Many parents feel strongly that where their child begins school will have long-term effects on their child’s education outcomes. Yet, despite this widely held belief surprisingly little longitudinal research on the role of school systems on students’ college outcomes has been conducted. In general, education research is a highly siloed endeavor, with little spillover acknowledged between the elementary school, secondary and postsecondary years. Education inputs and outputs are usually measured in close proximity to one another: the beginning and end of the school year, from one school year to the next, or on occasion from one level of schooling to the next. However, the cumulative nature of schools or the school environment is rarely assessed.
These studies examine whether school feeder patterns correspond to inequities in four-year college completion for students of certain race, ethnicity, gender, socio-economic status, and disability status. Such research illuminates why gaps in college completion persist despite policy and practice efforts to reduce such inequalities. My dissertation contributes significantly to the fields of education research and social welfare scholarship in numerous ways. These studies are the first to follow complete cohorts of public school students from elementary school through college and to examine differences in educational attainment based on disability status, race, gender and socio-economic status. These studies also highlight the differences that exist across subgroups of students net of tested achievement levels, and raises questions about the connection between test scores and later educational attainment. Lastly, my dissertation contributes to the opportunity gap and cumulative advantage literature by positing the original concept of postsecondary pathway, defined as the unique combination of education opportunities students engage between elementary school and college.
SUMMARY
Study one of the dissertation utilizes descriptive statistics and logistic regression analysis to provides a macro-level view of differences in the longitudinal education outcomes of academically similar 8th graders across demographic subgroups. The sample includes all Chicago Public Schools (CPS) eighth-graders (n=38,166) in 2004 and 2005, including students who are usually omitted from analysis, such as students attending alternative schools, and students with disabilities. Data include CPS administrative records, latent test scores derived by researchers at The UChicago Consortium on School Research, and college enrollment outcomes obtained for National Student Clearinghouse. Study two utilizes hierarchical linear models to estimate the overall between elementary-school difference in college completion prospects for five cohorts of CPS 8th graders (n=99,253), and identifies how much of this difference can be attributed to the high schools and colleges students later attend.
Results from study one indicate students with emotional or behavioral disabilities experience significantly reduced longitudinal education prospects compared to their peers. Results also indicate that the racial gap and poverty gap in education prospects widens as 8th graders move from high school graduation to four-year college enrollment, and eventually four-year college completion. Notably, the largest differences in education prospects ware for 8th graders in the bottom 20th percentile versus top 20th percentile of latent test scores, even after controlling for demographics and special education status. As test scores are heavily relied on in the district’s high school enrollment process, differences in longitudinal education outcomes by test score raise questions as to whether utilizing test scores so early on in a student’s career places them on an academic pathway of less postsecondary opportunity before they begin high school.
Results from study two suggest that once students test scores are controlled for, students with disabilities attend elementary schools where their academically similar peers without disabilities have similar prospects of earning a college degree. Such is not the case for Black and Latino students. After controlling for test scores and covariates, Black and Latino students are far more likely to attend elementary schools where their academic peers have drastically lower prospects of earning a college degree compared to their White and Asian peers. Results from study two also indicates that approximately one-third of the between elementary-school difference in students’ college completion prospects can be attributed to the high school students attend in the 9th grade, and that 60 percent of the variation at the high-school level can be explained by the institutional graduation rates of the colleges students ultimately attend. Such findings suggest elementary to high school to college enrollment patterns can account for a substantial portion of the between student difference in college completion prospects for academically similar students.