Beyond Transitions: Applying Optimal Matching Analysis to Life Course Research

Life course researchers have increasingly explored optimal matching analysis (OMA) as a tool for the analysis of sequences, such as sections of people's status biographies. OMA is usually employed in combination with cluster analysis ( CA) to create classifications of sequences. In this article, we introduce an analytic strategy that allows assessing the classification's internal validity. Using ideal typical sequence representations, we test different cluster algorithms and are able to optimise the fit to the data. An application analyses economic activity sequences collected for two British birth cohorts born in 1958 and 1970, investigating historical changes in passages to adulthood. The results suggest that passages into adulthood have become more diverse since the 1970s. The analytic strategy produced a classification with better fit than straightforward CA.