The Society for Social Work and Research

2014 Annual Conference

January 15-19, 2014 I Grand Hyatt San Antonio I San Antonio, TX

Do College Expectations Mediate the Effects of Parental College Savings On Four-Year College Attendance Among Hispanic Youth?

Thursday, January 16, 2014: 4:30 PM
HBG Convention Center, Room 002B River Level (San Antonio, TX)
* noted as presenting author
Hyun-a Song, MSW, Student, University of Pittsburgh, Pittsburgh, PA
William Elliott, PhD, Assistant Professor, University of Kansas, Lawrence, KS
Background and Purpose:

College education is seen as a necessary step toward upward social and economic mobility. A more educated work force also enhances the productive capacity of the economy. However, Hispanics have the lowest rate of college attendance among all major racial/ethnic groups. Compounding the problem, Hispanic youth enroll disproportionately in two-year colleges, as opposed to four-year colleges.

Since expectations influence current behavior as well as the achievement of future goals, Hispanics’ low college expectations appear to be implicated in their low college attendance. Studies show that college savings have not only financial but also psychological effects on college attendance. However, the impact of college savings on Hispanic youth’s four-year college attendance, as well as the psychological mechanisms at play between the two, has not yet been fully explored. Furthermore, the role of schools in college attendance has been largely ignored in previous studies.

The main purpose of this study is to identify the role of parental college savings in increasing four-year college attendance among Hispanic youth. We tested whether both parents and their children’s college expectations mediate the effect of college savings on four-year college enrollment. The moderating effect of school characteristics on the association between college savings and college attendance was also examined.   


We analyzed a sample of 2,750 Hispanic youth from the Education Longitudinal Study of 2002/2006. The ELS data set is multi-level: schools were the first-stage unit of selection, with 10th graders randomly selected within schools. Since students were not independent of schools, and since the outcome variable was binary, a two-level HGLM was estimated. The outcome variable was coded as 1 (4-year college attendance) or 0 (less than 4-year college). By using HGLM, we examined the moderating effects of school factors on the relationship between college savings and college attendance. To test the mediating effects of college expectations, the Baron and Kenney method was utilized. A multiple imputation approach was applied to deal with missing data.   


Findings show that Hispanic youth who have college savings are 1.5 times more likely to attend a 4-year college than otherwise similar youth without savings. Moreover, the association between college savings and attendance is significantly mediated by college expectations. Lastly, we found that school characteristics do not moderate the effect of the relationship between savings and college attendance.


A structured savings program, Child Development Accounts (CDAs), has been proposed as a promising asset-building and financial aid approach. Findings of this study suggest that CDAs can help Hispanic youth and parents accumulate savings for higher education, which may be effective strategies for enhancing their expectations and for achieving their educational goals, ultimately, for a better future.