science contentandmethods courses required by their program and completed a one-semester-long student-teaching practicum prior to this study.

Measurement Instruments

This study is limited to exploring beliefs about teaching science andmathematics. The Science Teaching Efficacy Belief Instrument (STEBI-B, Enochs & Riggs, 1990), a widely used instrument to assess self-efficacy beliefs of preservice teachers regarding science instruction inschools, was selected as the foundation for this study. The instrumentconsists of two subscales: PSTE and science teaching outcome expectan-cy (STOE). The PSTE consists of 13 Likert-type items, and the STOE

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 655

consists of 10 Likert-type items designed to document respondents’ levelof agreement or disagreement with a statement on a five-point scale (1=strongly disagree to 5=strongly agree). Only the PSTE subscale was usedin this study. The reliability, internal consistency, and alpha coefficient ofthe PSTE scale was reported to be 0.90 (Enochs & Riggs, 1990).

The Mathematics Teaching Efficacy Belief Instrument (MTEBI-B,Enochs, Smith, & Huinker, 2000) was developed by modifying theSTEBI-B to assess self-efficacy beliefs of preservice teachers regardingmathematics instruction in schools. Similar to the STEBI, the MTEBIconsists of two subscales: PMTE and mathematics teaching outcomeexpectancy (MTOE). The PMTE consists of 13 items, but unlike the STOEsubset of STEBI, the MTOE consists of eight items; two items from thisscale were dropped after the factor analysis to explore the structuralvalidity. Again, each PMTE item documents the respondent’s agreement/disagreement on a five-point scale, and only the PMTE subscale was usedin this study. The reliability, internal consistency, and α coefficient of thePMTE subscale was reported to be 0.88 (Enochs et al., 2000).

This study included only the personal teaching efficacy subsets (PSTEand PMTE) of the STEBI and MTEBI, because the goal of the study wasto investigate the teaching self-efficacy beliefs of participants inelementary mathematics and science. In addition, research since theintroduction of the STEBI and MTEBI has often cited that the outcomeexpectancy construct is usually viewed to be inconsistent amongpreservice teachers as opposed to the consistent results from the self-efficacy scales. Enochs and Riggs (1990) noted that outcome expectancyis a difficult construct to measure due to its complex nature. Mulhollandet al. (2004) argued that one of the reasons for the low reliability of theoutcome expectancy scales in recent studies is the evolving view ofteachers’ roles in today’s classrooms, which is likely to be much greaterfor preservice teachers. According to these authors, the outcomeexpectancy items emphasize a teacher-centered approach, while student-centered teaching is highly encouraged in current teacher preparationprograms. Mulholland et al. concluded that, “within this new mindset,teachers are learning facilitators and it is plausible that contemporarypreservice teachers cannot see the relevance of STOE items” (p. 327).

The STEBI-B and MTEBI-B were translated into Turkish by theauthor and a Turkish graduate student majoring in elementary scienceeducation at an American university. Turkish versions of the surveys werethen back-translated into English with the help of another Turkishgraduate student majoring in secondary science education at the sameAmerican university. Examination of the original versions and the back-

MURAT BURSAL656

translated versions by these experts indicated that the Turkish translationsof the surveys were parallel to the original surveys. The score range forthe 13-item PMTE and PSTE subscales were the same (13–65). Thereliability and validity of the Turkish versions of these surveys wereexplored with data from 140 Turkish preservice elementary teachers.Cronbach’s α coefficients were found to be 0.91 for the PSTE and 0.90for the PMTE scales. Confirmatory factor analyses to explore thestructural validity revealed that all PSTE items had factor loadings higherthan 0.36 in the Turkish version of the STEBI-B, and all PMTE items hadfactor loadings higher than 0.47 in the Turkish version of the MTEBI-B.

Data Collection and Data Analysis

The Turkish versions of the STEBI-B and the MTEBI-B wereadministered to all participants in the final week of the spring semester,after their completion of student teaching. All negative-worded itemswere reflected (5=1, 4=2, 3=3, 2=4, 1=5) to provide a consistent scorefor all items. Descriptive statistics were calculated to provide percentageresponse patterns, means, and standard deviations. Paired-sample t testswere used to compare the PSTE and PMTE scores, and 2x2 ANOVAtests were run to investigate possible gender and high school major effectson teaching self-efficacy scores. Cohen’s d effect sizes were used toexplore the standardized differences between the subgroups in the sample.

