Search

There is a widely recognised national trend for girls to outperform boys at all levels of compulsory schooling. With few exceptions, however, most recent research has reported that, in relation to academic performance at university, men are proportionately over-represented at the First Class level. A number of general hypotheses have been put forward to explain this phenomenon, including those that assume gender-linked differences in cognitive and/or personality traits. A smaller proportion of research has given explanatory primacy to the broad subject area studied. More specifically, it has been alleged that the over-representation of men within the First bracket is largely a function of a "compositional effect" whereby men achieve proportionately more Firsts as there are more of them within the First-rich Sciences. Based upon analysis of 1,707,408 students graduating between 1995 and 2002, this paper seeks to provide the most comprehensive exploration, to date, of this effect. It confirms that a substantial proportion of the "gender gap" can be explained with reference to the male propensity to take degrees in First-rich disciplines.

Although Boudons distinction between primary and secondary effects, and its associated rational action models of inequality of educational opportunity, have been more influential in the field of social stratification and mobility than in the sociology of education, there is good reason to reconsider the theoretical and practical implications of this approach. The investigation brings conceptual analysis and empirical research to bear on Boudons arguments in a manner that may be somewhat unorthodox. The theoretical arguments are developed in the context of a detailed empirical investigation of three transitions--age 10 to O-level, O-level to A-level, and A-level to degree--using the extensive 1970 British Cohort Study. It is concluded that primary and secondary effects should be recognised as methodological rather than theoretical concepts, that the techniques used to identify them are independent of rational action theory and that, contrary to an influential position, the evidence suggests that primary effects are more important than secondary effects in the generation of social disparities in access to education. (Contains 3 tables.)