| This research examines the determinants of the match between graduating high school seniors and postsecondary institutions in the United States. Individuals in the model choose whether and to which colleges they apply and, ultimately, whether/where to enroll. In the application stage of the process, individuals have expectations about their admission probabilities and they search nonsequentially for the portfolio of college applications that maximizes expected utility. Following application, individuals learn admissions outcomes and acquire more information about the colleges to which they applied before making a final enrollment decision or choosing to enter the labor force or community college. Each college in the model chooses a threshold admission rule that is a function of observable applicant characteristics and an unobservable (to the econometrician and applicant) match value. I assume that colleges are motivated by factors affecting institutional reputation or prestige. Modeling the full set of decisions made during this matching; process insures that parameter estimates are unbiased and captures the equilibrium effects that are crucial to conducting meaningful policy analysis.; The parameters of the structural model are estimated with maximum simulated likelihood. I use data on individuals' postsecondary choices from the National Education Longitudinal Study of 1988 (NELS) and institutional information from the Integrated Postsecondary Education Data System (IPEDS) and Barron's Profiles of American Colleges. Structural parameters represent the true underlying preferences of individuals if the model is correctly specified. As a result, the parameter estimates are invariant to changes in public policy and can be used to analyze policy changes that have not already been implemented, have been implemented on only a small scale, or are not captured by available datasets.; In the present study, predicted individual and institutional behavior is used to examine how the equilibrium match between individuals and colleges is affected by a ban on affirmative action in college admissions. I predict small decreases in minority representation at all four-year colleges as well as at the most competitive institutions following a simulated nationwide ban. I also examine the ability of three potential replacement policies to restore pre-ban diversity levels. |
