Cultural capital as an explanation of variation in participation in higher education

Abstract

Cultural capital is frequently referred to as a construct in the analysis of inequality in higher education. It has been suggested that variations in cultural capital contribute to social class differences in levels of participation, distribution of students between elite and other universities, and the likelihood of dropping out. However, recent analyses of quantitative data suggest that once students’ attainments are included in analysis of levels of participation the effects of social class disappear. One possibility is that cultural capital affects the likelihood of participation in higher education independently of the common measures of social class variation (parental occupation and education). In this analysis we include a measure of students’ cultural capital to investigate whether it exerts an effect on the likelihood of participation that is independent from students’ attainment. We also present and evaluate a practicable method of measuring students’ cultural capital.

Keywords:

• widening participation
• cultural capital

Acknowledgements

The authors are grateful to Jean Mangan and Jan Meyer for their advice in the preparation of this paper, and to Alice Sullivan and two anonymous referees for comments on earlier drafts of the paper.

Notes

1. A logistic regression examines the effect of a set of explanatory variables on a dependent variable that can only take one or two values (such as participating or not participating in higher education).

2. Factor analyses yield a set of results (in terms of ‘eigenvalues’) that show how much of the variance in the set of results is explained by each factor. On this basis the factors are placed in order starting with the factor that explains the highest proportion of the variance. Once the eigenvalue goes below one any additional factors are generally not adding much to the explanation of the overall variance, so an eigenvalue of one is commonly used as a cut‐off point.

3. Each factor can be associated with any of the items, so there can be overlap between the factors with a single item ‘loading’ more heavily on one factor than another. The factor analysis becomes easier to interpret if there is little overlap between the factors (more separated) so that each item is only associated with one factor.

4. Details can be found online (http://www.aimhigher.ac.uk/sites/practitioner/programme_information/about_aimhigher.cfm).