Modelling gender inequalities: A longitudinal analysis of repeated ordinal data

We propose a paper that will present longitudinal statistical models of gender inequalities. The paper will present the results of an analysis of three ordinal responses on gender inequality taken from the BHPS between 1991 and 2007. Cumulative logit [or proportional odds] models with correlated random effects will be presented.

These correlated random effects models provide the capacity to test formally whether there is significant association between the three ordinal responses.
The paper will address a number of sociological questions including: Is there one underlying dimension [or latent variable] measuring attitudes to gender inequalities? Does this dimension vary over time? What explanatory factors determine variations in patterns of responses?
The analysis will explore the possible effects of a wide range of explanatory variables including age, ethnicity, social class, marital status, household size, political affiliation and gender.