Imagine There’s No Politics: A Review of the Northern Report

In the fall of 2017, over 50 education researchers gathered to discuss their projects authorized under the auspices of a Regional Education Lab (REL) funded by the Institute of Education Sciences (IES). The convener interrupted the proceedings by stating, “Congress is considering a student privacy bill so draconian that it would render education research impossible. Who has good contacts with Republicans on Capitol Hill?” I, rather sheepishly, raised my hand. No other hand went up. I thought to myself, IES has a big problem.

Flash forward to late February of 2026, when the report Reimagining the Institute of Education Sciences: A Strategy for Relevance and Renewal was released at 4:59 pm. on a Friday. The report was authored by Amber M. Northern, on leave from the Thomas B. Fordham Institute to serve temporarily as a senior advisor to Education Secretary Linda McMahon. Northern’s job was to propose a rebuild of IES after the Department of Government Efficiency, in the first few months of the second Trump Administration, cut the IES staff by 88 percent and canceled just about every IES contract with an exit proviso. President Trump’s budget proposal for the next fiscal year, released two months after Northern’s report, would cut IES funding by 67 percent, mostly in research projects.

Like an emergency room physician performing triage after a mass casualty event, Northern consulted widely and moved energetically from program to program, determining what should and could be salvaged, reformed, or repurposed. I thank her for her yeoman’s service.

In the interest of full disclosure, I have worked with and for Northern and consider her a friend. I have also worked with and for IES in various capacities, for nearly its entire existence. I support the rigorous scientific evaluation of K–12 education interventions and the collection and posting of complete and accurate descriptive data on students, teachers, and schools. I have made a successful career out of doing so. I was among the dozens of experts consulted by Northern during the thorough evidence-gathering stage of her project. I think I am an informed judge of where IES has been, where it should go, and what it should be.

Northern’s core diagnosis is that IES is afflicted with “[a]n outdated research infrastructure and organization that limits quick insights, coordination across data sets, and innovative, non-traditional research models.” To cure IES, Northern recommends six “Big Shifts” that involve:

1. changing its scattershot approach to instead focus resources on a few big, urgent problems
2. standing up a single “super panel” to generate longitudinal survey data on multiple topics in place of the nearly dozen smaller panels, each focused on particular questions and subpopulations and proudly sporting its own obscure acronym
3. encouraging states with shared regional interests in an education research topic to join in Tocquevillian voluntaristic groups to petition IES to study their thing
4. directing “the focus of the research work towards practicality, innovation, and relevance”
5. creating a “research hub” to render the 10 RELs more responsive, timely, and better coordinated
6. re-orienting the What Works Clearinghouse from its current function as an arbiter of which studies are scientifically rigorous and which interventions are “evidence-based” to a producer of highly accessible guides to sound educational practice

Northern further highlights the “bedrocks of IES”: its political independence, scientific integrity, statistical data-gathering infrastructure, and sponsorship of rigorous research across diverse settings. The bulk of the 84-page report reviews a parade of horribles, organized by the divisions of IES: the National Center for Education Statistics, the National Center for Education Research, the National Center for Education Evaluation and Regional Assistance, and the National Center for Special Education Research. For each problem Northern highlights, she helpfully offers a variety of potential solutions.

There is much to applaud in Northern’s report. The Shift 2 suggestion of a “super panel” and the Shift 6 re-orientation of the much-maligned What Works Clearinghouse (WWC)—often sneeringly called “the nothing works clearinghouse,” make sense. Still, WWC would offer practice guides on only a modest number of topics, if compelling evidence were required to justify such directives. If not, if weak bodies of evidence sufficed to justify a practice guide, then the guides would seem to be no better than encouraging educators to go with their gut. This is just one of the many tradeoffs inherent in Northern’s recommendations.

Currently, there are a few big questions in K–12 education policy, involving chronic absenteeism, early and adolescent literacy, civic knowledge and skills, school choice, and the promise and perils of artificial intelligence in the classroom. It would be beneficial for the federal government’s education sciences agency to focus its limited resources on those big questions, as proposed in Shift 1. But Shift 3, allowing states to band together and petition for studies on the education issues that concern them, has the potential to undermine the laser focus of Shift 1, given inevitable dissensus among states regarding the education issues that require urgent attention.

A key question that is, shall we say, “finessed” in Northern’s report is who, ultimately, decides what IES funds? She describes a process whereby policymakers in individual states request studies of questions salient to them. What happens when that process yields 15–20 different topics? Who narrows that list down to the five or six focus areas called for by Shift 1? I suspect political imperatives would require that IES address all the questions requested by the states, thereby continuing the scattershot approach Northern rightfully bemoans.

Directing research towards a north star of “practicality, innovation, and relevance” (Shift 4) undoubtedly will come at some cost to scientific rigor, IES’s previous north star. These two priorities inevitably need to be balanced. Rigorous findings that are irrelevant are, well, irrelevant, but relevant findings that are not grounded in solid evidence can be downright harmful, as we have seen with the tremendous damage done by the education field’s uncritical embrace of the “balanced literacy” and “three cuing” approaches to (not) teaching literacy.

It is not clear how the research hub (Shift 5) would discipline the RELS to be timelier and more responsive, especially since the improved coordination also called for from the RELs slows down production processes. Innovation and AI are offered as solutions to much that ails IES, though the federal government isn’t known for being especially innovative, and the vast promise of AI remains unproven.