RESULTS

The percentages of participants who agreed/strongly agreed with thepersonal teaching efficacy items of the MTEBI and STEBI are given inTables 1 and 2. The original forms of the items were used in the study,but negative items are rewritten in positive form in Tables 1 and 2 forconvenience in reporting. While the majority of the participants expressedconfidence in teaching both elementary mathematics and science, theagreement percentages for the MTEBI items were higher than those forthe STEBI items. Compared to agreement percentages greater than 80%for all MTEBI items, more than 20% of the participants did not agree thatthey will teach science as well as other subjects (item 2); that they knewthe necessary steps to teach science concepts (item 3), and they will havethe skills to teach science in the future (item 9). Furthermore, compared to82% agreement with MTEBI item 10 (Will invite the principal to evaluate

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 657

mathematics teaching), only 61% expressed the same level of confidencefor their science teaching.

The PMTE and PSTE scores by gender and high school major and thenumber of students in each group are given in Tables 3 and 4. As wasseen in the comparison of Tables 1 and 2, comparison of the

TABLE 1

Percentages of participants who agreed or strongly agreed with the MTEBI items

Item Agreement (%)

Will find better ways to teach mathematics 96Will teach mathematics as well as other subjectsa 92Knows how to teach mathematical concepts 86Will be effective in monitoring mathematics activitiesa 91Will teach mathematics effectivelya 96Understands mathematical concepts 97Will be able to explain how mathematical solutions worka 90Will be able to answer students’ mathematics questions 98Will have the skills to teach mathematicsa 95Will invite the principal to evaluate mathematics teachinga 82Will be able to help students understand mathematical conceptsa 95Will welcome student questions 98Knows what to do to turn students on to mathematicsa 94

aNegative item rewritten in positive form

TABLE 2

Percentages of participants who agreed or strongly agreed with the STEBI items

Item Agreement (%)

Will find better ways to teach science 91Will teach science as well as other subjectsa 74Knows the steps to teach science concepts 68Will be effective in monitoring experimentsa 80Will teach science effectivelya 83Understands science concepts 85Will be able to explain science experimentsa 84Will be able answer students’ science questions 87Will have the skills to teach sciencea 79Will invite the principal to evaluate science teachinga 61Will be able to help students understand science conceptsa 86Will welcome student questions 97Knows what to do to turn students on to sciencea 83

aNegative item rewritten in positive form

MURAT BURSAL658

corresponding cell means from Tables 3 and 4 indicates that the meanPMTE score for each subgroup (male, female, mathematics/sciencemajor, other major) was higher than the corresponding mean PSTE score.The paired-sample t test result (pG .001) and the large effect size (d=0.64)value for the comparison of the overall mean PMTE (57.28) and PSTE(52.50) scores show that the mean PMTE score was significantly higherthan the mean PSTE score. However, a significant (pG .01) positivecorrelation (r=.54) was calculated between the PMTE and PSTE scores.

Comparison of the gender/major cell means reveals that (a) mathematics/science major females had higher mean scores than mathematics/sciencemajor males and (b) females with other high school majors had slightly lowermean scores than their male peers. Considering the mean scores of males andfemales, regardless of high school major, females had slightly higher PMTEand PSTE scores than the males. Furthermore, comparison of high schoolmajors reveals that mathematics/science major students had higher meanscores than other major students in all possible comparisons.

Two-way ANOVAs were conducted to investigate the significance ofthe gender and high school major effects on PMTE and PSTE scores. A 2x2ANOVA design, involving male and female students and mathematics/science and other majors, was used to explore the main effects and

TABLE 3

Participants’ mean PMTE scores by gender and high school major

Mathematics/Science Major Other Major Total

M SD n M SD n M SD n

Female 59.26 5.18 31 55.58 6.59 31 57.42 6.17 62Male 57.77 5.64 26 56.74 7.18 39 57.15 6.58 65Total 58.58 5.40 57 56.23 6.90 70 57.28 6.36 127

