How do Affirmative Action Bans Affect Student Racial and Socio-economic Composition?

Open Access
Liu, Huacong
Graduate Program:
Higher Education
Doctor of Philosophy
Document Type:
Date of Defense:
May 30, 2017
Committee Members:
  • Liang Zhang, Dissertation Advisor
  • Liang Zhang, Committee Chair
  • John Jesse Cheslock, Committee Member
  • Alicia C Dowd, Committee Member
  • Mark John Roberts, Outside Member
  • affirmative action
  • student-body composition
  • low-income students
  • synthetic control analysis
This study investigates the consequences of state-wide affirmative action bans on students’ college enrollment outcomes. In particular, I examine four states that have banned the consideration of race in college admissions in public colleges and universities in the past decade, Michigan, Nebraska, Arizona, and New Hampshire. The enactment of these bans provides quasi-experimental variation in affirmative-action policies. The bans come from a mix of sources, ranging from voter initiatives to executive orders and court rulings. The variation in time and location of these bans provides useful variation in identifying the effect of banning affirmative-action policies on college enrollment outcomes. This dissertation consists of two parts in which I utilize data from two different sources, the Integrated Postsecondary Education Data System (IPEDS) and the Current Population Survey (CPS) between 1999 and 2014 to examine the effects of these bans. In the first part, I examine these relationships using the IPEDS which contains institutional level data. Findings suggest that although overall the bans do not significantly decrease racial minorities enrollment at 4-year public colleges and universities, a ban is associated with a significant reduction of racial minorities enrollment in Michigan and New Hampshire at public 4-years. Beyond the impact on racial minority students, the bans also change the socioeconomic composition of the freshman classes. In particular, the bans increase low-income students’ college enrollment in Michigan, Arizona, and New Hampshire. Moreover, this study utilizes a quantitative case study method - the synthetic control method to conduct a close-up examination of the four public flagship universities. Findings demonstrate that for a competitive public flagship, as is the case for the University of Michigan, a ban decreases the racial diversity of its freshman class. For non-competitive public flagships (i.e. the University of Nebraska, the University of Arizona, and the University of New Hampshire), the ban changes the socioeconomic composition of the student body - low-income students’ representation at these campuses increases in years following the bans. The rest part of the dissertation uses the CPS data and examines the effects of the bans across racial and income groups. Findings suggest that the bans affect low-income students differently than their effects on high-income students across racial groups. In particular, the bans decrease low-income racial minorities’ chances of college enrollment more than they do for high-income ones. Yet, for white students, the bans increase low-income students’ chances of college enrollment more than they do for high-income ones. These findings shed new light on the discussion of the impact of affirmative action policies on the intersection of race and income-based class.