Modeling the consequences of the transition to parenthood: Applications of panel regression methods

Abstract

The transition to parenthood is a topic of substantial interest to family researchers across the social sciences, and many theoretical paradigms have been invoked to understand how it affects men’s and women’s lives. While early empirical scholarship on the transition to parenthood relied on cross-sectional data and methods, the increasing availability of panel data has opened up new analytical pathways—including the possibility to track the same individuals over time as they approach and experience parenthood and their children grow older. By making full use of longitudinal data, researchers can both improve estimation of the consequences of parenthood, as well as advance knowledge by testing more nuanced and complex theoretical premises involving time dynamics. In this article, I present an overview of panel regression models, a family of specifications that can be leveraged for these purposes. In doing so, I discuss the data requirements, advantages and disadvantages of different models, pointing to useful examples of published research. The approaches considered include random effects and fixed effects panel regression models, specifications to model linear and nonlinear time dynamics, and specifications to handle dyadic data structures. The use of these techniques is exemplified via an application considering the effect of motherhood on time pressure using long-running panel data from an Australian national sample, the Household, Income and Labour Dynamics in Australia Survey (n = 68,911 observations; 10,734 women).