TABLE 4

Participants’ mean PSTE scores by gender and high school major

Mathematics/Science Major Other Major Total

M SD n M SD n M SD n

Female 56.39 5.86 31 50.16 7.73 31 53.27 7.49 62Male 53.54 7.47 26 50.56 8.14 39 51.75 7.96 65Total 55.09 6.73 57 50.39 7.91 70 52.50 7.74 127

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 659

interaction effect. The nonsignificant gender main effect indicates that themales and females had similar PMTE (F=0.02, p=.89) and PSTE (F=0.85,p=.36) scores. However, the significant high school major main effectsuggests that mathematics/science high school majors had significantlyhigher PMTE (F=4.35, p=.04) and PSTE (F=12.02, p=.001) scoresthan other majors. No significant interaction effect of gender and highschool major was found in either model. To explain the practicalsignificance of the mean PMTE and PSTE differences between themathematics/science and other major students, Cohen’s d effect sizeswere calculated. A medium effect size (d=0.38) was calculated for themean PMTE difference compared to a large effect size (d=0.64) for themean PSTE difference.

DISCUSSIONS AND CONCLUSIONS

The findings indicate that a majority of Turkish female and malepreservice elementary teachers in this study graduated from their teachereducation program with adequate self-efficacy beliefs to teach elementarymathematics and science. There has been an ongoing teacher-educationreform process in Turkey, and these results suggest that recent reformshave been successful in terms of preparing more self-efficaciouselementary teacher candidates, especially in teaching mathematics.

The significant correlation (r=.54) between the PMTE and PSTEscores is in agreement with the data reported in previous studies (Utley etal., 2005) and indicates that (a) participants’ self-efficacy beliefs inteaching mathematics and science are associated and (b) a strongerbackground in one area may influence self-efficacy beliefs in both areas.However, when the self-efficacy beliefs in these two areas are compared,participants were found to exhibit significantly lower self-efficacy inteaching science than in teaching mathematics. This result signals thatprospective Turkish elementary teachers do not believe they are equallywell prepared to teach all the contents as they are expected to.

Females were usually reported to have lower mathematics and scienceteaching self-efficacy than males in previous studies conducted indifferent countries (Brownlow et al., 2000; Ho et al., 2000; Riggs,1991; Zettle & Raines, 2000). However, consistent with the findings ofstudies conducted with Turkish middle school teacher candidates (Gencer& Cakiroglu, 2007; Isiksal & Cakiroglu, 2005), it is heartening to observethat the Turkish female preservice elementary teachers in this study werefound to have slightly higher PMTE and PSTE scores than their male

MURAT BURSAL660

peers. A possible explanation would be the more positive attitudes offemales toward elementary teaching, but this finding also suggests thatthere might be other factors significantly affecting teacher candidates’beliefs. Goodchild and Grevholm concluded there are possibly otherfactors that may rather have a greater effect than gender and that “thecrude distinction on the basis of sex may merely hint at a cause whichmay have gender-related characteristics or bias.” (2009, p. 171).

Since all participants in this study had taken the same number ofcollege-level courses, there was no variability among the number ofcollege mathematics and science credits. This result is consistent withprevious studies that had reported nonsignificant correlations between thenumber of college mathematics/science courses and preservice teachers’self-efficacy beliefs to teach these courses (Feistritzer & Boyer, 1983;Stevens & Wenner, 1996; Wenner, 1993; Westerback, 1982). On theother hand, consistent with the conclusion of Sarikaya et al. (2005), thesignificant differences in PMTE and PSTE scores appear to be related toparticipants’ high level of participation from their school mathematics andscience backgrounds, which are frequently undervalued in elementaryteacher education. Although they had taken the same college courses,students with mathematics/science high school majors were found to havesignificantly higher PMTE and PSTE scores than students with other highschool majors. Therefore, another important finding of this study is thatTurkish preservice elementary teachers’ mathematics and science teach-ing self-efficacies are related to their high school majors and, therefore, totheir high school mathematics and science preparation.

When the magnitude of the mean PMTE and PSTE differencesbetween the mathematics/science majors and other majors is compared,a larger effect size (d=0.64) and a smaller p value for the PSTE meandifference (p=.001), compared to the corresponding values for the PMTEmean difference (d=0.38, p=.04), indicate that the mean PSTE differencebetween the two groups is larger than the mean PMTE difference. Asnoted earlier, while almost the entire sample had three years of highschool mathematics, only mathematics/science majors had taken sciencecourses after Grade 9. For this reason, finding the PSTE mean differencebetween different majors to be larger than the PMTE mean difference isan expected and natural result of the Turkish high school system.

