The relevance of sources of support for teachers’ self-efficacy beliefs towards students with special educational needs 1. Introduction The move to include students with special educational needs (SEN) in mainstream education is one of the priorities of educational reform agendas in many countries. In Flanders (Belgium) the aim is to implement a more inclusive school system, but it faces resistance from practitioners. After all, implementing a more inclusive school system requires teachers to adjust their practices and adopt new tasks that come with the job of instructing SEN students (Pijl & Frissen, 2009). These practical concerns have undermined the implementation of inclusive education (IE) (Burke & Sutherland, 2004), and have led to a growing research interest in measuring teachers’ self-efficacy (TSE) in teaching SEN students in mainstream education (Sharma, Loreman, & Forlin, 2012). Teachers self-efficacy regarding inclusive education TSE refers to teachers’ beliefs about their capability to “organize and execute the courses of action required to produce given attainments” (Bandura, 1997, p. 3). Research has indicated the importance of high TSE for teaching SEN students in mainstream classes (e.g., Brownell & Pajares, 1999). Highly self-efficacious teachers make greater teaching effort, which leads to better student performance and provides a successful experience for teachers, thus further improving their levels of self-efficacy (Tschannen-Moran & Woolfolk Hoy, 2001). Sources of support for teaching students with special educational needs The availability of sources of support at class and school level has repeatedly been associated with more positive attitudes (e.g., Avramidis & Norwich, 2002). Avramidis and Norwich (2002) divide sources of support for SEN students into three categories: (1) physical support (e.g., teaching materials, IT equipment, adjusted physical environment), (2) human support (e.g., learning support assistants), and (3) school environment (e.g., class size, planning time). Chiner and Cardona (2012) confirm that teachers who have more human support and material resources, comparable to physical support, are more positive towards IE than those who have less support and fewer resources. Avramidis and Kalyva (2007) point out that “teachers may not hold ‘negative attitudes’; rather they may not see solutions to problems they feel are outside their competence or control” (p. 385). Therefore, sources of support seem to be crucial for TSE in relation to SEN students. The relation between teacher self-efficacy and sources of support A review of teacher perceptions regarding the inclusion of SEN students by Scruggs and Mastropieri (1996), concluded that many teachers have concerns about IE and believe that supports are necessary to succeed. In other studies (e.g., Kuyini, Desai, & Sharma, 2018), these sources of support are described as concerns (e.g., concerns from teachers about a lack of, resources, funding and training to teach SEN students). If these support sources are not available, then they become concerns for teachers when teaching SEN students. Recent studies (e.g., Ahsan, Sharma, & Deppeler, 2012) found a link between self-efficacy and concerns about teaching students with SEN. Using a measure of general efficacy toward inclusive practices, Ahsan et al. (2012) found that pre-service teachers with higher levels of self-efficacy showed lower levels of concerns towards IE. Considering these findings, it is interesting to discover which sources of support affect TSE in relation to SEN students. This study The aim of this study is twofold: (1) to compare teachers’ student-specific self-efficacy levels in relation to students with and without SEN, and examine the contribution of teacher-related (gender, years of teaching experience, grade level, experience in special education, experience as a care teacher or (internal) student counsellor, and training (an additional degree in care)) and student-related variables (none, one or multiple diagnoses, and type of disability) in predicting TSE; and (2) to examine how sources of teacher support that favour including SEN students can influence TSE. 2. Materials and methods Data for the current study was collected from January to February 2018, from 692 regular school teachers; 291 teachers from 59 primary schools and 401 teachers from 61 secondary schools in Flanders (Belgium). To ensure the sample was representative, the following variables were taken into account to select a variety of schools: school network , geographical spread, and school size. To measure TSE in relation to students with and without SEN, teachers were asked to complete the Dutch version of the Student-Specific TSE Scale (Zee & Koomen, 2015). They were asked to complete the scale twice, first for the first student on their class list without SEN (i.e., a student who does not receive additional care) and second for the first student on their class list with SEN (i.e., a student for whom the additional care within the school is not sufficient). A six-point Likert-type scale (1 = ‘totally disagree’ to 6 = ‘totally agree’) was used. To analyse TSE in relation to SEN students and the sources of support used for these students, the teachers were asked to indicate to what degree they made use of certain sources of support for the described SEN student), using a four-point Likert-type scale (1 = ‘(almost) never’ to 4 = ‘very often’). The sources of support surveyed were based