A nonlinear simultaneous probit-model for the investigation of decision-making processes: modelling the process of setting up a family in partnerships

This article proposes an approach for modelling the interactive influences of two (or more) actors in decision-making processes. We use a nonlinear simultaneous probit-model and show how the problem of identification for estimating the relative effects of the actors can be solved. The formal model will be applied to examine the decision-making process for setting up a family in partnerships. We model a trivariate distribution consisting of the wifes’ desire to have a child or disposition toward the generative decision, the husbands’ disposition and the joint generative decision. We show how the parameter can be used to assess the relative importance of both partners’ dispositions for the decision, the reciprocal influence of both partners’ dispositions within the interaction process, and the relevance of both partners’ biographical contexts in relation to their own disposition as well as to that of the partner. The analysis is based on a three-stage estimation strategy which is implemented in MECOSA 3 and we use data from the Bamberger married couple panel.