The observed significant high school major effect is also consistentwith the study of Stevens and Wenner (1996) that found a significantcorrelation between the number of high school courses and teaching self-efficacy beliefs. Therefore, the significant difference between the PMTEand PSTE scores appears to be related to the lack of high school science

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 661

preparation of non-mathematics/science majors, who constitute about halfof the Turkish preservice elementary teacher population (55% of thesample in this study).

In light of the findings of this study, it is clear that prior learning andmathematics/science experience needs to be considered when planningthe mathematics and science content components of elementary teachereducation programs. While all mathematics/science majors in Turkey takefull science and mathematics courses during high school, their non-mathematics/science major peers have very limited science and mathe-matics background due to the limited number of courses. It should beexpected that these differences during the high school education ofTurkish teacher candidates are most likely to have serious consequencesin college and possibly in their future teaching career.

IMPLICATIONS

The data from different Turkish high school majors indicate that a one-type approach to elementary teacher preparation programs do not help allTurkish elementary teacher candidates. While preservice teachers withmathematics/science high school majors are likely to develop higher self-efficacy beliefs about teaching mathematics and science, due to theirstrong content area background, a considerable number of preserviceteachers with other high school majors lack similar high self-efficacybeliefs, especially in teaching science. This systemic problem might beresolved by changes to teacher education and high school programs.

Unlike many other college programs, Turkish elementary educationprograms are open to all three most common high school majors:mathematics/science, social science, and Turkish/mathematics integrated.Therefore, the preservice elementary teacher population consists ofstudents with distinctive high school content experiences; it is worrisomethat no special consideration is provided to these teacher candidates. Itappears that prior learning needs to be considered and additionally,alternative mathematics and science pedagogical content courses shouldbe developed and added to the elementary teacher education programs.

It is more disturbing that most Turkish high school students are notaware of the consequences of high school majors when they make thiscrucial decision at the end of Grade 9. Therefore, Turkish educators andpolicy makers should consider a more flexible high school educationalsystem where students can take high school courses from variousdisciplines that would lead to a general/liberal arts major. Instead of

MURAT BURSAL662

rigid high school curricula, more elective courses open to all students(especially science-related courses) would help reduce their anxieties inthese areas. Instead of requiring students to make their major decision atan early age, allowing them to switch between majors at any time andawarding majors based on total credits at the end of high school, wouldtransform the rigid Turkish educational system to be a much moreflexible, effective one.

Another recommendation relates to current teacher candidates andteacher education programs. Current rigid teacher-education programscould be transformed into specialized certification or limited teachingassignments. If non-mathematics/science majors do not feel confidentenough to teach science and mathematics, a possible solution would benot requiring them to teach these topics in elementary schools. Instead,these students could be encouraged to teach topics in which they haveprevious high school experience, such as social science, Turkish, etc. Asimilar recommendation would apply to mathematics/science majors whodo not have high self-efficacy in teaching topics other than mathematicsand science.

REFERENCES

Ashton, P. T., & Webb, R. B. (1986). Making a difference: Teachers’ sense of efficacyand student achievement. White Plains: Longman.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.Psychological Review, 84, 191–215.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.Battista, M. T. (1986). The relationship of mathematics anxiety and mathematicalknowledge to the learning of mathematical pedagogy by pre-service elementaryteachers. School Science and Mathematics, 86, 10–19.

Bauer, S. M., & Toms, K. (1990, February). Science in the elementary schools: Teachers’perceptions of test-mandated curriculum reform. Paper presented at the annual meetingof the Eastern Education Research Association, Boston, MA.

Bleicher, R. E., & Lindgren, J. (2009). Variable relationships among different sciencelearners in elementary science-methods courses. International Journal of Science andMathematics Education, 7, 293–313.

Briscoe, C., & Stout, D. (2001). Prospective elementary teachers’ use of mathematicalreasoning in solving a lever mechanics problem. School Science and Mathematics, 101,228–235.

Brownlow, S., Jacobi, T., & Rogers, M. (2000). Science anxiety as a function of genderand experience. Sex Roles: A Journal of Research, 42, 119–31.

