Preservice teacher efficacy: cross-national study.

Abstract

This cross-national study assessed the relationship between two groups (i.e., US and Turkey) of matched preservice teachers regarding their perceptions of self-efficacy toward teaching students with learning disabilities and their perceptions regarding personal efficacy and general teaching efficacy, as measured by the Teacher Efficacy Scale. All possible subsets regression analyses indicated a model for each preservice teacher group pertaining to their perceptions of self-efficacy toward teaching students with learning disabilities and their perceptions regarding personal teaching efficacy.

Research in the area of cross-national comparisons of preservice and inservice teachers' self-efficacy beliefs validates their importance as a mediating factor influencing instructional decisions (Clark & Artiles, 2000) and accommodating students with learning and behavioral problems in the context of classroom practice (Haynes, Hook, & Macaruso, 2000). Haynes et al. (2000) compared the perceptions of elementary teachers in the United States (US) and Japan regarding the strengths and weaknesses of students according to criteria used for identifying students with learning disabilities (LD) in the US. Results indicated a similar pattern across the two groups regarding teachers' perceptions of student ability in the areas of reasoning and mathematics. Significant between-group differences occurred in the areas of listening, speaking, reading/writing, and social and study skills. In another cross-national study conducted by Clark and Artiles (2000), significant differences were found in attributional responses between US (n = 97) and Guatemalan (n = 59) teachers regarding their perceptions of (a) student ability (high vs. low), (b) students' degree of expended effort (high vs. low), and (c) students' categorical label (LD vs. non-LD) when presented with vignettes depicting a student as having failed a typical classroom test.

Other cross-national research assessing preservice teachers' beliefs about mathematics indicates that the context for teacher preparation (i.e., teacher education programs) can both facilitate and constrain preservice teachers' self-efficacy (Wagner, Lee, & Azgun-Koca, 1999). In a recent study undertaken by Wagner et al. (1999), preservice teachers (n = 106) from the US, Turkey, and Korea completed a questionnaire designed to measure their beliefs about teaching mathematics and about the efficacy of their teacher preparation program. The US preservice teachers' responses indicated a stronger degree of self-confidence regarding their ability to teach mathematics than did the Turkish and Korean preservice teachers. In addition, the responses of the Korean preservice teachers indicated that they were the least confident among the three groups regarding their ability to teach mathematics. Analysis of preservice teachers' responses regarding the efficacy of their teacher training programs indicated that the US preservice teachers believed their teacher education program provided adequate preparation for the teaching of mathematics. In contrast, the Turkish and Korean preservice teachers expressed reservations regarding the degree that their programs provided sufficient groundwork to teach mathematics in classrooms.

Findings from the above cross-national studies suggest that the instructional-based variables of student characteristics (i.e., LD) and content area (i.e., mathematics) differentially impact teachers' beliefs across countries and cultures. As such, these inquiries are useful in identifying cross-national patterns and may limit stereotype biases associated with categorically labeling student performance. Indeed, utilizing samples representing two countries that do not share a similar referral and placement process for labeling students' performance (i.e., LD vs. Non-LD) minimizes variance in teachers' responses that are associated with label stereotypes (MacMillan, Jones, & Aloia, 1974; Semmel & Gao, 1992). However, the bulk of these studies typically have compared teacher beliefs across countries (i.e., cross-national studies) with respect to learning disabilities (e.g., Clark & Artiles, 2000; Haynes et al., 2000) or successful mathematics instruction (e.g., Wagner et al., 1999), or have compared teacher perceptions within a nation (i.e., within-national studies) in the context of standards-based instruction in mathematics (e.g., Collins & Gerber, 2001; Collins & Onwuegbuzie, 2002). To date, no formal research appears to exist involving cross-national comparison of the degree to which LD characteristics impact preservice teachers' self-efficacy beliefs regarding instruction. One exception is a study conducted by Collins and Erktin (2002). These researchers conducted a cross-national study comparing preservice teachers from the US and Turkey with respect to the extent that their beliefs and predictions regarding their instructional practice are influenced by characteristics associated with LD. Results indicated that preservice teachers (US and Turkey) expressed less confidence in their efficacy (i.e., personal efficacy) at addressing students demonstrating poor motivation and poor strategy use in the context of learning in contrast to efficient motivation/strategy use.

