We examine how students of varying abilities sort (and are sorted) into colleges of varying qualities. Our data indicate substantial amounts of both academic undermatch (high-ability students at low-quality colleges) and academic overmatch (low-ability students at high-quality colleges). Student application and enrollment decisions, rather than college admission decisions, drive most deviations from academic assortative matching. Financial constraints, information, and the public college options facing students all affect this sorting, but mainly via college quality rather than the match between ability and quality. More informed students attend higher-quality colleges, even when doing so involves overmatching.

Existing studies of the effects of college quality on wages typically rely on a single proxy variable for college quality. This study questions the wisdom of using a single proxy given that it likely contains substantial measurement error. We consider four econometric approaches to the problem that involve the use of multiple proxies for college quality: factor analysis, instruments variables, a method recently proposed by Lubotsky and Wittenberg, and a GMM estimator. Our estimates suggest that the existing literature understates the wage effects of college quality and illustrate the value of using multiple proxies in this and other similar contexts.

This article decomposes the participation process of a prototypical program into eligibility, awareness, application, acceptance, and enrollment. With this decomposition, we determine the sources of unequal participation for different groups and demonstrate that variables often have very different effects at different stages in the participation process. Our analysis shows that personal choices substantially affect participation and that awareness of program eligibility is a major source of variation in participation.

This paper re-creates three of the samples from LaLonde’s famous 1986 paper that began the literature on “within-study designs” that uses experiments as benchmarks against which to assess the performance of nonexperimental identification strategies. In particular, we recreate the experimental data for the target group of women on welfare from the National Supported Work (NSW) Demonstration and two of the corresponding comparison groups drawn from the Panel Study of Income Dynamics (PSID). The loss of these data resulted in the (sizable) subsequent literature devoting its attention solely to the NSW men. In addition to repeating LaLonde’s analyses on our recreations of his files for the AFDC women, we apply (many of) the estimators from later papers by Dehejia and Wahba and by Smith and Todd to these data. Our findings support the general view in the literature that women on welfare pose a less difficult selection problem when evaluating employment and training programs. They also call into question the generalizability of some of the broad conclusions that Dehejia and Wahba and Smith and Todd draw from their analyses of the NSW men.