Segregation indices can be modelled statistically by using bootstrap re‐sampling along with multilevel modelling. Computer software is now available to accomplish this straightforwardly. Examples are provided from measurements of social‐class segregation among Scottish secondary‐school pupils. The modelling shows that some of the differences in segregation between communities can be attributed to characteristics of the communities. Extensions of the model would allow the tracking of changes in segregation over time, for example to assess the impact of policies to reduce segregation.
Notes
The research was supported by the Centre for Educational Sociology (University of Edinburgh), a Designated Research Centre funded by the Economic and Social Research Council (C00280004), and by the Canadian Social Sciences and Humanities Research Council. The analyses and conclusions do not necessarily reflect the views or policies of these or any other organisations. We are grateful to Stephen Raudenbush and three anonymous reviewers for their comments on an earlier draft of this paper.