This study proposes a new framework and verifies its propositions on how teacher self-efficacy (TSE) grows. Bandura (1977) proposed four sources of self-efficacy, including performance accomplishment, vicarious experience, verbal persuasion, and physiological states. Most of the studies on the sources of TSE proceeded from the four-sources theory (Morris, Usher, & Chen, 2017; Usher & Pajares, 2008). However, the results of these studies remain empirically and theoretically unsatisfactory. We argue that the four-sources theory needs to be refined since there are overlaps and internal relations among the four sources. This is because Bandura did not differentiate the concepts of actions, outcomes, and the interpretation of outcomes in the selection of words describing the four sources.

The new framework clearly differentiate and definde actions, outcomes, and the interpretation of outcomes, specifically considering the characteristics of teaching activities. The framework is derived from social cognitive theory (Bandura, 1997), attribution theory (Weiner, 1985), and locus of control theory (Rotter, 1966). We define actions as goal-directed behaviors. Regarding TSE, it refers to any behavior with goals of improving teaching capability. Teaching-related actions include teaching, observing others teaching, reading materials on teaching, and listening to lectures on teaching. Outcomes are the consequences of actions (Bandura, 1997, p. 22; 2001, p. 6). We argue that outcomes are objective information that teachers derive from their actions. As a consequence of teaching-related actions, resultant, evaluative, knowledge, physical, and physiological outcomes are produced. Physical and knowledge information play an intermediary role, evaluative and physiological information play a supplementary role, and resultant information plays the final role in the growth of TSE. The interpretation of outcome information is the process that teachers attach meaning to this information.

The formation of TSE is a cognitive processing of information. Teachers combine various outcome information and make a judgment on their teaching capability. We argue that it takes three steps to reach a judgment of self-efficacy. These are goal comparison, attribution of discrepancy, and self-efficacy appraisal. The formation of TSE goes as this: Teachers take teaching-related actions, and “action will produce certain outcomes” (Bandura, 1977, p. 193), teachers process outcome information to form self-efficacy, which motivates them to take actions again (Bandura, 1988, 1997). The process above repeats until TSE is stabilized.

The first implication of the framework is that we should focus on outcome information, especially resultant information. Existing studies show that TSE is associated with students’ orderly behavior (Lee, Dedrick, & Smith, 1991; Newmann, Rutter, & Smith, 1989), student engagement (Depaepe & König, 2018; Guo, Justice, Sawyer, & Tompkins, 2011), student achievement (Guskey, 1987; Klassen & Tze, 2014). Second, we argue that the level of correspondence between TSE and resultant information in terms of teaching tasks is related to the strength of the relationship between the two. Third, many external factors influence educational goal attainment and TSE, including student abilities (Guskey, 1987; Raudenbush, Rowan, & Cheong, 1992), teaching resources (Tschannen-Moran & Hoy, 2007), distributed and instructional leadership (Fackler & Malmberg, 2016; Liu, Bellibaş, & Gümüş, 2021).

We test the propositions using the Teaching and Learning International Survey 2018 (TALIS2018) data, which include most of the European countries (OECD, 2019). First, we choose the resultant information on students’ cognitive activation, teacher-student relation, and disciplinary climate. Second, we choose class autonomy as an external factor to demonstrate its impact on TSE. Third, TALIS provides information on TSE in classroom management, instruction, and student engagement, we test whether the level of correspondence between TSE and resultant information in terms of teaching tasks is related to the strength of the relationship between the two. Forth, we test the generality of relationships mentioned above across countries.

We use the OECD TALIS 2018 dataset to test the hypotheses elaborated above because it includes survey questions on TSE and other relevant variables (OECD, 2019). The TALIS 2018 data include teachers from 48 countries or economies, most of which are in Europe. At least a minimum sample of 200 schools were randomly drawn from each participating country or economy. At least a minimum of 20 teachers were randomly sampled from each participating school. The final dataset used in the study includes 46 countries or economies, with sample sizes ranging from 827 to 6439 teachers. The method invented by Tschannen-Moran and Woolfolk Hoy (2001) was used to measure TSE in the TALIS 2018 (Ainley & Carstens, 2018). Teaching tasks were categorized into three specific areas: classroom management, instruction, and student engagement. The variables on disciplinary climate, teacher-student relation, cognitive activation, classroom autonomy are latent variables, which were measured through expressed opinions on a set of items. These items were chosen based on theoretical underpinnings (Ainley & Carstens, 2018). The calculation of scores of the constructs went through internal consistent analysis, confirmatory factor analysis (CFA), measurement invariance testing, final scale modeling, and scale score computation. A multilevel model was used for each country or economy because teachers were sampled via the two-strata strategy in the TALIS and teachers were clustered in schools (Snijders & Bosker, 2012). Specifically, this study uses the multilevel model to test the relationships between students’ cognitive activation, teacher-student relation, disciplinary climate, class autonomy and TSE. The dependent variable is TSE, including TSE in the three task areas. The independent variables of interest are teacher-student relation, cognitive activation, disciplinary climate, and class autonomy. Gender, teaching experience and its square are included as controlling variables. We only use a random intercept multilevel model without a random slope.

