LAUSR

Summary The aim of this paper was to establish whether the differences in the risks of union dissolution between cohabitors and non-cohabitors in Uganda have converged over time using event history data. Data were collected in 2013 from 1200 women in central Uganda using retrospective methods. Of these, 839 provided information on three types of first union: women who married directly (without first cohabiting), those who married following cohabitation and those who were still cohabiting. The data were analysed using decrement lifetable analysis. Though the analysis indicated a small difference in the timing of first union dissolution for women who married directly, no evidence was found that the difference in the risk of union dissolution between cohabitors and non-cohabitors had converged over the 9-year period following first union. Women’s union/marriage status, number of living children in a union, parental union status and birth cohort were found to significantly influence the timing of union dissolution. Overall, the rate of union dissolution was fairly high, regardless of type of union.