Inclusive education has gradually become the mainstream movement around the world since the Salamanca Statement on Principles was published (UNESCO, 1994), and offering equal educational opportunities to all children has been reinforced as a global agenda (United Nations, 2006; United Nations General Assembly, 2015). It goes without saying that teachers play an important role in implementing inclusive education. To date, a number of studies have highlighted factors that are associated with teachers when they develop inclusive classroom (e.g., Avramidis & Norwich, 2002; de Boer, Jan Pijl, & Minnaert, 2011), and teachers’ self-efficacy in implementing inclusive education is the one which attracts researchers’ attention (e.g., Malinen et al., 2013; Meijer & Foster, 1988).

The term “self-efficacy” was introduced by Bandura (1977) in his social cognitive theory and it was defined as one’s belief that she or he can perform effectively in a certain situation. In addition, Bandura (1997) indicated that self-efficacy is composed of four sources. The first source is mastery experience, which is experience of successes or failures in specific situation and the effect of which on self-efficacy depends on the process and the effort in overcoming obstacles. The second source is vicarious experience, that is, modeling the attainments of others. Efficacy beliefs are enhanced or deteriorated depending on group norms and one’s relationship with others. Verbal persuasion is the third source that is often understood as evaluative feedback from others. The forth source of self-efficacy is psychological and affective states. When people judge their capabilities, they sometimes utilize somatic information based on their psychological and affective states. For example, higher stress level and negative emotional proclivities can reduce self-efficacy belief (Bandura, 1997). Previous studies have supported the Bandura’s idea that mastery experience is the most powerful source among four sources of self-efficacy (Tschannen-Moran & Hoy, 2007; Usher & Pajares, 2008).

Several variables such as gender or ethnicity have been found to influence the role of different sources of self-efficacy (Usher & Pajares, 2008). Klassen (2004a) investigated the differential impact of the sources on the mathematics efficacy beliefs by using a sample of Indo Canadian and Anglo Canadian Grade 7 students. In a regression analysis, vicarious experience and verbal persuasion predicted mathematics efficacy significantly in Indo Canadian students but not in Anglo Canadian students. The results may indicate that the self-oriented sources (mastery experiences and somatic and emotional states) was experienced more in an individualist society, on the other hand, the other-oriented sources (vicarious experience and verbal persuasion) work stronger in a collectivist society (Klassen, 2004a).

Even though teachers’ self-efficacy for inclusive practices has been investigated in several studies, there have been few that focused on their sources. Thus, this paper attempts to examine which sources are related to teachers’ self-efficacy in implementing inclusive education. There are three aims for this study:

1)      To test a construct validity of the Sources of Teacher Self-Efficacy (STSE) scale that was used in this study.

2)      To assess what profiles of sources of self-efficacy come out among the teachers in this study.

3)      To address what is the relationship between four sources of self-efficacy and teachers’ self-efficacy for inclusive practices.

From the second and third aims, we formulated the following hypotheses.

Hypothesis 1: According toprevious studies (Bandura, 1997; Tschannen-Moran & Hoy, 2007; Usher & Pajares, 2008), mastery experiences might be the most influential source among the four sources.

Hypothesis 2: The self-oriented sources (mastery experiences and somatic and emotional states) might work strongly in Finland as it is an individualist society (Klassen, 2004a).

Participants A total of 730 primary and comprehensive school teachers from Finland answered the questionnaire about their self-efficacy for inclusive practices and the sources of self-efficacy. Over half of participants are female (63.0% female, 20.1% male) (Mage=45.77, SD = 9.02). Measures Teachers' self-efficacy for inclusive practices was measured using the Teacher Efficacy for Inclusive Practices (TEIP) scale (Sharma, Loreman, & Forlin, 2012). The scale consists of 18 items which can be divided into three sub-scales: 1) "Efficacy in instruction" examines participants' efficacy beliefs in using suitable strategies when they implement inclusive education; 2) "Efficacy in collaboration" addresses participants' efficacy beliefs in collaborating with students' guardians and other school staffs; and 3) "Efficacy in managing behaviour" reveals participants' perception of efficacy in coping with students who show problematic behaviour. Sources of self-efficacy were assessed using the STSE scale which is a newly developed scale for the ProKoulu project in Finland (Malinen, 2014). The scale consists of 16 items asking how much the four sources have affected participants' perception of capabilities in four different educational domains (instruction, behaviour management, collaboration and student engagement). The items can be divided into four subscales that are: 1) mastery experience; 2) vicarious experience; 3) verbal persuasion; and 4) affective state. Analysis All the analyses were conducted using Mplus version 7.0 statistical program for Mac. The model fit was evaluated using Comparative Fit Index (CFI), Root Mean Squared Residual (SRMR) and Root Mean Square Error of Approximation (RMSEA). Cutoff value close .95 for CFI, .09 for SRMR and .06 for RMSEA were signs of good fit according to the previous study (Hu & Bentler, 1999). At the first stage, Confirmatory Factor Analysis (CFA) was conducted to test the measurement model of the two scales. Since the STSE scale has a unique structure which addresses the four sources in the four educational domains, the Multi-Trait Multi-Method (MTMM) model was adapted for the STSE scale. In the second stage, regression analysis was conducted to examine whether a certain source predicts teachers' self-efficacy. Since the primary trait factors in the STSE scale have high correlations with each other, a Cholesky decomposition approach (de Jong, 1999) was adapted in this study. The method allows researchers to deal with a problem of multicollinearity and to find a unique contribution of the highly correlated factors.

