For several decades, research on opportunities in education has shown differences in school careers according to socio-economic and ethnic-cultural backgrounds and has tried to explain these differences (see e.g. Berkowitz, Moore, Astor & Benbenishty, 2017). The quality of teaching is one of the most central but also complex factors in this research (Opdenakker & Van Damme, 2000; Hargreaves & Fullan, 2012; Nye, Konstantopoulos, & Hedges, 2004). A literature review by Severiens, Wolff and van Herpen (2014) illustrates the complexity of teaching competency in urban schools as it comprises five specific areas of expertise. Wubbels, den Brok, Veldman and van Tartwijk (2006) also show how diversity in classrooms put a higher and complex demand on teachers.

The higher demands and specific expertise needed can explain, at least partly, why the differences in careers according to student background are so persistent. One of the possible ways forward is suggested by school effectiveness research. Effective schools in urban areas show characteristics of so-called learning organizations (see e.g., Muijs, Harris, Chapman, Stoll & Russ, 2004). They seem to operate as professional learning communities (PLCs) (Hargraves & Fullan, 2012). When teachers collaborate and try to improve their teaching by focusing on student performance, they may be addressing the higher demands more successfully. In other words, schools that operate as a PLC may increase the quality of teaching and, subsequently enhance student performance in diverse classrooms. We have to be careful with this claim though, as most of the research that shows this link is based on small-scale qualitative research (see Muijs et al, 2004). More quantitative research is needed to find out more about the challenges and demands in diverse classrooms and the possible effects of PLCs.

The reports of the OECD on the Teaching and Learning International Survey (TALIS) data (2014a; 2016) do not directly address the question about the relationships between composition (in terms of SES and language diversity), teacher self-efficacy and the mediating role of PLCs. The reports do show that teaching in diverse classrooms seems to be more demanding, but not always and not in every respect. Teachers in diverse classrooms “encounter more difficulties in creating an orderly learning environment in their classroom.“ (OECD, 2016, p. 116; see also OECD 2014a). And, only in some countries are higher percentages of low achieving pupils or pupils with behavioural problems related to lower scores of teachers on self-efficacy (OECD, 2014a). The OECD (2014a) also reports that in-school relationships seem to reduce the relationship between the percentage of low-achieving students on the one hand and self-efficacy on the other hand. We aim to investigate a similar relationship, with the direct indicators of PLCs and more classic composition variables (SES and home language).

This will done by conducting secondary analyses on the Dutch data in TALIS 2013. The rationale for choosing the Dutch data is the increasing emphasis of Dutch educational policy on professional capacity in schools, among which the encouragement of establishing professional learning communities (OCW, 2011). This is expected to increase educational quality in general. In the current paper, we will investigate the effects of PLCs in schools that vary according to composition to find out whether the claim that PLCs are specifically effective in urban schools can be supported.

The following research questions will be answered: Is there a relationship between school composition and teacher self-efficacy? Is this relationship mediated by the presence of professional learning communities?

The concepts of school composition, self-efficacy as well as PLC’s will be operationalized on the basis of the TALIS conceptual framework and its measures.

Respondents: 1912 teachers and 199 school leaders from 127 schools have participated (Van der Boom & Stuyvenberg, 2013). Response rates are 80% among schools and 76% among teachers. The level is ISCED 2, 55 % is female and the average age is 34 years. Procedure: the data are collected in two steps, first schools were randomly selected and then teachers within schools. The questionnaires were administered in the period May to June in 2013. (see for more information OECD, 2014b). Instruments: School composition was indexed by four categories on the percentage of pupils from low SES backgrounds and by the percentage of pupils with a different home language (1= none, 2 = 0-10%, 3 = 11-30% and 4 > 30%). Self-efficacy was measure by three scales: to what extent are teachers feel proficient in classroom management (4 items, alpha is .852), instruction (4 items, alpha is .646) and establishing student engagement (4 items, alpha is .740). To what extent the school shows characteristics of a professional learning community (PLC) was measured by two scales: exchange and coordination for teaching (TCEXCHS: 4 items, alpha is .636 and professional collaboration (TCCOLLS: 4 items, alpha is .576) (OECD, 2014). Analyses: Two-level regression models were tested to investigate the relation between each Level 1 predictor separately (i.e., exchange and coordination for teaching and professional collaboration) and Level 2 predictors (Language and SES) on the three dependent self-efficacy variables (i.e., classroom management, instruction and student engagement). Depending on whether random slopes for the dependent variables added to the model fit, further testing for cross-level interactions was applied. Two levels were accounted for: the teacher level and the school level. The models were tested in the following order: 1. In the 0-model, averages and the variance components at each of the two levels were calculated 2. Level 1 predictors “two indicators of PLCs” were included 3. Level 2 predictors “Language” (three dummy variables) and SES (three dummy variables) were included, each with “none” as reference group 4. Random slopes model: Random slopes for the dependant variables 5. In case the random slopes model showed an improved fit, cross-level interactions were tested

