Who Drops Out from College? A Study of Social Origin at a Midwestern Teaching University

Abstract

Previous research has examined the degree to which social origins affect college completion, but few have studied the association of family background, social class, and neighborhood contexts with regard to the rate of four-year college attrition. To fill this gap, this study utilizes rich administrative data on first-time (students who have not completed any post-secondary courses), full-time freshman cohorts (2007-2014) from a four-year Midwestern teaching university which provided information on students’ demographic information, including parental education and income, academic performance, and family background via admission and Federal Application for Federal Student Aid (FASFA) applications. I supplement these with secondary data on students’ county and high school socioeconomic characteristics. Linear probability and hazard models are estimated. Primary amongst the findings is that parental education is the significant predictor in dropping out of college rather than parental income, even when controlling for academic preparation and a variety of other family and neighborhood variables. Being a first-generation student, someone who does not have a parent with a college degree, significantly and substantially increases the likelihood of dropping out of college, as does being male. To a lesser though still significant extent, county unemployment also predicts retention or withdrawal, suggesting the importance of neighborhood effects. The results imply the value of cultural rather than economic capital transmission in students’ college success and lend further evidence for the widening class inequality gap regarding college completion. This study is especially significant for educational sociologists and higher education retention programs, providing empirical data from which to draw to create targeted intervention for potentially at-risk freshman.