Cakiroglu, J., Cakiroglu, E., & Boone, W. (2005). Pre-service teacher self-efficacy beliefsregarding science teaching: A comparison of pre-service teachers in Turkey and theUSA. Science Educator, 14, 31–40.

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 663

Cantrell, P., Young, S., & Moore, A. (2003). Factors affecting science teaching efficacy ofpreservice elementary teachers. Journal of Science Teacher Education, 14, 177–92.

Chin, C. (2005). First-year pre-service teachers in Taiwan: Do they enter the teacherprogram with satisfactory scientific literacy and attitudes toward science? InternationalJournal of Science Education, 27, 1549–1570.

Cox, C., & Carpenter, J. (1989). Improving attitudes toward teaching science and reducingscience anxiety through increasing confidence in science ability in inservice elementaryschool teachers. Journal of Elementary Science Education, 1, 14–34.

Downing, J. E., Filer, J. D., & Chamberlain, R. A. (1997, November). Science processskills and attitudes of preservice elementary teachers. Paper presented at the annualmeeting of the Mid-South Educational Research Association, Memphis, TN. (ERICDocument Reproduction Service No. ED416191)

Enochs, L. G., & Riggs, I. M. (1990). Further development of an elementary scienceteaching efficacy instrument: A preservice elementary scale. School Science andMathematics, 90, 694–706.

Enochs, L. G., Smith, P. L., & Huinker, D. (2000). Establishing factorial validity of themathematics teaching efficacy beliefs instrument. School Science and Mathematics,100, 194–202.

Feistritzer, E. C., & Boyer, E. L. (1983). The conditions of teaching: A state-by-stateanalysis. Princeton: The Carnegie Foundation for the Advancement of Teaching.

Gencer, A. S., & Cakiroglu, J. (2007). Turkish preservice science teachers’ efficacybeliefs regarding science teaching and their beliefs about classroom management.Teaching and Teacher Education, 23, 664–75.

Goodchild, S., & Grevholm, B. (2009). An exploratory study of mathematics test results:What is the gender effect? International Journal of Science and MathematicsEducation, 7, 161–82.

Graeber, A., Tirosh, D., & Glover, R. (1989). Pre-service teachers’ misconceptions insolving verbal problems in multiplication and division. Journal for Research inMathematics Education, 20, 95–102.

Hadfield, O. D., Littleton, C. E., Steiner, R. L., & Woods, E. S. (1998). Predictors ofpreservice elementary teacher effectiveness in the micro-teaching of mathematicslessons. Journal of Instructional Psychology, 25, 34–47.

Harper, N. W., & Daane, C. J. (1998). Causes and reductions of math anxiety in pre-service elementary teachers. Action in Teacher Education, 19, 29–38.

Ho, H., Senturk, D., Lam, A. G., Zimmer, J. M., Hong, S., Okamoto, Y., et al. (2000). Theaffective and cognitive dimensions of math anxiety: A cross-national study. Journal forResearch in Mathematics Education, 31, 362–79.

Isiksal, M. (2005). Pre-service teachers’ performance in their university coursework andmathematical self-efficacy beliefs: What is the role of gender and year in program? TheMathematics Educator, 15, 8–16.

Isiksal, M., & Cakiroglu, E. (2005). Teacher efficacy and academic performance.Academic Exchange Quarterly, 9, 28–33.

Manning, P. C., Esler, W. K., & Baird, J. R. (1982). How much elementary science isreally being taught? Science and Children, 19, 40–41.

Marshall, J. C., Horton, R., Igo, B. L., & Switzer, D. M. (2009). K-12 science andmathematics teachers’ beliefs about and use of inquiry in the classroom. InternationalJournal of Science and Mathematics Education, 7, 575–96.

MURAT BURSAL664

Milli Eğitim Bakanlığı [Ministry of National Education of Turkey]. (2007). High schoollesson schedules. Retrieved November 13, 2007, from http://ogm.meb.gov.tr/ [inTurkish]

Moore, J. J., & Watson, S. B. (1999). Contributors to the decision of elementary educationmajors to choose science as an academic concentration. Journal of Elementary ScienceEducation, 11, 37–46.

