Critically Evaluating the Effectively Maintained Inequality Hypothesis

Abstract

This paper uses data for England to test the effectively maintained inequality (EMI) hypothesis that individuals from ‘high’ and ‘low’ socioeconomic backgrounds have qualitatively different modal educational destinations at a given educational level. In so doing, the paper highlights how a focus on modal educational destinations seriously detracts from the usefulness of the EMI hypothesis as a basis for identifying qualitative educational inequalities. First, tests of the EMI hypothesis are shown to be of questionable reliability because they involve calculating the predicted probabilities of different educational destinations based on ultimately arbitrary operationalisations of ‘high’ and ‘low’ socioeconomic background, with more polarised formulations being more likely to find in favour of EMI. Second, tests of the EMI hypothesis are shown to be of questionable validity in that it is possible to find in favour of EMI even when the degree of qualitative inequality is negligible and to find against EMI even when the degree of qualitative inequality is substantial. These limitations have been recognised by the originator of the EMI hypothesis but dismissed as unimportant. However, this paper argues that these limitations are so serious that analysts seeking to identify qualitative inequalities in education should discard the focus on modal educational destinations advocated by the EMI hypothesis.