Partnership formation and dissolution over the life course: applying sequence analysis and event history analysis in the study of recurrent events


  • Satu Helske University of Jyväskylä
  • Fiona Steele London School of Economics
  • Katja Kokko University of Jyväskylä
  • Eija Räikkönen
  • Mervi Eerola



partnership formation, partnership dissolution, sequence analysis, event history analysis, repeated events


We present two types of approach to the analysis of recurrent events for discretely measured data, and show how these methods can complement each other when analysing co-residential partnership histories. Sequence analysis is a descriptive tool that gives an overall picture of the data and helps to find typical and atypical patterns in histories. Event history analysis is used to make conclusions about the effects of covariates on the timing and duration of the partnerships. As a substantive question, we studied how family background and childhood socio-emotional characteristics were related to later partnership formation and stability in a Finnish cohort born in 1959. We found that high self-control of emotions at age 8 was related to a lower risk of partnership dissolution and for women a lower probability of repartnering. Child-centred parenting practices during childhood were related to a lower risk of dissolution for women. Socially active boys were faster at forming partnerships as men.

Author Biographies

Satu Helske, University of Jyväskylä

PhD Student, Depatment of Mathematics and Statistics

Fiona Steele, London School of Economics

Professor, Department of Statistics

Katja Kokko, University of Jyväskylä

Research Director, Gerontology Research Center, Department of Health Sciences

Eija Räikkönen

Post Doc Researcher, Faculty of Education

Mervi Eerola

Professor, Department of Mathematics and Statistics