Mulholland, J., Dorman, J. P., & Odgers, B. M. (2004). Assessment of science teachingefficacy of preservice teachers in an Australian university. Journal of Science TeacherEducation, 15, 313–31.

Palmer, D. H. (2006). Durability of changes in self-efficacy of preservice primaryteachers. International Journal of Science Education, 28, 655–71.

Putney, L. D., & Cass, M. (1998). Preservice teacher attitudes toward mathematicsimprovement through a manipulative approach. College Student Journal, 32, 626–32.

Quinn, R. J. (1997). Effects of mathematics methods courses on the attitudes and contentknowledge of pre-service teachers. Journal of Educational Research, 91, 108–113.

Ramey-Gassert, L., & Shroyer, M. G. (1992). Enhancing science teaching self-efficacy in preservice elementary teachers. Journal of Elementary ScienceEducation, 4, 26–34.

Riggs, I. M. (1991). Gender differences in elementary science teacher self-efficacy. Paperpresented at the annual meeting of the American Educational Research Association,Chicago, IL.

Rule, A. C., & Harrell, M. H. (2006). Symbolic drawings reveal changes in preserviceteacher mathematics attitudes after a mathematics methods course. School Science andMathematics, 106, 241–58.

Sarikaya, H., Cakiroglu, J., & Tekkaya, C. (2005). Self-efficacy, attitude and scienceknowledge. Academic Exchange Quarterly, 9, 38–43.

Sherwood, R. D., & Westerback, M. E. (1983). A factor analytic study of the state-traitanxiety inventory utilized with pre-service elementary teachers. Journal of Research inScience Teaching, 20, 225–229.

Stepans, J., & McCormack, A. (1985). A study of scientific conceptions and attitudestoward science of prospective elementary teachers: A research report. (ERICDocument Reproduction Service No. ED266024)

Stevens, C., & Wenner, G. (1996). Elementary pre-service teachers’ knowledge andbeliefs regarding science and mathematics. School Science and Mathematics, 96, 2–9.

Swars, S. L., Daane, C. J., & Giesen, J. (2006). Mathematics anxiety and mathematicsteacher efficacy: What is the relationship in elementary preservice teachers? SchoolScience and Mathematics, 106, 306–15.

Tekkaya, C., Cakiroglu, J., & Ozkan, O. (2004). Turkish pre-service science teachers’understanding of science and their confidence in teaching it. Journal of Education forTeaching, 30, 57–66.

Thipkong, S., & Davis, E. (1991). Pre-service elementary teachers’ misconceptions ininterpreting and applying decimals. School Science and Mathematics, 91, 93–99.

Tooke, D. J., & Lindstrom, L. C. (1998). Effectiveness of a mathematics methods coursein reducing math anxiety of pre-service elementary teachers. School Science andMathematics, 98, 136–39.

Tosun, T. (2000). The beliefs of pre-service elementary teachers toward science andscience teaching. School Science and Mathematics, 100, 374–379.

MATH AND SCIENCE TEACHING SELF-EFFICACY BELIEFS 665

Utley, J., Moseley, C., & Bryant, R. (2005). Relationship between science andmathematics teaching efficacy of preservice elementary teachers. School Science andMathematics, 105, 82–88.

Wenner, G. J. (1993). Relationship between science knowledge levels and beliefs towardscience instruction held by pre-service elementary teachers. Journal of ScienceEducation and Technology, 2, 461–468.

Westerback, M. E. (1982). Studies on attitude toward teaching science and anxiety aboutteaching science in pre-service elementary teachers. Journal of Research in ScienceTeaching, 19, 603–616.

Young, B. J., & Kellogg, T. (1993). Science attitudes and preparation of pre-serviceelementary teachers. Science Education, 77, 279–291.

Yüksek Öğretim Kurulu [Higher Education Council of Turkey]. (1998). Teacherpreparation. Retrieved May 5, 2007, from http://www.yok.gov.tr/content/view/16/52/[in Turkish]

Zettle, R. D., & Raines, S. J. (2000). The relationship of trait and test anxiety withmathematics anxiety. College Student Journal, 34, 246–258.

Faculty of Education, Elementary Education DepartmentCumhuriyet UniversitySivas, 58140, TurkeyE-mail: mbursal@cumhuriyet.edu.tr

MURAT BURSAL666