Social Discrimination in Classrooms: The Contribution of a Social Networks Approach to Theory and Methods, and Empirical Evidence

Abstract

Policy makers have promoted ethnically mixed school populations, the underlying assumption being that an ethnic mix improves interethnic contacts and consequently prevents social discrimination. However, scientific research did not decisively support this assumption, partly because of methodological shortcomings. In this paper an investigation is presented of social discrimination within classroom networks. The social networks tradition entered the field in the 1980s, and initially focused on methodological topics, particularly on the measurement of ethnic segregation, and inspired the development of new statistical models. However, the consequences were far beyond methodological. Our investigation specifically adapted central constructs and developed operational network definitions for opportunity, ethnic mix, ethnic boundaries, segregation, and social discrimination. Some other changes came about more implicitly: the focus changed from prejudices to actual behaviour; the dyad became the central unit; and social discrimination by minority members obtained the same status in research as discrimination by majority members. Moreover, our analysis of three datasets of classroom networks, led to some striking results. While most majority members had only few positive relationships with minority members in their classroom, minority members had ample relationships with majority members. However, while majority members (almost) did not discriminate socially, minority members did. The ethnic classroom composition had no effect on social discrimination. In contrast, minority members discriminated more and majority members less in neighbourhoods with a larger proportion of minority members. None of the current mainstream theories seems to be able to explain these findings. Other mechanisms seem to affect social discrimination more strongly than the development of prejudices. Also, the findings illustrate why mixing projects can have positive outcomes for minority members, even when these projects do not reduce discrimination.