From marriage to cohabitation: family formation and parenthood in Denmark

Mette Lausten, SFI - Danish National Centre for Social Research
Siddhartha Baviskar, SFI - Danish National Centre for Social Research
Mette Deding, SFI - Danish National Centre for Social Research
Vibeke Jakobsen, SFI - Danish National Centre for Social Research

The purpose of this paper is to analyse the changing relationship in Denmark between family status and parenthood across cohorts. The norm from ‘the good old days’ of getting married before having children has almost vanished in favour of having children before getting married – 46 % of children in Denmark today are born outside of marriage. We analyse the changing relationship over three birth cohorts using event history analysis. The data used are from Danish administrative registers. We have information on three birth cohorts: all men and women born in Denmark in 1954, 1966 and 1978. Immigrants are not included in the sample as well as Danes not continuously present in Denmark. First, we map the relationship between family status (marriage versus cohabitation) at the birth of children for women and men from the three cohorts. This mapping will provide a much-needed overview of the changing patterns from the perspective of children being born. Next, we analyse changes across the three cohorts in the pathways to family formation and parenthood applying event history analysis in two steps. First, we focus on the transition from being single to being a couple using competing risks duration model where the two transitions modelled are to marriage and cohabitation, respectively. Second, we focus on the transition to parenthood also applying a duration model. This analysis is limited to couples where the probability of having a first child differs between married and cohabitating couples. For some cohabitating couples, having the first child is preceded by marriage which will also be taken into account in the modelling. Both nonparametric and parametric models including socio-economic variables will be estimated. We will estimate the models separately for each of the three birth cohort because we want to focus on changes in the pathways over time.