This article develops and applies a stochastic two-sided matching model to analyze marriage patterns in the United States using 1 percent samples from the 2010 and 2019 American Community Survey, accessed via the Integrated Public Use Microdata Series. This approach disentangles two sources of change in marriage patterns over time: individuals’ preferences for partner characteristics (“forces of attraction”) and the numbers and composition of potential partners (“partner availability”). As illustrated by our empirical application, the model provides a flexible and unified analytical framework to address a broad range of relevant questions in marriage research, offering valuable new perspectives on marriage dynamics and facilitating future research, despite the limitation that the model does not separately identify individual-specific preferences.
Yuan et al. (Thu,) studied this question.