At a time when policy makers are faced with budget constraints, the idea of tying financial aid to desirable outcomes has a lot of surface appeal. But a new study -- one of a series being conducted to test the concept -- shows the limits of the approach.
The study,  published by the research group MDRC and part of a larger project  financed by the Bill & Melinda Gates Foundation and other sources, explores the effects that grants tied to measures of enrollment, persistence and academic performance have on low-income students at two New York City community colleges. Unlike some of the other studies in the series, which provide students with extra student services and other enhancements as well as scholarships, the New York study provided grants alone (on top of the students' federal and state need-based grants) to students at Borough of Manhattan Community College and Hostos Community College, both in the City University of New York System.
Based on their enrollment and completion levels, study participants received grants of up to $1,300 for each of two semesters, and some received a similar-sized grant for a third (summer) semester. In each term, a student received $200 for registering for six or more credits, another $450 if still enrolled by the middle of the term, and $650 for achieving a grade of C or better (or the equivalent in developmental courses) in at least six credits. (The control group, of course, got no grants at all.)
The grants did appear to have some impact.
Students who received the performance-based grants were likelier than those who did not to enroll full time during the semesters they received the aid. And the students who received aid for the summer were more likely to enroll in that term, and earned more credits, than did those who did not receive grants for that term.
But by and large the effects did not last. The study tracked students for a full additional year beyond the year in which they received the performance-based grants, and by the end of that period, the average grant recipient had neither earned any additional credits nor registered for more semesters. "This suggests that while the program was effective when students were eligible for scholarships, the effects on enrollment and credits earned dissipated after the program ended," the authors write. (The relatively small number of grant recipients at Hostos, a much smaller institution where the students are older and the program was housed in a student services division of the college, did accumulate more credits than their peers did, the authors note.)
Reshma Patel, project and data manager for the Performance-Based Scholarship Demonstration Project, suggested that the New York project might be showing the least results of the various sites in the larger study because of its "bare bones" nature, since participants received only grant funds and no other support. (Studies at other sites are testing the provision of advising and tutoring services, the power of longer terms and larger amounts of scholarships, etc.) While results for most of the other sites have not yet been published, she said, "we have had consistent findings across the sites, in terms of improvements in credit accumulation."
Optimistic as Patel is that the studies, taken together, will ultimately show benefits of tying aid to performance in some fashion, she made clear that the project's intent is not to influence debate over how federal aid such as Pell Grants should be awarded.
"These grants are going to an already Pell-eligible student population, but students with high levels of unmet need," Patel said. "These results don't speak to [possible changes in Pell Grants]. A lot of community foundations and other providers are looking for ways to make scholarship dollars that right now are not very purposeful a little more purposeful. This could be a viable solution to repurpose those dollars."