Association for Public Policy Analysis & Management

As a primary type of financial aids to support student success in higher education, the total amount of federal student loans has been increasing during the past decades. Whether federal loans are effective in student completion, which is specified as baccalaureate degree attainment in this study, is a question worth exploring. We reviewed literatures regarding the status quo and factors influencing students’ baccalaureate degree attainment, as well as the positive and negative impact of federal loans. Additionally, we examined the gap between community college transfer students and four-year native students, illustrating that these two groups of students are different in terms of academic performance, financial status, college experience, etc. These differences may contribute to different impacts of federal loans on their degree attainment.

As one of the most commonly accepted concepts in education economics, human capital theory explained how both individuals and the society invest in higher education in order to make “profits” out of it. Becker (1993) pointed out that education is one of the most important investments in human capital, and financially disadvantaged families could adopt the practice of borrowing to finance their higher education. A second theory focuses on community college transfer students’ baccalaureate degree completion and its influential factors. Examining this specific population, Wang (2009) categorized influential factors into three groups, including precollege characteristics, college experiences, and environmental factors.

By employing the Education Longitudinal Study of 2002, we aimed to answer two research questions: 1) What is the impact of student borrowing federal student loans on baccalaureate degree attainment? 2) For community college transfer students and four-year native students, do they respond differently to federal loan’s impact on baccalaureate degree attainemnt? To answer these questions, we utilized Inverse Probability of Treatment Weighting (IPTW) to eliminate bias between the community college transfer group and the four-year native group, and to include all the data without imputation (Hogan & Lancaster, 2004). While the propensity score is e_i = prob (z = 1|Xi), the IPTW is defined as W_i = Z_i/e_i + (1- Z_i)/(1-e_i) (Austin, 2011). 

After incorporating IPTW, we included a set of logistic regressions to test the relationship between federal loan taking, baccalaureate degree attainment, and sector of first attended institution: logit(Y) = a + b₁X₁ + b₂X₂ +… + b_iX_i + ε. Four models were used to explicit different groups of variables. Preliminary results indicate that taking federal loans has significantly positive impact on the probability of baccalaureate degree attainment. For each $1,000 increase in cumulative federal loan, the probability increases 1.6 percent. Cumulative amounts of parent loans and subsidized loans are also positively associated with degree attainment; however, no such relationship was found for unsubsidized loans. Moreover, all four models show that with the support of federal loans, students that attended community colleges first and transferred to four-year institutions actually have a higher probability of obtaining a baccalaureate degree than attending four-year institutions directly. This study provides both research and practical implications regarding financing baccalaureate degree and the recently released “Free Community College” proposal.