Consequences of Racial School Segregation for College Outcomes

• In Event: 63.018 - Promising Scholarship in Education: Dissertation Fellows and Their Research

Abstract

Nearly 60 years after racial school segregation was deemed unconstitutional, it continues to be an issue in American education. Students in minority-segregated schools on average have less access to key educational building blocks, including highly qualified teachers, rigorous curricula, and small class sizes. Research to date generally finds negative consequences for attending minority-segregated schools on academic achievement and lasting impact on educational attainment and occupational outcomes (Coleman et al., 1966; Wells & Crain, 1994; Mickelson 2008; Mickelson & Nkomo, 2012).
Most studies of racial school segregation measure exposure to racially segregated schools (black- or minority-concentrated schools) at one point in time. This characterization is problematic for two reasons. First, it unrealistically simplifies student experiences by ignoring changing policy and demographic contexts, including increasing district-wide school choice policies, the elimination of race as a possible factor in admissions decisions, and the rollback of court-ordered desegregation plans, a rising proportion of minority students and increasing minority movement to the suburbs. Second, it ignores key temporal aspects of segregation exposure, including whether the highest levels of segregation are experienced in earlier or later grades, whether exposure is stable or changing over time, and the total length of time exposed. As a result, we do not yet understand the consequences of variation in the timing, sequencing, and duration of segregation exposure on college outcomes. At a time when college is critical for economic mobility and when minority students continue to lag behind in college enrollment and completion (Fry, 2011), evidence on these issues can provide important directions for educational policy.
The primary objective of this paper is to estimate the effect of differences in the timing, sequencing, and duration of exposure to black school segregation on postsecondary educational outcomes. Exposure to school segregation during adolescence can directly affect college enrollment and completion through several mechanisms, including school quality or information networks. However, segregation exposure in early adolescence may also indirectly affect college outcomes by affecting characteristics of students that change over time, such as course-taking levels, educational aspirations, or development of problem behaviors, which may in turn influence whether students experience school segregation in later grades and ultimately affect the probability of enrolling in and completing college. Traditional methods for analyzing the effect of school segregation on college outcomes are not suitable when students can be exposed to school segregation at different points and for different lengths of time, and when student characteristics that are related to segregation exposure and college outcomes also vary over time. Controlling for time-varying student characteristics may remove the indirect effects of segregation exposure on postsecondary outcomes, understating its effect. This dissertation uses an innovative method of causal inference—marginal structural models—that incorporates the direct pathways through which segregation exposure affects students’ college outcomes as well as the indirect pathways. Findings will extend our understanding of the consequences of school segregation by estimating the causal effect of three important temporal dimensions of exposure—timing, sequencing, and duration.

Author

• Siri Warkentien, Child Trends