Quantitative approaches to researching ethnicity and difference: how can we trouble and interrogate the categories we are stuck with?

This is a recording of a presentation given at Methods in Dialogue: Researching Diversity, 9 February 2011 at the University of Manchester. Approaches to analysing surveys inevitably rely on the generation of categories to make sense of the data generated. Such categories imply fixed boundaries between homogenous and intuitive groupings of ethnicity (in this case), class, gender etc; but everyday experience point to meaningful diversity within such categories. One way of troubling ethnic categories is to query their explanatory utility within statistical models; how well does the ‘ethnicity’ variable explain variance in outcomes when put up against other variables? In a related manner, ethnic categories can be troubled by exposing heterogeneity within them, so revealing diversity within a category, and similarities across categories, allowing the ‘placement’ of boundaries between ethnic categories to be queried. And ethnic categories can be interrogated by refining them, examining the emergence of new categories, and comparing their significance across contexts and locations. Here comparisons across nations, the study of differing experiences across generations and, particularly, studies of ‘mixed’ ethnicities are useful. Nevertheless, I hope to demonstrate that if we are prepared to interrogate quantitative categories, they are a useful means to mark out important differences across socially meaningful groups.