Sherod Thaxton (UCLA; Google Scholar), A Comment on Sander and Steinbuch's “Mismatch and Bar Passage: A School-Specific Analysis”:

Richard Sander and Robert Steinbuch’s, “Mismatch and Bar Passage: A School-Specific Analysis,” offers a statistical analysis of the “law school mismatch hypothesis” in an effort to explain racial differences in the likelihood of passing the bar examination. Sander and Steinbuch claim their analysis is an improvement on prior studies of mismatch—which relied on data from the Law School Admission Council’s (LSAC) Bar Passage Study (BPS)—because their data are (a) more recent (the BPS is nearly 25 years old) and (b) permit the construction of a school-specific measure of mismatch than was not possible with the BPS.

These improvements, notwithstanding, the same conceptual and methodological mistakes previously identified in the empirical literature on mismatch analyzing the BPS data are present in Sander and Steinbuch’s study of school-specific data, thereby calling into question Sander and Steinbuch’s conclusion that “the mismatch effects in their models can therefore account for the large disparities in bar passage across racial lines.” This paper identifies six problems with Sander and Steinbuch’s empirical analysis and attempts to propose a sensible solution, or set of solutions, to each problem.