Studying historical occupational careers with multilevel growth models

In this article we propose to study occupational careers with historical data by using multilevel growth models. Historical career data are often characterized by a lack of information on the timing of occupational changes and by different numbers of observations of occupations per individual. Growth models can handle these specificities, whereas standard methods, such as event history analyses cannot. We illustrate the use of growth models by studying career success of men and women, using data from the Historical Sample of the Netherlands. The results show that the method is applicable to male careers, but causes trouble when analyzing female careers.