Closing the Gap?: Admissions & Enrollments at the Texas Public Flagships Before and After Affirmative Action

This paper uses administrative records to evaluate the impact of eliminating affirmative action in Texas on admissions and enrollments at the state's two most selective public universities during the 1990s. Although Texas is rapidly becoming a majority minority state, the demographic profile of the two public flagships has failed to keep pace with the growth of minority groups among college-age students. Unless qualified minority students who are admitted to the public flagships actually enroll, both institutions will weaken their reach in educating a leadership class for the State's rapidly growing minority population. Changes in admissions probabilities during the 1990s reveal substantial changes in the structure of opportunities for African American, white, Hispanic and Asian American applicants. On the one hand, H.B. 588 leveled the playing field for students ranked in the top decile of their class by equalizing their nearly equal odds of admission before the Hopwood decision. This change benefited high achieving minority students who may have been rejected for low test scores or poor essays before the top ten percent law was implemented in 1998. On the other hand, Hopwood reversed the favorable admission probability enjoyed by minority applicants who graduated in the second decile of their class or below, but the non-overlapping groups simulation indicates that no more than three percent of all admitted students were affirmative admits at either institution. The Hopwood reversal benefits white applicants, and especially Asian Americans seeking admission to UT. Moreover, contrary to public criticisms alleging that H.B.588 privileges high performing students who attend low performing schools, the admission probability of students from the major feeder high schools who ranked in the second decile actually rose at both institutions, but most especially at UT. The ban on affirmative action did have a chilling effect on enrollment odds of minority students admitted to the public flagships. For African Americans, Hispanics and Asian Americans admitted to A&M, the net odds of enrollment relative to whites, which were well below unity for all groups before Hopwood, fell after 1996. At UT the odds ratios for enrollment conditional on admission, which were below parity before 1996, changed very little for Hispanics and Asian Americans, and dropped slightly for African Americans. Admitted white applicants are significantly more likely to enroll than African Americans and Hispanics granted admission to UT, and Asian Americans accepted to UT after 1997 were about 16 percent more likely to matriculate than their white statistical counterparts. The enrollment analyses forcefully demonstrate that by itself, the top ten percent policy is NOT an alternative to race sensitive admissions; rather, it is a merit-based admission plan that emphasizes high school academic achievement in the admission decision while de-emphasizing standardized achievement tests for top ranked students. In the absence of financial support to needy students coupled with a vigorous outreach program to high schools populated by minority and economically disadvantaged students, the top ten percent policy will not diversify campuses of selective universities.