| Dongbin Kim
UC Los Angeles
Multilevel analysis of the effect of loans on students' degree attainment: Differences by student and institutional characteristics
FINAL REPORT:
Among current trends of student financial aid, the increasing reliance on loans to pay for higher education is particularly noticeable. Considering that the central objective of financial aid is to provide equal educational opportunity to students, it is crucial to examine if loans promote educational opportunity, not just measured in terms of access (getting students in the college door), but in terms of success (completing a degree). Nevertheless, there is no extensive research that examines the distinct effects of loans on students' degree attainment by individual as well as institutional characteristics. Therefore, this study investigated how borrowing money is related to degree attainment with particular focus on students' race, parental income, and institutional characteristics. Hierarchical Generalized Linear Model (HGLM) was used to clarify the effects of individual and institutional level variables on degree attainment. The data of this study was derived from the Beginning Postsecondary Student (BPS) Survey in 1989/90 and 1993/94.
From the HGLM analysis using loan amounts as a continuous variable (increases of $1,000), it became clear that loan amount had a significant positive impact on studentsÕ degree attainment. However, the positive impacts of higher loans were greater for low-SES students, followed by middle-SES students. No significant impact of loan amount was found for the rates of degree attainment among high-SES students.
From the second HGLM analysis using loan amounts as a categorical variable, the distinctive differences in the effect of loans on degree attainment by loan amount were clarified: students who had loans more than $11,000 were more likely to complete a bachelorÕs degree than students who did not have loans. In contrast, students who had loans less than $11,000 were less likely to complete a degree than their counterpart who did not have loans. However, it became clear that the face value of a loan amount is different from its perceived value for subgroups of students when interaction terms between loans and student SES, and between loans and student race, were analyzed. If Black students had loans more than $11,000, they were significantly more likely to complete a degree than other Black students who did not have loans. SES also plays a significant role in determining the effect of loans on degree attainment: if low SES students had more than $11,000 loans, they were significantly more likely to complete a degree than their counterpart low-SES students who had no loans. Of the students who have loans of $11,000 or more, middle-SES students have the highest rates of degree attainment, followed by low-SES and lastly high-SES students.
Institutional-level analysis indicates that students respond differently to increases in tuition by race and SES. For all students, higher tuition increases rates of degree attainment (i.e., students' price sensitivity is a positive direction). By contrast, high tuition decreases the rates of degree attainment for low-SES students with loans and Black students with loans of $11,000 or more. Thus when low-SES or Black students need to pay more tuition, they tend not to complete a degree.
A recent statistic shows that only 17% of low-income students received a bachelorÕs degree by age 24 compared with 52% of upper income students. Degree completion rates for African-American are also far lower than those for Asians and Whites. In this context, the findings from this study are particularly promising because they suggest that there is room for improving the chronic disparities in degree attainment rates by race or SES. With the information, policy makers will have an opportunity to provide financial aid programs that can significantly increase degree attainment rate, especially for Black and low SES students.
Back to Funded Dissertation Grants Page |