Trends in Income-Based Inequality in Postsecondary Education

Abstract

There has been widespread concern about both widening educational attainment gaps between children from upper- and lower-income families and widening disparities in parental behaviors that may be associated with widening attainment gaps. This dissertation comprises two studies that examined income-based gaps in attainment and parental behaviors between adolescents from families at the 90th and 10th percentile ranks of the income distribution for three cohorts of high school students between 1992 and 2013. In the first paper, I examined changes over time in income-based gaps in parental involvement in the college-going process for two high school cohorts. Between the high school class of 1992 and the class of 2004, I found parents’ aspirations for their adolescent’s educational attainment increased across all income groups, with the largest increases among parents of low-income students, resulting in a narrowing of the gap. In contrast, income gaps in parents’ financial investments for college and student college entrance exam preparation and test taking widened. I examined potential explanations for growing gaps and found support primarily for changing associations between income and parental involvement rather than rising income inequality. In the second study, I documented income-based trends in college enrollment and college completion among cohorts comprising students enrolled in high school as of 9th or 10th grade. I found that the 90/10 college enrollment gap narrowed between 1992 and 2004 and then remained stable between 2004 and 2013. Gaps in four-year college and highly selective college enrollment also narrowed, but only slightly, and then stabilized. Finally, I found that the college completion gap narrowed between 1992 and 2004. These findings stand in contrast to some earlier studies. In attempting to reconcile results across different datasets, I showed that prior estimates of the widening of the enrollment gap between 1980 and 2000 were overstated because the relationship between enrollment timing and income has changed over time, and estimates of gap changes between cohorts were sensitive to the age/timing of when college enrollment was measured.