Additionally, preservice teachers perceived that they would have to expend a higher degree of instructional effort (i.e., outcome expectancy) in order for students with poor affect to reach grade level expectation in mathematics. Similarly, preservice teachers had lower expectations regarding student performance (i.e., outcome expectancy) for students demonstrating poor strategy use and poor motivation in contrast to characteristics typifying efficient motivation/strategy use in the context of learning. Findings also revealed that the Turkish sample and US sample exhibited differential response patterns with respect to perceived effectiveness of grouping strategies in response to student characteristics. Specifically, the Turkish sample recognized LD characteristics as distinct from efficient motivation/strategy characteristics in their perceptions regarding effective practice. Conversely, the US sample was consistent in their perceptions regarding effective practice across the three sets of student characteristics. All effect sizes were large (Cohen, 1988). The goal for conducting this current research was to identify aspects of preservice teacher beliefs regarding instruction of students with LD that are relatively similar and dissimilar utilizing culture (US and Turkey) as the medium. The present investigation's purpose was to assess the relationship between preservice teachers' perceptions regarding their self-efficacy toward teaching students with LD and their perceptions regarding personal efficacy and general teaching efficacy, as measured by the Teacher Efficacy Scale (TES; Gibson & Dembo, 1984).

This investigation's focus on preservice teachers' efficacy beliefs in the self-regulation of behavior is based on social learning theory posited by Bandura (1977b, 1997). In the educational literature, teachers' self-efficacy beliefs have been conceptualized as a two-dimensional construct defined as outcome efficacy of teaching and personal efficacy of teaching (Bandura, 1977a; Gibson & Dembo, 1984). Outcome efficacy of teaching defines teaching as an influential factor in elevating student achievement. Teachers demonstrating high levels of outcome efficacy select challenging activities and are self-motivated to pursue their goals regardless of obstacles. In contrast, teachers with low levels of outcome efficacy perceive difficult activities as a challenge to their teaching competence. When faced with instructional challenges, these teachers reduce their levels of expended effort. Personal efficacy characterizes teachers as individuals with a high degree of confidence in their teaching abilities and placing a high value on their abilities to affect student performance (Gibson & Dembo, 1984). In contrast, teachers with low levels of personal efficacy place a low value on their abilities to affect student performance and experience helplessness when faced with challenges.

Method

Participants

Data for this study were collected as a component of a larger study that was conducted by Collins and Erktin (2002). To provide a cross-national perspective, two groups of matched primary and secondary preservice teachers (n = 52) representing a public university in Turkey and a similar size university located in the southeastern section of the US were surveyed. The groups were matched on their perceptions regarding effective teaching, as measured by a composite score on the TES (Gibson & Dembo, 1984). Both groups of teachers were enrolled in teacher certification programs that provide both theoretical training and field experiences. Both programs offer courses that address individual differences in the areas of child development, classroom management, behavioral interventions, and strategy training. Practicum courses also were offered to provide students access to realistic classroom experiences.

Instruments and Procedures

Participants were administered a packet containing a cover letter that guaranteed confidentiality regarding their responses and explained the purpose and importance of their participation in the study. All preservice teachers completed the Student Teacher Assessment of Mathematics Instruction (STAMI) questionnaire (Collins & Gerber, 2001). The STAMI questionnaire was developed to measure quantitatively preservice teachers' levels of belief toward mathematics instruction of students with LD. The questionnaire consists of three subscales: level of teacher confidence, level of expended effort, and expectation of a student's grade level success in the context of practice. Preservice teachers were presented nine vignettes describing characteristics of hypothetical students. These vignettes were designed to measure preservice teachers' belief systems with respect to different student self-regulatory characteristics (i.e., poor strategy use vs. poor motivation vs. efficient motivation/strategy use). Preservice teachers' beliefs in response to the vignettes were measured using a 7-point Likert-type scale, ranging from "very low" to "very high." For the present investigation, Cronbach coefficient alphas ranged from .70 to .86 for these subscales' scores.