Testing measurement model The theoretically driven CFA model for the TEIP scale and MTMM model for the STSE scale were conducted. All factor loadings with the primary factors of the TEIP scale and the trait factors of the STSE scale were statistically significant, showing adequate fit (CFI = .957, RMSEA = .037 and SRMR = .044). Thus, the results indicated there was a construct validity for each of the two scales. Testing Cholesky model Cholesky regression models were conducted to answer the second and third research questions. The results showed that the four sources of self-efficacy explained 56% of the variance in teacher self-efficacy. Regarding the unique contribution of each source, mastery experience was the strongest independent predictor from the other sources (β = .72, p < .001). The result is in accord with Bandura (1997) indicating that mastery experience is the most powerful source of self-efficacy which means Hypothesis 1 was supported. Vicarious experience and affective state did not have any such variation that was independent from others and that could explain teacher self-efficacy. However, verbal persuasion was an independent predictor (β = .10, p < .05) although it explains only 1% of variance. Since there is no significant relationship between affective state and teachers' self-efficacy, Hypothesis 2 was not supported in this study. Consequently, it is evident from the data that the subscale of mastery experience was the most influential source and that verbal persuasion had small effect on teachers' self-efficacy. It is pointed out that how the four sources affect self-efficacy might be different from different culture (Klassen, 2004b). Therefore, a comparative analysis using Japanese and Finnish data will be conducted in a further study.

Avramidis, E., & Norwich, B. (2002). Teachers' attitudes towards integration/inclusion: a review of the literature. European Journal of Special Needs Education, 17(2), 129-147. 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: Freeman. de Boer, A., Jan Pijl, S., & Minnaert, A. (2011). Regular primary schoolteachers' attitudes towards inclusive education: a review of the literature. International Journal of Inclusive Education, 15(3), 331-353. de Jong, P. F. (1999). Hierarchical regression analysis in structural equation modeling. Structural Equation Modeling, 6(2), 198-211. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1-55. Klassen, R. M. (2004a). A cross-cultural investigation of the efficacy beliefs of South Asian immigrant and Anglo Canadian nonimmigrant early adolescents. Journal of Educational Psychology, 96(4), 731-742. Klassen, R. M. (2004b). Optimism and realism: A review of self-efficacy from a cross-cultural perspective. International Journal of Psychology, 39(3), 205-230. Malinen, O.-P. (2014). Improving teachers' sense of professional efficacy: Empirical investigation on the sources of teacher efficacy. Paper presented at the 2nd Global Teacher Education Summit, Beijing Normal University, China. Malinen, O.-P., Savolainen, H., Engelbrecht, P., Xu, J., Nel, M., Nel, N., & Tlale, D. (2013). Exploring teacher self-efficacy for inclusive practices in three diverse countries. Teaching and Teacher Education, 33, 34-44. Meijer, C. J. W., & Foster, S. F. (1988). The effect of teacher self-efficacy on referral chance. The Journal of Special Education, 22(3), 378-385. Sharma, U., Loreman, T., & Forlin, C. (2012). Measuring teacher efficacy to implement inclusive practices. Journal of Research in Special Educational Needs, 12(1), 12-21. 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. UNESCO. (1994). The Salamanca statement and framework for action on special needs education. Retrieved from http://www.unesco.org/education/pdf/SALAMA_E.PDF United Nations. (2006). Convention on the rights of persons with disabilities. Retrieved from http://www.un.org/disabilities/documents/convention/convoptprot-e.pdf United Nations General Assembly. (2015). Transforming our world: The 2030 agenda for sustainable development. Retrieved from http://www.un.org/ga/search/view_doc.asp?symbol=A/70/L.1 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.