The purpose of this article is to investigate whether family background and the choice of vocational field explain the observed gap in dropout rates from vocational upper secondary education between natives and children of immigrants in Denmark and to investigate ethnic and gender differences in educational choices.
A parsimonious version of Cameron and Heckman's (2001) dynamic statistical model of educational progression is estimated. By parceling educational attainment out into a series of transitions by grade, the model is able to identify barriers to educational progression and to determine the effects of explanatory variables at different stages of the educational career. In addition, the model is able to accommodate the institutional structure of an educational system with parallel branches of study at the upper secondary level and to control for educational selection and unobserved heterogeneity.
The main findings are: that family background variables do not explain the gap in dropout rates between natives and children of immigrants; that the dropout rates from different vocational fields are significantly different but affect natives and children of immigrants equally; that girls do better than boys in immigrant families; and that intergenerational mobility is greater among children of immigrants than natives.
The statistical model used is not available in any standard statistical package. For the purpose of this paper it was coded in GAUSS. Furthermore, the model demands fairly large data sets to be useful in empirical research.
The analysis provides more detailed information about differences in educational attainment between population groups than most previous studies.
