To meet the special educational needs (SEN) of students in regular classes, policy actors allocate additional personnel resources toward inclusive schools. Though the distribution of personnel resources within schools evades direct political control, indirect modes of governance are conceivable (Altrichter, 2016). Policy actors can use different modes to allocate personnel resources: input-based on the number of students with SENs or according to a through-put model based on the size of student body (Banks et al., 2015). Subsequently, school principals direct resources to classes. Empirically it is unclear, if the mode of allocation on policy level (input-based vs. through-put) entails differences in the intra-school distribution of resources. To address this desideratum, we tested two hypotheses: (H1) the more children with SEN in a class, the more resources are directed towards this class. (H2) This connection differs according to the mode of allocation on policy level. Additionally, we analyzed if the patterns found were stable over time. Data on the intra-school distribution of resources was collected in N = 76 inclusive primary schools in school years 2012/2013 (t1) and 2013/2014 (t2) in eastern Germany. Personnel resources for those schools were allocated through-put based (to cover SENs in the area of learning, speech, or behaviour; LSB), and input-based (for other SENs). The school principals answered a questionnaire (t1 and t2), listing the distribution of personnel quotas for each class (e.g., How many hours per week is a special teacher paid from lump-sumLSB in class 2b?) as well as the students with SEN in each class. To test our hypotheses, we calculated correlations between personnel quotas and the number of students with SENLSB and SENother on school level by pooling them over classes. Using Fisher-z-transformed coefficients we tested if the correlations’ means differed significantly (p < .05) from zero (H1) and if the correlation coefficients differed in relation to the allocation mode (H2). It turned out, that the correlation between the number of children with SENLSB and through-put based resources was not significant at t1, while at t2 it was significant for one quota (r = .27). The input-based resources were significantly linked to the number of children with SENother in class at t1 and t2 (.50 > r < .82). The differences between the correlations were significant at both measurement points. The results indicate that modes of personnel assignment on policy level influence the distribution of resources to classes on school level.

Altrichter, H. (2016). Governance in education: Stocktaking governance reforms and initiatives over the last two decades. Background paper commissioned by UNESCO for the conference on Education Policies for 2030: Governance, School Leadership and Monitoring and Evaluation as levers for change, UNESCO Headquarters, Paris; UNESCO). Banks, J., Frawley, D., & McCoy, S. (2015). Achieving inclusion? Effective resourcing of students with special educational needs. International Journal of Inclusive Education, 19, pp. 926-943.