Academic Undermatch: Investigating Substantial Undermatch in a Nationally-Representative Sample

Background/Purpose:

There are a number of factors that influence a low-income student’s ability and opportunity to attend college. What if both the ability and the opportunity are present but the student still chooses to attend a less selective postsecondary institution? This phenomenon is known as “college undermatch”.  Undermatch occurs when a student’s academic ability allows them to attend a college or university that is more selective than the one they ultimately choose. Undermatch may perpetuate educational and economic inequality and inhibit social mobility for low-income students.

Much of the present literature fails to differentiate between academic undermatch and substantial academic undermatch (i.e., undermatching that spans at least two college selectivity categories). For example, a student who attends the University of Michigan rather than Stanford would “undermatch” but a student who attends a community college rather than Stanford would “substantially undermatch.” Thus far, there has been little research conducted on current, nationally-representative data on the predictors of low-income students who substantially undermatch compared to students who match or undermatch.

Method:

We utilized the High School Longitudinal Study (HSLS 2009) which contains high-achieving, low-income 1,380 students (i.e., student have at least a cumulative 3.5 grade point average and family falls at/below 185% of the federal poverty level). The study provides information on participants’ educational experiences, sociodemographic background, and college preparation and choice. This study conducted logistic regression (with Holm-Bonferonni corrections; weighted) to investigate several potential reasons why a high-achieving student would substantially undermatch (reference group, n = 263) compared to students who match (n = 648) and undermatch (n = 94).

Results:

Compared to the students who matched, students in the substantial undermatch group were more likely to be female (b = .34, p < .05) and English language learners (b = 1.13, p < .05), and less likely to have a parent with a college degree (b = -.80, p < .001). Substantial undermatch students were less likely to believe they could qualify for financial aid based on their academic achievement compared to match (b = -1.16, p < .001) and undermatch (b = -1.57, p < .01) students. Compared to match students, substantial undermatch students were more likely to believe they could not afford an in-state four-year college (b = -.43, p < .05). Substantial undermatch students were less likely to cite reputation (b = -.90, p < .001) and more likely to cite being close to home (b = .49, p < .05) as reasons to choose a college compared to students who matched.

Conclusions/Implications:

This study is one of the first to consider reasons why low-income students would substantially undermatch. One of the main reasons appears to be concern for college cost. Students who substantially undermatch do not appear to prioritize a college’s reputation and may want to stay close to home. Interestingly, this study did not find many differences between undermatch and substantial undermatch students. College choice is complex and there are a number of other reasons, not investigated here, that could provide useful insight.