Self-efficacy (Level 1 predictors): classroom management, instruction and student engagement Both indicators of a learning organization, Exchange and Coordination for Teaching and Professional Collaboration were positively related to all three teacher efficacy scales (classroom management, instruction and student engagement), which reflects that teachers who gave higher ratings on the PLC indicators were more confident about their teaching skills. Composition (Level 2 predictors): Language and SES The percentage of students with a different native language did not affect teacher self-efficacy in any of the scales. However, teachers who worked at schools with a higher percentage students with a low SES background felt less confident in their ability to engage students than teachers who worked at schools with a lower percentage of students with a low SES background. A small effect of SES was found on instruction, reflecting that teachers working in schools with 11-30% students with a low SES background are less confident about their instructional abilities than teachers who work at school without children with a low SES background. In conclusion, secondary analyses of the Dutch TALIS 2013 data show that teaching in diverse classrooms is indeed more demanding: teachers feel less confident about their ability to engage their students and less confident about their instruction. At the same time, schools that operate as a PLC seems to increase teachers’ self-efficacy. In other words, schools that encourage teachers to collaborate and observe each other’s classrooms, and have them perform joint activities, seem to increase their skills in classroom management, instruction and student engagement. Mediation analyses, that are currently being conducted, will show the possible mediating effect of PLCs on self-efficacy in schools that vary according to composition.

Berkowitz, R., Moore, H, Astor, R., & Benbenishty, R. (2017). A Research Synthesis of the Associations between Socioeconomic Background, Inequality, School Climate, and Academic Achievement. Review of Educational Research, 87(2), 425-469. doi.org/10.3102/0034654316669821 Hargreaves, A. & Fullan, M. (2012). Professional capacity. Transforming teaching in every school. New York: Teachers College Press. Muijs, D., Harris, A., Chapman, C., Stoll, L. & Russ, J. (2004). Improving Schools in Socioeconomically Disadvantaged Areas – A Review of Research Evidence. School Effectiveness and School Improvement, 15(2), 2, 149-175. Nye, B., Konstantopoulos, S., & Hedges, L. V. (2004). How Large Are Teacher Effects? Educational Evaluation and Policy Analysis, 26 (3), 237 257. OCW (2011). Actieplan Leraar 2020. [Action plan Teacher 2020]. Den Haag: OCW (Ministry of Education). OECD (2014a). Talis 2013 Results: An International Perspective on Teaching and Learning, TALIS, OECD Publishing. http://dx.doi.org/10.1787/9789264196261-en OECD (2014b). Technical Report TALIS, 2013. OECD Publishing. Download op 31-1-2018. https://www.oecd.org/edu/school/TALIS-technical-report-2013.pdf OECD (2016), School Leadership for Learning: Insights from TALIS 2013, TALIS, OECD Publishing, Paris. http://dx.doi.org/10.1787/9789264258341-en Opdenakker, M. & Van Damme, J. (2000) Effects of Schools, Teaching Staff and Classes on Achievement and Well-Being in Secondary Education: Similarities and Differences Between School Outcomes. School Effectiveness and School Improvement, 11(2), 165-196. DOI: 10.1076/0924-3453(200006)11:2;1-Q;FT165 Severiens, S., Wolff, R. & Herpen, S. van (2014). Teaching for diversity. European Journal of Teacher Education, 37(3), 295-311 doi 10.1080/02619768.2013.845166 Van der Boom, E. & Stuyvenberg, M. (2013). Teaching and Learning International Survey (Talis) 2013. Nationaal rapport Nederland. Den Haag: Ministerie van OCW. Wubbels, T., Brok, P.den, Veldman, I., & Tartwijk, J. van (2006). Teacher interpersonal competence for Dutch secondary multicultural classrooms. Teachers and teaching: Theory and practice, 12, 407-433.