Seminar, 12 hours/PhD level
This workshop introduces sequence analysis for social science research. Sequence analysis is a set of tools to analyze sequences of categorical states that represent processes that unfold over time. These tools were first introduced to social sciences in the form of optimal matching analysis by Andrew Abbott who imported this approach from biology and computer science. Sequence analysis has attracted increasing attention in the social sciences for the analysis of longitudinal life-course data, as labor market careers, transitions to adulthood, or family formation. More recently, sequences are scrutinized in other fields of social sciences such as geography, archaeology, and economics This workshop covers longitudinal data management (only briefly) and basic techniques of sequence analysis to (1) describe and (2) visualize sequences, (3) to identify typical patterns among a set of sequences using cluster analysis, and (4) to examine the antecedents and consequences of these patterns. Recent approaches combining sequence analysis with event history analysis and propensity score matching will be also discussed. All analytical steps are demonstrated with hands-on examples using Stata (for data preparation and basic sequence analysis using commands from the SQ and SADI ados) and R (TraMineR and ancillary packages).