European Population Conference, Sept 1-4, 2010, Vienna, Austria
June 1, 2010
A clear preference for repartnering in the form of a cohabiting rather than a marital union has been noted by studies examining repartnering of lone parents (Böheim and Ermisch, 1998) as well as repartnering of all individuals (Ermisch and Francesconi, 2000; Wu and Schimmele, 2005; Skew, Evans and Gray, 2009). However, what is unknown for lone mothers is how the effects of explanatory covariates may differ depending on the type of new partnership a lone mother enters. This paper uses data collected by the British Household Panel Survey (BHPS) and employs discrete-time multinomial logistic hazard models to simultaneously model entry into marriage and cohabitation. Initially, results suggest that the route of entry into lone motherhood has a strong impact on the choice of new union type; those who entered through the dissolution of a marriage are more likely to form a marriage than those entering through the breakdown of a cohabiting union or through giving birth whilst single and never-married. However, much of this is found to result from previously married lone mothers reconciling with their former spouse. Removing all those that reconcile with a former partner from the model considerably reduces the magnitude of this difference and renders it statistically insignificant. Overall, age is important for both marriage and cohabitation, but number of children, ethnicity and social class are only important in relation to forming a marriage and mental health and employment status only significant with respect to forming a cohabiting union.