The empirical part of this study has made several contributions to the literature on TSE. First, the results of data analysis show that educational results, such as cognitive activation, supportive teacher-student relation, and disciplinary climate are significantly associated with the growth of TSE. Second, class autonomy is significantly associated with the growth of TSE. Third, this study shows that specific resultant information has more predictive power on TSE in similar task areas. For example, resultant information on students’ cognitive activation has a stronger relationship with TSE in student engagement than TSE in classroom management, disciplinary climate shows a stronger negative association with TSE in classroom management than TSE in student engagement and instruction, and class autonomy has a stronger association with TSE in instruction than in classroom management and student engagement. Forth, this study shows that TSE theory is cross-cultural. The theory of self-efficacy was created in the context of western culture. Much research was done in western countries (Fackler, Malmberg, & Sammons, 2021). It was debated whether or to what extent the theory is applicable to eastern culture (Yada et al., 2019). The findings of this study reveal that the development of TSE shares similar conclusions across different countries regardless of cultural orientations, with a few exceptions. Finally, many studies on TSE were limited by small samples or non-random sampling (Morris, Usher, & Chen, 2017; Usher & Pajares, 2008). This study’s sample was randomly drawn from the population of lower secondary teachers in each of 46 countries or economies. Thus, its findings are more robust than those of small-sample studies.

Ainley, J., & Carstens, R. (2018). Teaching and Learning International Survey (TALIS) 2018 Conceptual Framework (No. 187; pp. 1–108). Paris: OECD Publishing. Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W.H. Freeman and Company. Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1–26. Depaepe, F., & König, J. (2018). General pedagogical knowledge, self-efficacy and instructional practice: Disentangling their relationship in pre-service teacher education. Teaching and Teacher Education, 69, 177–190. Fackler, S., & Malmberg, L. E. (2016). Teachers’ self-efficacy in 14 OECD countries: Teacher, student group, school and leadership effects. Teaching and Teacher Education, 56, 185–195. Fackler, S., Malmberg, L. E., & Sammons, P. (2021). An international perspective on teacher self-efficacy: Personal, structural and environmental factors. Teaching and Teacher Education, 99, 103255. Guskey, T. R. (1987). Context variables that affect measures of teacher efficacy. Journal of Educational Research, 81(1), 41–47. Klassen, R. M., & Tze, V. M. C. (2014). Teachers’ self-efficacy, personality, and teaching effectiveness: A meta-analysis. Educational Research Review, 12, 59–76. Morris, D. B., Usher, E. L., & Chen, J. A. (2017). Reconceptualizing the sources of teaching self-efficacy: A critical review of emerging literature. Educational Psychology Review, 29(4), 795–833. Newmann, F. M., Rutter, R. A., & Smith, M. S. (1989). Organizational factors that affect school sense of efficacy, community, and expectations. Sociology of Education, 62(4), 221. OECD. (2019). TALIS 2018 and TALIS Starting Strong 2018 User Guide. Paris: OECD. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80(1), 1–28. Snijders, T. A. B., & Bosker, R. J. (2012). Multilevel analysis: An introduction to basic and advanced multilevel modeling (2nd ed.). London: SAGE Publications Ltd. Tschannen-Moran, M., & Hoy, A. W. (2001). Teacher efficacy: Capturing an elusive construct. Teaching and Teacher Education, 17(7), 783–805. Tschannen-Moran, M., & Hoy, A. W. (2007). The differential antecedents of self-efficacy beliefs of novice and experienced teachers. Teaching and Teacher Education, 23(6), 944–956. Usher, E. L., & Pajares, F. (2008). Sources of Self-Efficacy in school: Critical review of the literature and future directions. Review of Educational Research, 78(4), 751–796. Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(4), 548–573.