Preservice teachers also completed the TES (Gibson & Dembo, 1984). The TES is a quantitative measure of teachers' perceptions regarding effective teaching. This instrument consists of two scales. The first scale, representing nine items, measures personal teaching efficacy, or belief that one has the skills and abilities to induce student learning. The second scale, comprising seven items, represents a teacher's perception of general teaching efficacy, or belief that any teacher's ability to induce change is significantly limited by variables outside her/his control. Both scales contain 6-point Likert-type items. For the present research, Cronbach coefficient alphas pertaining to the respondents' scores on the personal teaching efficacy subscale were .88 (US sample) and .70 (Turkish sample), and on the general teaching efficacy subscale were .39 (US sample) and .61 (Turkish sample).

Results and Discussion

Because the focus was primarily about the relationship between preservice teachers' perceptions of their efficacy regarding the teaching of students with LD and their perceptions regarding teacher efficacy, analysis of survey responses was conducted using all possible subsets (APS) multiple regression (Thompson, 1995). This form of analysis involved a series of separate regressions that were conducted to identify the best possible combination of independent variables that predicted the dependent variables. The maximum proportion of variance explained (R2) was the criterion used to determine the inclusion of the independent variables in the model. Utilizing R2 as a criterion for developing the model differentiates APS multiple regression from the much criticized stepwise regression (Hocking, 1976). Two separate APS regression analyses were performed for each sample (i.e., Turkish and US). For each sample, one of the APS regression analyses involved personal teaching efficacy as the dependent variable, as measured by a composite measure of the preservice teachers' perceptions regarding their teaching efficacy (Gibson & Dembo, 1984). The other APS regression analysis employed the general teaching efficacy subscale scores of the TES as the dependent variable. For all four APS models, the independent variables were preservice teachers' responses associated with their efficacy toward teaching students with LD (Collins & Gerber, 2001).

The APS regression analyses yielded a model pertaining to the personal teaching efficacy scale (belief that one has the skills and abilities to induce student learning) for each preservice teacher group sample. For the Turkish sample, preservice teachers' perceptions regarding their levels of confidence and levels of expended effort in response to LD characteristic associated with poor strategy use were statistically significantly associated with personal teaching efficacy (R2 = .26, p < .03). Using Cohen's (1988) criteria, the proportion of variance explained reflects a large effect size. These results indicate that the Turkish preservice teachers who reported high levels of confidence as an instructional response to the LD characteristic of poor strategy use (Beta = .29) tended to have high scores on the personal efficacy scale. Furthermore, Turkish preservice teachers who reported high levels of expended effort as an instructional response to the LD characteristic associated with poor strategy use (Beta = .32) tended to have high scores on the personal efficacy scale.

For the US sample, preservice teachers' perceptions regarding their levels of confidence in response to LD characteristic of poor motivation as well as their levels of expectation of student grade level success in response to LD characteristic of poor strategy use were statistically significantly associated with personal teaching efficacy (R2 = .37, p < .01). The effect size for this model, as measured by R2, was large (Cohen, 1988). These results indicate that the US preservice teachers who reported high levels of confidence as an instructional response to the LD characteristic of poor motivation (Beta = .44) tended to have high scores on the personal efficacy scale. Furthermore, US preservice teachers who reported high levels of expectation of student grade level success (Beta = .22) tended to have high scores on the personal efficacy scale. These results are encouraging because the responses of both preservice teacher groups are consistent with the theory of self-efficacy in the context of educational literature (Bandura, 1977a; Gibson & Dembo, 1984). Specifically, efficacious teachers are confident that they can teach difficult students, place a high value on their ability to affect student performance, and when faced with instructional challenges increase their levels of instructional effort.

