Screening with Tests

College admissions tests like the SAT are a crucial element of a process that is pivotal for the lives of millions of American families every year. The United States is exceptional in that these tests are provided by private companies that earn revenues from fees paid by students who take the tests. The Principal Investigator is studying how revenue generation shapes the design of these tests and in turn how the design impacts the admission prospects of students across the spectrum of economic and educational circumstances of American students. Numerous distortions are engendered by the profit motive. Tests are less informative about student ability than would be ideal. Students with strong educational backgrounds are bunched into near-perfect scores making it impossible to further distinguish them. Tests are designed to make it difficult for students with weaker educational backgrounds to achieve scores that show their true ability. Policies like test-optional, regulatory intervention, or even a centralized publicly administered test are examined. The research combines the frontier theoretical tools of information design and mechanism design to study the optimal design of a screening test. Novel features of this screening environment create new challenges for the analysis. For example, the problem fails the classical revenue equivalence framework necessitating an alternative approach for characterizing optimal menus. Moreover, the pattern of binding incentive constraints at an optimum is dramatically different than the typical setting. The research therefore brings a whole new set of analytical tools that will be useful for furthering the study of screening tests, for example in application to the variety of certification intermediaries like pharmaceutical testing, account auditors, insurance and underwriting. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.