on Avramidis and Norwich (2002) (i.e., physical and human support) and on availability within the Flemish education system. An exploratory factor analysis was performed on the sources of support to explore multi-dimensionality. Two factors were retrieved, ‘easily accessible sources’: information and advice, materials, and emotional support; and ‘coöperative sources’: observation and feedback, supervision/coaching, co-teaching/team teaching, support for this student in the classroom, support for this student outside the classroom. All data analyses were conducted using SPSS software. A paired samples T-test was performed to compare student-specific TSE in relation to teaching students with and without SEN. To examine the contribution of teacher-related variables in predicting TSE a series of multilevel models were fitted. To investigate the contribution of student-related variables in predicting teachers’ student-specific self-efficacy beliefs towards SEN students, a set of one-way ANOVAs were carried out. (Multilevel modelling was not needed because each teacher rated only one SEN student.) To identify whether the use of support was predictive for teachers’ student-specific self-efficacy levels in relation to SEN students, multiple regression analyses were performed using Easily accessible sources and Cooperative sources as predictors. 3. Discussion and conclusion Regarding the first research aim, it was found that teachers have significantly lower levels of TSE when teaching SEN students compared to students without SEN. When looking at teacher-related variables that affect these levels of self-efficacy, remarkable results were found with regards to gender and grade level. Compared to male teachers, female teachers generally tend to have higher levels of self-efficacy for teaching students without SEN, but not for students with SEN. At the grade level, we found that compared to secondary school teachers, primary school teachers have higher levels of self-efficacy towards students without SEN, but lower overall levels of self-efficacy towards SEN students, With regard to student-related variables for students with SEN, TSE is lower regarding students with multiple diagnoses compared to students with one diagnosis, and lower regarding students with socio-emotional and/or behavioural disorders compared to students with learning disabilities. For the second research objective, we found that the more Cooperative sources of support were used the higher the levels of student-specific self-efficacy. No predictive value for Easily accessible sources of support on teachers’ student-specific self-efficacy was found. These findings emphasize the importance of human support sources as described by Avramidis and Norwich (2002), but only those sources of support characterised by in-depth cooperation between two teachers or a teacher and an internal or external counsellor. We assume that making these forms of support more available to teachers will make them more competent to deal with students with SEN and will, therefore, reduce resistance to a more IE system. However, Cooperative sources of support may not be sufficient for all students (e.g., students with multiple diagnoses) or all teachers. Further research on boundary conditions of Cooperative sources of support is needed. References Ahsan, M. T., Sharma, U., & Deppeler, J. M. (2012). Exploring Pre-Service Teachers' Perceived Teaching-Efficacy, Attitudes and Concerns about Inclusive Education in Bangladesh. International Journal of whole schooling, 8(2), 1-20. 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. doi:10.1080/08856250210129056. Avramidis, E., & Kalyva, E. (2007). The influence of teaching experience and professional development on Greek teachers’ attitudes towards inclusion. European Journal of Special Needs Education, 22(4), 367-389. Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: Worth Publishers. Brownell, M. T., & Pajares, F. (1999). Teacher efficacy and perceived success in mainstreaming students with learning and behavior problems. Teacher Education and Special Education, 22(3), 154-164. Burke, K., & Sutherland, C. (2004). Attitudes toward inclusion: Knowledge vs. experience. Education, 125(2). Chiner, E., & Cardona, M. C. (2013). Inclusive education in Spain: how do skills, resources, and supports affect regular education teachers’ perceptions of inclusion? International Journal of Inclusive Education, 17(5), 526-541. Kuyini, A. B., Desai, I., & Sharma, U. (2018). Teachers’ self-efficacy beliefs, attitudes and concerns about implementing inclusive education in Ghana. International Journal of Inclusive Education, 1-18. doi:10.1080/13603116.2018.1544298. Pijl, S. J., & Frissen, P. H. A. (2009). What policymakers can do to make education inclusive. Educational Management Administration & leadership, 37(3), 366-377. doi:10.1177/1741143209102789. 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. Schwab, S. (2019). Teachers’ student-specific self-efficacy in relation to teacher and student variables. Educational Psychology, 39(1), 4-18. doi:10.1080/01443410.2018.1516861. Tschannen-Moran, M., & Woolfolk Hoy, A. W. (2001). Teacher efficacy: Capturing an elusive construct. Teaching and teacher education, 17(7), 783-805. Zee, M. & Koomen, H. M. Y. (2015). Student-specific teacher self-efficacy: Investigating the factorial, convergent, and concurrent validity of a new instrument. Poster presented at the biennial meeting of SRCD, Philadelphia, PA.