Class Inequality and Egalitarian Reform

Abstract

Major studies of recruitment to higher education find the association with class to be stable over time, and on this basis conclude that egalitarian reforms have been ineffective. In contrast, measures of inequality for the same data show the distribution of attractive positions to have evened out considerably. The difference in results is explained by how the two sets of measures are affected by changes in the marginals of the class-education table, in particular the increased provision of higher education. It is argued that measures of inequality are more relevant for the question of the effectiveness of reform than measures of association, whether linear or loglinear.

Linear versus logistic regression when the dependent variable is a dichotomy

“…This pattern of monotonous increasing values for PR is different from that of both the linear and the loglinear measures. For this reason it has been argued that neither of these two classes of measures are suited for analyses of inequality (Hellevik 1997(Hellevik , 2000(Hellevik , 2002.…”

Inequality Versus Association in Educational Attainment Research: Comment on Kivinen, Ahola and Hedman

“…Their faith in OR obviously has not been shaken by my assertion of a fallacious inference from the specific property of being 'invariant under row or column multiplication' to the general claim of insensitivity to any changes in the marginals (Hellevik 1997:394;2000:82). Nor by the serious problem pointed out with regard to the description of fundamental differences between linear and loglinear measures of association (Hellevik 1997(Hellevik :388: 2000.…”

“…I have suggested using sampling models to represent the process of selection to attractive positions, with quota sampling as an example (Hellevik 1997(Hellevik :390: 2000. Stratitied random probability sampling may also serve as a model.…”

Inequalities in educational opportunities in France: educational expansion, democratization or shifting barriers?

“…Even if the successive transitions remain marked by social inequalities, it is not without social or political importance that pupils from underprivileged groups are more educated and that education is more equally distributed. Expansion may thus gradually eliminate inequalities in schooling, modifying the framework within which social bias has historically operated in the allocation of pupils to education levels (Hellevik 1997).…”