Do Colleges Use High School Accountability Information to Inform Undergraduate Admissions?
*Names in bold indicate Presenter
In this paper, we combine publicly available school accountability data from the State of Florida as well as micro-application data the state’s flagship university to estimate the effect public signals of high school quality on application behaviors, admissions probabilities, and the requisite academic criteria needed to gain admissions. We capitalize on the conditions for a natural experiment generated by the exogenously chosen grade cut scores on a continuous underlying measure of school performance.
We use a regression discontinuity research design and estimate the effects of attending a school that just received a lower letter grade, relative to schools that just received a higher grade to see whether being in a lower-performing school impacted a student’s ability to gain admission. Our study is uniquely positioned given the postsecondary application incentives in the State of Florida. Since 1997, Florida has offered in-state students a large tuition subsidy (Bright Futures scholarship) which creates a strong fiscal incentive for public high school graduates remain in state for their postsecondary education. Using data from five cohorts of first-time college applicants we first model the propensity for students to apply. We then estimated the differential effects of school quality on the probability of admission.
Initial evidence suggests that attending school with a lower signal of quality may impact a student’s selection to apply to the state flagship institution or not. Specifically, we find evidence students enrolled in a “B” school, rather than an “A” school, exhibited higher SAT/ACT scores and high school GPA. After accounting for these initial application differences, we found little to no evidence that postsecondary institutions use signals of school quality to make admissions decisions. Results indicate that signals of school quality do not directly impact the postsecondary admissions process, but rather influence the process by restricting applications from students from “lower quality” high schools. This paper demonstrates promise to contribute to literature on the spillover effects of K-12 policies within higher education. Next, we aim to integrate statewide administrative data from the State of Florid to estimate the differential effects by postsecondary institution type.