The Department of Government Efficiency (DOGE) has told the research arm of the Department of Education to cancel contracts, close centers, eliminate laboratories, cut research grants, and ask staff to resign. Plaintiffs have filed lawsuits seeking to halt the orders. As the legal battle progresses through the courts, it is worth considering whether the Institute of Education Sciences (IES) is a perpetrator of waste and abuse or a valuable research organ.

To shed light on the question, I counted the share of the most important education studies since 2002 that received IES funding. To see if funded proposals had a political bias, I also looked at the ideological coloration of their findings.

Answers in brief: IES funded four of the top 11 studies (36 percent); the ideological coloration of funded proposals: light red.

You would not guess these facts from the rhetoric employed on both sides of the current debate. DOGE says its wasteful contracts have allocated $4.6 million “to coordinate zoom and in-person meetings,” $3 million “to write a report that showed that prior reports were not utilized by schools,” and $1.4 million “to physically observe mailing and clerical operations.” Mark Schneider, the just retired IES director concedes that “far too many . . . projects seem designed to advance the careers of academics who publish obscure findings in obscure journals rather than support student achievement.”

Another former IES official, Thomas Weko, instead finds the cuts “shocking” and “pointless.” An anonymous insider condemns them as a “destruction of knowing what works for kids.” Adam Gamoran, president of the William T. Grant Foundation and Biden nominee for IES director, says it’s wasteful to eliminate research that can make schools more efficient and effective. He points to the “science of reading” as a powerful, new instructional approach inspired by IES research and dissemination.

The debate is what prompted me to look deeper into how many major studies are indebted to IES for financial support and whether they have a detectable bias.

My judgments draw upon a consensus report Nora Gordon and I wrote for the Hoover Institution entitled What We Know about Our Schools, which was released last October. Nora is an economist at Georgetown University who leans moderately to the left; I am a political scientist with a more conservative bent. We found ourselves pretty much in agreement about “what we know.” Our judgments about what we know were based mainly upon research of an experimental or quasi-experimental nature. We focused on U.S. education from pre-school to high school, not the higher educational system. We did not include in our analysis the considerable literature on curriculum (such as the science of reading), as neither of us is a subject-matter specialist.

I searched through our report to find the eleven studies that seemed to be the most important and influential published since 2002, when IES opened its doors (see sidebar). I then looked to see which studies included an acknowledgement of financial support from IES or from a center it sponsored.

Four of the 11 research projects—36 percent—acknowledged IES support. Of the remaining seven, five received fiscal backing from private foundations or from another governmental agency; two do not include any mention of external financial support.

The four IES-funded studies appear to be free of progressive or other ideological bias, though many findings receive more applause in conservative quarters than liberal ones. For example, conservatives welcomed the finding by Durkin and his colleagues that state-funded pre-school education has an adverse effect on a child’s social and educational outcomes by 6th grade. Anstreicher and colleagues report that school desegregation yielded positive effects on student learning in the South but not in other parts of the country. Angrist and colleagues found positive effects on longer-term educational outcomes of charter schools in Boston. Chen and Harris find that higher enrollments in charter schools benefit all students district-wide. Dee and Wyckoff find performance pay lifted student achievement in Washington, D.C.

Studies that did not receive IES support were slightly more likely to generate findings liberals applaud. Reardon reports that the socioeconomic divide in achievement widened dramatically during the last half of the 20th century. Jackson finds a positive effect of additional school spending on the long-term economic outcomes of students from low-income families, a result widely used by teacher organizations and liberal advocacy groups in state school finance cases. Woessmann and Hanushek tell us that student achievement in adolescence is correlated with a country’s rate of economic growth. Chetty and colleagues show that a child who has an effective rather than an ineffective teacher in elementary school will enjoy more years of education and higher pay when they become an adult. Dee and Jacob find gains in math, but not in reading, from the accountability system established by the federal law, No Child Left Behind. Hartney reports that teacher unions have long been heavily subsidized by federal, state, and local governments.

Considered altogether, the findings from the non-IES studies are diverse but somewhat more likely to fit well with the agenda of left-leaning advocacy groups than those funded by IES. Even so, the number of cases is too small to draw definite conclusions. What is clear is that there is no evidence that IES selects out progressive studies for funding.

I draw three conclusions. First, IES has a history of supporting high-quality research projects that have had a substantial impact on what we know about our schools. Second, education research will survive even if IES resources are substantially reduced, as private and other governmental programs provide adequate sources of financial support.

Third, and most important, IES should continue the bulk of its data collection and longitudinal surveys, such as the National Assessment of Educational Progress. Nearly every one of the eleven major studies draw upon IES surveys or its other data collection activities. IES data remain an essential source of basic information about the operation of K–12 and pre-school programs.

Collecting information on the state of American education was the first task given to the Office of Education when it was established in 1867. It remains IES’s most important job. Just as the Commerce Department gathers information on the state of the U.S. economy and the Bureau of the Census tracks demographic trends, so IES tells us what is happening in schools. Americans need to know that public school enrollments are falling, that chronic absenteeism is now rampant in public schools, that the per pupil cost of education is on the rise, and that learning tanked when schools closed during the pandemic. None of this evidence would be as irrefutable had we not a national data-collection system.

Unfortunately, some of DOGE’s proposed cuts are aimed at IES’s data collection function. That mistake needs to be corrected by Linda McMahon, the 13th Secretary of Education. Above all, she must protect the Department of Education’s information-gathering capacity.

Paul E. Peterson is the Henry Lee Shattuck Professor of Government and director of the Program on Education Policy and Governance at Harvard University and a senior fellow at the Hoover Institution at Stanford University. He welcomes your reactions to this post by email at paul.peterson@educationnext.org; responses will be curated and shared periodically. His Education Exchange podcast is available with a new episode each Monday.