BCSER: IID: Evaluating Letters of Reference to Engineering Doctoral Programs for Racial and Gender Bias
PERIOD:
TO
Dr. Lisa Lattuca will serve as a mentor to PI Dr. Allyson Flaster (PhD '17) on this project that considers how the content of letters of recommendation may have an effect on admittance to engineering doctoral programs for applicants from minoritized backgrounds. Dr. Flaster is a graduate of the Marsal School's Center for the Study of Higher and Postsecondary Education and is an Assistant Research Scientist at U-M's Institute for Social Research.
Faculty largely determine which college students can continue on to STEM research careers, and thus who can contribute to scientific knowledge production. One of the key ways they exercise this gatekeeping role is by deciding who to recommend and admit to graduate programs. Since STEM Ph.D. program enrollment is highly unequal by race and gender, it is important to examine how practices exercised during graduate admissions can perpetuate enrollment disparities. The PI will use information from interviews with engineering faculty and an extensive dataset of applications to engineering doctoral programs to develop a corpus of language associated with potential for success in select engineering fields. The corpus will be combined with existing language dictionaries to identify implicit bias in written materials. Then, using natural language processing techniques on a large scale, the PI will use these resources to evaluate if the language used in the letters to describe applicants from over- and under-represented racial and gender backgrounds is systematically different.
Primary Investigator(s)
