About the Study


The Wisconsin Scholars Longitudinal Study is the first statewide longitudinal study of the impact of private need-based financial aid on college persistence and graduation. The Fund for Wisconsin Scholars (FFWS), through a generous donation of $175 million from John and Tashia Morgridge, awarded 1,200 grants to low-income college students attending the 42 University of Wisconsin System and Wisconsin Technical College System institutions. Those chosen will receive $3,500 per year if they attend a 4-year college, and $1,800 per year if they attend a 2-year college. Each recipient will be eligible to continue receiving the grant for up to 10 semesters. Federal aid rules often require reductions in federal aid when private aid is awarded; therefore, some students will receive the grant in the form of reduced loans and others in the form of cash. A new group of recipients will be selected every fall.

The main purpose of the FFWS grant is to increase college completion among low-income students. Similar to many other states, Wisconsin has experienced declines in the college continuation rates of its high school graduates over the last 15 years, while rates of bachelor’s degree completion have remained relatively steady, with slightly more than half of entering college students finishing a degree. But rates of degree completion are much lower among students from lower-income families. Wisconsin is widely considered a “low tuition” state, yet like most states around the nation its public colleges and universities have witnessed regular increases in tuition that have been unmatched by increases in state support for financial aid. As a result, like 49 other states, Wisconsin was recently given an “F” in affordability by a national report card. On the whole, Wisconsin is a fairly typical state in its struggles to improve educational attainment and close achievement gaps while confronting declines in state support and affordability.

Past research suggests that financial aid can promote college success, however the policy making community needs to know much more about the amount and kind of aid that increases completion the most, for which students, and under what circumstances. In this study we will therefore answer four main policy-relevant questions: (a) what are the impacts of private grants on college persistence and graduation?; (b) how do the impacts vary according to the way in which the grant is distributed to students (reduced loans versus increased cash)?; (c) how do the impacts vary by student and college characteristics?; and (d) how and why do the impacts arise? We will address these questions using a longitudinal, mixed-methods analysis and creating, for the first time, a large-scale comprehensive database that integrates student financial aid data, college transcripts, surveys, and in-depth interviews with students. Our findings will have direct implications for both governmental and private grant programs and provide an important basis for addressing current debates about the future of financial aid in the state and nationwide.

What Makes WSLS Unique

Contribution #1: Gold-Standard Evidence.

Financial aid is received by the neediest students, those who are already less likely to finish college. So it’s often hard to know whether differences in graduation rates among those who get aid and those who don’t are attributable to aid—or something else. The WSLS is a randomized experiment of aid, where grants are given via random assignment. This means that any observed differences in outcomes can be directly, and appropriately, attributed to the effects of financial aid.

Contribution #2: Implementation at Scale.

While a few experiments of financial aid have been conducted in the past, they usually include a few hundred students at a small number of colleges. This makes it hard to identify effects on target populations.

The WSLS includes 6,000 students attending 42 colleges and universities across the entire state of Wisconsin. Four-year flagships, comprehensives, two-year open-access and two-year technical colleges are all included. Student participants reside in urban, suburban, exurban, and rural areas and comes from a wide range of family backgrounds. This will enable WSLS to produce evidence on differential impacts of aid for students of color, by college type, etc.

Contribution #3: High External Validity.

Experiments are usually conducted by first recruiting participants, and then randomly assigning them to treatment and control conditions. This helps ensure a high level of internal validity, but since the effects are based on a group that volunteered to participate, it’s often difficult to extrapolate to the larger population and take the intervention to scale. This is especially difficult if one is interested in understanding differential effects of financial aid.

The WSLS has high levels of both internal and external validity since potential participants were identified using administrative data. This means they needed to take no action of their own to receive the treatment (additional aid), and thus are like other aid recipients who need only to complete the FAFSA to be eligible. Policymakers can therefore learn much more about aid’s effects and extrapolate to other populations using WSLS data.

Contribution #4: Need-based Criteria.

All experimental evaluations of aid to date are of merit-based or performance-based programs. This is part of a widespread and unfortunate trend in aid that prioritizes academic merit over financial need in the distribution of resources.

The financial aid given in the WSLS is strictly need-based. Like the largest aid program, the federal Pell, it is based solely on expected family contribution, and does not include any merit or performance incentives that could restrict access for disadvantaged populations. Since it is distributed to existing Pell grant recipients, the aid tested in the WSLS is essentially a test of doubling the size of the Pell.

Contribution #5: Targeted at College Completion.

Most financial aid research focuses on inducing students to attend college. But large inequalities in college completion persist, and the ability of aid to increase graduation rates is unknown. Studies tend to estimate only the total effect of aid (effect on access + effect on completion) which means that policymakers do not have needed information on the most effective timing of awards.

The WSLS is given to college freshman in their first semester. As such it is targeted at altering the prospects of completing college, for those who already choose to attend. Turning more entrants into graduates is a specific goal of the program.

Contribution #6: Serving the Traditional Student.

The MDRC Opening Doors experiment focuses on whether aid can increase achievement among older students. While that is a group deserving of more attention, it represents a small fraction of college students nationwide. After decades of investment in financial aid for traditional-aged college students, we still do not know if it works. It’s about time. The WSLS provides money to first-time, full-time students who start college within 3 years of high school graduation. Therefore, these are students in the midst of a transition to adulthood, between the ages of 18-22, who are most likely to accrue immediate and long-term benefits from a college degree.

Contribution #7: Comparing and Measuring Effectiveness.

Most studies of aid identify effects of a single intervention—most commonly a scholarship program. But they do not compare the effects of different forms of that intervention, which could help policymakers decide what will work best for their state.

This study involves a private grant which will reach some students in the form of cash, and others in the form of loan reduction. We will produce comparisons of the relative effectiveness of each, and calculate the cost effectiveness of both approaches.

Contribution #8: Knowing Why Aid Works.

Research on financial aid is typically limited in its ability to explain how and why effects occur. The WSLS employs a longitudinal mixed-method approach to the study of financial aid. Led by a team of sociologists, economists, and psychologists, the data comes from administrative sources, surveys, and in-depth interviews. Each estimation of an effect will be buttressed by an explanation of how and why it occurs.

Contribution #9: Long-Term Commitment.

Without long-term commitment to provision of the grants being studied, long-term outcomes cannot be observed. WSLS is studying a program created by a $175 million endowment from John and Tashia Morgridge. As long as their recipients are in school, they will continue to receive the grant.

Contribution #10: Enduring Effects.

The relationship between income, education, and lifetime earnings, health, and longevity are poorly understood. But in a time of scarce resources, our investments need to be targeted in ways that can pay off in multiple ways over time. Typical studies of financial aid only look at very short-term outcomes, usual a year or two of college persistence.

The WSLS focuses first and foremost on degree completion, and secondly on identifying the multitude of ways in which college pays off. We intend to track participants for a minimum of 10 years, and hopefully much longer. Just as the first Wisconsin Longitudinal Study of 1957 generated some of the most important information about the role of education in social mobility at a time when college-going was not the norm, the WSLS has the potential to provide needed information about the effectiveness of higher education interventions in an era when policymakers are looking for ways to make college accessible to all.