Working Paper: NBER ID: w14275
Authors: Jesse Rothstein; Albert Yoon
Abstract: An important criticism of race-based higher education admission preferences is that they may hurt minority students who attend more selective schools than they would in the absence of such preferences. We categorize the non-experimental research designs available for the study of so-called "mismatch" effects and evaluate the likely biases in each. We select two comparisons and use them to examine mismatch effects in law school. We find no evidence of mismatch effects on any students' employment outcomes or on the graduation or bar passage rates of black students with moderate or strong entering credentials. What evidence there is for mismatch comes from less-qualified black students who typically attend second- or third-tier schools. Many of these students would not have been admitted to any law school without preferences, however, and the resulting sample selection prevents strong conclusions.
Keywords: No keywords provided
JEL Codes: I21; J15; K30
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| affirmative action does not harm black students with moderate or strong entering credentials (J78) | employment outcomes (J68) |
| affirmative action does not harm black students with moderate or strong entering credentials (J78) | graduation rates (I23) |
| affirmative action does not harm black students with moderate or strong entering credentials (J78) | bar passage rates (K29) |
| students attending highly selective schools (I23) | better outcomes than equally qualified students attending less selective schools (I24) |
| less-qualified black students (I24) | evidence of mismatch effects (C52) |
| sample selection bias (C83) | observed outcomes for less-qualified black students (I24) |
| data indicate that affirmative action benefits rather than harms black students (J78) | particularly those with moderate or better qualifications (J24) |