Memo: Retention Statistics Deserve Nuanced Analysis

A January 2009 article in the Chronicle of Higher Education quoted a higher-education marketing consultant as saying, “We know statistically if a student can barely afford college as a freshman, they almost never come back as a sophomore.”

This statement seemed reductionist to us, based on the retention research the firm has has conducted for clients over the years. We went back to look at the relationship we found between financial aid and retention, and came up with the following results.

It is true that high need and low grant aid (high gap) predict low retention when considered independently from other variables. If this is the definition of “barely able to afford college,” then the statement is supported by our data as well, at least at face value.

However, high gap is often highly correlated to a number of other factors that also predict low retention, particularly academic preparedness. Typically, the highly-gapped students are the least prepared academically (otherwise they would be getting merit awards to meet some of that gap), and often apply later in the cycle, when need-based awarding is less generous. Both of these factors are known predictors of poor retention, and when included in a regression equation, usually prove to be more significant than any variables related to grant or gap.

In laymen’s terms, this result suggests that there are other factors that predict both high gap and poor retention, and that the poor retention of highly-gapped students is a correlation rather than a cause of low retention. Of course there are individual cases where affordability really is the problem, but in aggregate, for the clients we have studied, affordability alone does not appear to be the driving force in the attrition of highly gapped students.

The notable exception is when students earning a low GPA (in the bottom one third to one half of the freshman class) in their first semester are considered independently. In several of the institutions we have studied, high gap is a significant predictor of attrition for this cohort

These are the students where the problem is really value; they are struggling to afford an education that is not helping them to succeed. For high-achieving students, however, our results generally suggest that even families who are likely struggling to afford college are willing to absorb the cost because the results are worth it.