For decades, U.S. higher education enrollments have been stratified with students from wealthier households consistently attending postsecondary institutions at higher rates than low-income students. The disparity in postsecondary participation rates by family income is a systemic issue (M. J. Bailey & Dynarski, 2011), meaning the phenomenon is the result of a combination of factors within society rather than one factor alone. Guided by a critical theory perspective and the assumptions behind Perna’s (2006) proposed conceptual model for student college choice research, the current study sought to examine the extent to which policies and practices at the postsecondary institution level may be contributing to the inequity in higher education enrollments by family income. Based upon theory and findings from prior research, the presumed causal effects of tuition costs, student financial aid, selectivity, and proximity on the average amount of enrolled undergraduate students’ aggregate capital (i.e., cumulative available resources) was constructed in one hypothesized structural model. Four-year, public postsecondary institutions were the unit of analysis.
The purpose of the current study was to test the hypothesized structural model. Using data from the 2011-12 National Postsecondary Student Aid Study, student-level data were combined by postsecondary institution to establish a final sample of N = 330 (rounded to the nearest ten) four-year, public postsecondary institutions. Structural equation modeling (SEM) was performed to (a) assess the overall fit of the hypothesized structural model to the sample data; (b) determine the amount of variance in the average amount of enrolled undergraduate students’ aggregate capital that could be explained by the hypothesized structural model; and (c) identify the direct, indirect, and total effects among the variables included in the hypothesized structural model. Though a confirmatory SEM analysis indicated the hypothesized structural model was a poor fit to the data, respecifications to the model via exploratory SEM analyses revealed a modified hypothesized structural model that was, provisionally, considered a good fit to the data. Findings from this study supported the assumption that the majority of the policies and practices included in the hypothesized structural model, including non-need-based (merit-based) gift aid awards funded by state governments and postsecondary institutions, are positively related to the average amount of enrolled undergraduate students’ aggregate capital and, therefore, may be creating barriers to enrollment for low-income students.
To foster low-income student enrollment, college administrators at four-year, public postsecondary institutions are encouraged to reduce tuition increases and incorporate financial need as a criterion for institutional gift aid awards. In addition, for leaders at non-selective institutions, focusing recruitment efforts on students who reside within close proximity to their institutions is recommended. Recommendations for future research include validating the modified hypothesized structural model, examining other models, and continued investigations of the measurement of the multidimensional construct, aggregate capital.