Analysing Social Rhythms by Optimal Matching: Working week schedules of UK 2000
Abstract
Sequence analysis is a useful tool for analysing time diary data, which not only provide detailed information about frequency and duration of daily activities (e.g. paid work, housework, leisure and rest), but also sequences of these activities. We conduct optimal matching analyses (OM) on “workweek-grid” data from the UK Time Use Survey (2000-01) and the French Time Use Survey (1998-99). Both surveys provide detailed information about the scheduling of paid work over a week with 15-minute time slots. A special case of Optimal matching (Dynamic Hamming Matching) with no insertion and deletion operations is used in order to avoid altering the structure and length of workweeks. This variant of OM is applied in two steps. First, we build a typology of workdays by applying OM on the days people worked. Second, based on the typology of workdays, we build a simplified workweek and then a secondary level of empirical typology, i.e. the typology of workweeks. Results show the typologies of workdays and workweeks are very similar in the two countries, but the proportions of these various types of workdays and workweeks differ, especially for the case of part-time workweeks.