"This research is not partisan"

What are the long-term returns on public schooling? Does transitional kindergarten lead to better academic outcomes? Why don't eligible college students take advantage of valuable financial aid programs? These questions and others help us understand what works—and why—in education.

Many of these questions went unanswered until the Michigan Education Data Center (MEDC) was established in 2018 by the Ford School's Education Policy Initiative (EPI). 

MEDC is one of only a handful of state data systems nationwide that democratize data. By providing secure, equitable access to 20 years of insights into Michigan's learner —from birth through postsecondary education—MEDC enables researchers and policymakers to answer critical questions that improve student outcomes. 

This is possible only through a partnership among the Michigan Departments of Education; Lifelong Education, Advancement, and Potential; and the Center for Educational Performance and Information, along with Michigan State University's Education Policy Innovation Collaboration.

Filling the federal data gap

For decades, national research and policy decisions relied on comprehensive federal data. But with significant cuts to the U.S. Department of Education's data capacity, researchers and policymakers must now look to states for the insights needed to design and improve educational programs. "This type of research is not partisan or political," says Kevin Stange, Ford School professor and EPI co-director. 

We want to help children succeed, so we want to know what works. If a program helps students, we want to know that. If it doesn't, we also want to know that—because we don't want to spend money on a big program that is not effective."

Kevin Stange

Stange notes that the state's priority is to run educational programs, and they don't have capacity to carry out the in-depth analyses needed to evaluate them.

State data powers decision-making

The work that MEDC and a handful of initiatives in other states are doing is more critical than ever for driving evidence-based education policy. Findings from MEDC data can now be found in more than 150 publicly available reports from researchers across the United States—and have shaped education policy in important ways.

One example involves transitional kindergarten, a state-funded program providing an extra year of early education for children who need additional time to develop social, emotional, and academic skills. Over half the school districts in Michigan now offer the program known as Young 5s. Research using MEDC data found the program expands access, significantly improves kindergarten readiness, and leads to better math scores through third grade.

At the other end of the education continuum, MEDC data has enabled analysis of Michigan's largest state-run financial aid program, the Tuition Incentive Program (TIP), for community college students who received Medicaid at some point.

Stange's research found that only 14% of eligible high school graduates and 29% of eligible community college students use the program, largely because many don't realize they qualify. These findings prompted the state agency to increase efforts to educate students and schools about TIP.

Researchers also explore outcomes beyond the classroom. Using MEDC data, Jason Baron, an assistant professor of economics at Duke University, illustrates the return on public school investments. He has found that students in schools receiving extra funding at the elementary level are substantially less likely to be arrested in adulthood; for every $1 invested in elementary schools, Michigan saved more than $2 in criminal justice costs. His research also shows that investing in teachers and reducing class sizes is more effective than investing in infrastructure.

Baron, who has worked closely with Ford School professor Brian Jacob since his postdoctoral fellowship at U-M, notes the strength of EPI's partnership with the state. "I have found Michigan's state leaders to be incredibly responsive to research and empirical work. We often work with them to think about, 'what do they want to know?' These partnerships are the path to really good research."

Michigan's blueprint

MEDC's impact is inspiring efforts nationwide. At major education and data conferences, EPI is regularly called upon to share its approach: strong university-government collaborations, aligning research with state priorities, and translating findings into real-world action. 

Four states—Michigan, North Carolina, Texas, and Tennessee—have operated education data centers for more than 10 years, says Nicole Wagner Lam (MPP '10), associate director of EPI. All four received State Longitudinal Data System grants from the U.S. Department of Education to build the data infrastructure.

"Without these grants, none of the centers would have been able to do this work," Lam says, noting local data partnerships exist in New York City and Los Angeles as well. 

As more states recognize the need for robust data infrastructure, EPI's approach serves as a blueprint for future collaborations. EPI is a driving force in creating a national network of education research practitioners to fill the federal data gap. 

"Our goal is to build a specialized learning community for education leaders who want to strengthen their research-practice partnership's data infrastructure. We aim to share best practices and create a repository of resources to support them," Lam explains. "Looking ahead, the absence of federal data has made state and local partnerships more important than ever." 

We're helping states, districts, and universities across the country see what's possible—and how their own efforts can lead directly to better outcomes for students."

By Rebecca Cohen (MPP '09)

CLOSUP turns 25 STEM-ming the gap  Could new transmission lines lower your electricity bill? 

← Read all stories in the spring 2026 edition State & Hill