Elizabeth Bruch (University of Michigan)
03/26/2012, at noon in room 6050 ISR-Thompson.
For over a century, sociologists and demographers have documented spousal similarity on the basis of education, race/ethnicity, income, physical attractiveness, and other status attributes. This work has been largely descriptive, focused on more on documenting patterns of endogamy and homogamy rather than explaining them (Kalmijn 1998). The disjunction between the descriptive focus of empirical studies of marriage patterns and theories about the underlying mechanisms is largely due to limitations in available data and methods. I use unique data from an on-line dating website to analyze individuals’ preferences for mates in the early stages of mate search. The overarching goal is to understand how mate preferences might aggregate to generate observed patterns of assortative mating. I first estimate statistical models describing the likelihood that a man or woman contacts a member of the opposite sex, given one's own attributes as well as the attributes of potential matches. I then explore how individuals’ preferences for mates interact with the population joint distribution of attributes to generate matching outcomes. I use a static, two-sided matching model to explore how changes in the underlying distribution of available mates—for example, an increase in the average education of women or parity in the educational distributions of blacks and whites—influence observed matching and aggregate patterns of homogamy.