Doctoral Dissertation Research: The Role of Recommendation Letters in Selective College Admissions

University Of Notre Dame, Notre Dame IN

Investigators

Abstract

This dissertation builds on research regarding school effects, status attainment, and stratification by examining the role of teacher recommendations in selective college admissions. Through the process of effectively maintained inequality, parents may attempt to secure advantages for their children by sending them to well-resourced schools that are believed to provide an advantage in learning and life-outcomes. According to school effects research, though, school characteristics account for a relatively small proportion of the variation in student outcomes. Using a mixed-methods approach, this dissertation tests whether letters of recommendation written by teachers at well-resourced schools provide an advantage in highly selective college admissions relative to equally qualified applicants from less privileged schools. Letters of recommendation are posited to provide an advantage to students from privileged high schools in selective college admissions. These letters are a mechanism through which school characteristics directly affect the likelihood of college admission that is not mediated by student achievement. This project introduces new methods and data to sociological inquiry: computational linguistic analysis, which lies at the intersection of computer and social sciences. Key concepts and methods will be translated and applied in a way that is accessible and useful to sociologists. Data generated by this project will eventually be made available to other scholars and substantive findings will be shared with practitioners. Qualitative interviews with 15-30 admissions officers at highly selective colleges and universities will establish the role of teacher evaluations in the admissions process. The core of the project is a collection of nearly 25,000 teacher evaluations that were submitted to a top research university which are linked to limited biographical information and achievement indicators of applicants. A sample of these letters is analyzed using traditional qualitative methods and the entire corpus will be quantitatively examined using computational linguistic analysis. Finally, an experiment will test the findings of the qualitative interviews and measure how differences in letters of recommendation associated with high school characteristics impact admissions decisions at highly selective universities.