Current results indicated no relationship between beliefs about instruction for students with LD and the general teaching efficacy subscale of the TES for both samples of preservice teachers. This also is encouraging because it suggests that both preservice teacher groups' responses to LD characteristics were not impacted by their beliefs that a teacher's ability to induce change is limited by variables outside his/her control. However, these results may be attributed to the low score reliabilities for both preservice teacher groups on the general teaching efficacy scale. Indeed, researchers have criticized the psychometric properties pertaining to the TES (Henson, in press; Witcher, Onwuegbuzie, Collins, Witcher, James, & Minor, 2005). Given that both samples of teachers were enrolled in university teacher preparation programs in their respective countries (US and Turkey) at the time of data collection, further research should assess the degree to which the responses of both preservice teacher groups are associated with their perceptions regarding the efficacy of their teacher education programs. Results of cross-national research support the relationship between preservice teachers' beliefs about teaching specific content and their beliefs concerning the efficacy of their teacher preparation program (Wagner et al., 1999). The degree that a causal relationship exists between the variables in this current study and the preservice teacher preparation programs in which both samples were enrolled is outside the scope of the present study because of the type of data collected. Replications of this study utilizing larger cross-national samples, using other academic subjects as the context for instruction, incorporating other measures of preservice teacher efficacy (e.g., perceptions of the efficacy of their teacher education programs), and asking respondents to provide a reason for each of their responses to LD characteristics (i.e., poor strategy use and poor motivation) will elucidate the multivariate relationships among the variables of culture, beliefs, student learning characteristics, and effective teacher preparation programs.

References

Bandura, A. (1977a). Social learning theory. Englewood Cliffs, NJ: Prentice- Hall.

Bandura, A. (1977b). 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: W. H. Freeman.

Clark, M. D., & Artiles, A. J. (2000). A cross-national study of teachers attributional patterns. The Journal of Special Education, 34(2), 77-89.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York: John Wiley.

Collins, K. M. T., & Erktin, E. (2002, April). Preservice teachers' beliefs toward instruction of students with learning disabilities. A cross-national comparison. Paper presented at the annual meeting of the American Educational Research Association, New Orleans, LA.

Collins, K. M. T., & Gerber, M. M. (2001). Teachers' beliefs about mathematics reform: Instructional implications for students with learning disabilities. Research in the Schools, 8(2), 59-70.

Collins, K. M. T., & Onwuegbuzie, A. J. (2002). Assessing standards-based curricula for students with learning disabilities. Academic Exchange Quarterly, 6(2). (No. 2095-lw). Available: http://www.higher-ed.org/AEQ/

Gibson, S., & Dembo, M. H. (1984). Teacher efficacy: A construct validation. Journal of Educational Psychology, 76, 569-582

Haynes, C., Hook, P., & Macaruso, P. (2000). Teachers' skill ratings of children with learning disabilities: A comparison of the United States and Japan. Annals of Dyslexia, 50, 215-238.

Henson, R. K. (in press). Relationships between preservice teachers' self-efficacy, task analysis, and classroom management beliefs. Research in the Schools.

Hocking, R. R. (1976). The analysis and selection of variables in linear regression. Biometrics, 32, 1-50.

MacMillan, D. L., Jones, R. L., & Aloia, G. F. (1974). The mentally retarded label: A theoretical analysis and review of research. American Journal of Mental Deficiency, 79, 241-261.

Semmel, M. I., & Gao, X. (1992). Teacher perceptions of the classroom behaviors of nominated handicapped and nonhandicapped students in China. The Journal of Special Education, 25(4), 415-430.

Thompson, B. (1995). Stepwise regression and stepwise discriminant analysis need not apply here: A guidelines editorial. Educational and Psychological Measurement, 55, 525-534.

Wagner, S. L., Lee, H. J., & Azgun-Koca, S. A. (1999, January). A comparative study of the United States, Turkey, and Korea: Attitudes and beliefs of preservice teachers toward mathematics, teaching mathematics, and their teacher preparation programs. Paper presented at the annual meeting of the Association of Mathematics Teacher Educators, Chicago, IL.

Witcher, L. A., Onwuegbuzie, A. J., Collins, K. M. T., Witcher, A. E., James, T. L., & Minor, L. C. (2005). Preservice teachers' efficacy and their beliefs about education. Manuscript submitted for publication.

Kathleen M. T. Collins, University of Arkansas at Fayetteville

Kathleen M. T. Collins is an assistant professor at the University of Arkansas at Fayetteville.