.png)
Ever since self-appointed watchdogs from Elon Musk’s Department of Government Efficiency began slashing jobs and contracts at the U.S. Department of Education in February, Mark Schneider has served as a valuable touchstone, helping put the radical budget and programmatic changes in context.
But while some of the cuts are, in his words, “dumb,” and show a lack of experience among the cost-cutters, Schneider has also pushed against many critics’ assertions that the Trump administration will effectively destroy the agency. In his view, the cuts offer an opportunity “to clean out the attic” of old, dusty policies and revitalize essential research functions. Those include the National Assessment of Educational Progress, which he maintains has lost its way and grown prohibitively expensive while in some cases duplicating the work of independent researchers.
A conservative who has held top roles in both of the last two Republican administrations, as well as the most recent Democratic one, he’s the ultimate education insider — Schneider’s conversations often invoke an alphabet soup of government agencies, contractors and think tanks. Yet he’s unusually candid about his time in government, especially now that he is no longer there.
Get stories like this delivered straight to your inbox. Sign up for The 74 Newsletter
A political scientist by training, Schneider has spent nearly two decades in education research. He served three years as commissioner of the National Center for Education Statistics under President George W. Bush, then spent a decade as a vice president at the non-profit American Institutes for Research. He returned to government in 2018, appointed by President Trump to lead the Institute of Education Sciences, and stayed on until 2024 under President Biden.
Through it all, he has remained an independent voice even while in office, telling The 74 in 2023, for instance, that the reason Biden hadn’t fired him along with other Trump appointees was that education research wasn’t considered important enough for the president to bother.
Over the course of six years at IES, he tried — mostly unsuccessfully, he admits — to reform the department into “a modern science and statistics agency.” He’s honest about his limitations, saying he “tried really hard to modernize the place” without much success. While Musk and his cost-cutters last month took a chainsaw to IES, he observes, when he led it, “I didn’t even have a scalpel. I had a dull butter knife.”
While many education advocates are decrying Trump’s bid to eliminate the Education Department, Schneider has said carefully breaking it up could actually produce “a more efficient, dynamic and responsive school system — all things the Department of Education has been hard-pressed to do.”
Schneider sat down for a wide-ranging conversation last week with The 74’s Greg Toppo. They discussed the difficulties of reforming what he considers a hidebound agency, the opportunities of starting over and what the future might hold for NAEP, also known as the Nation’s Report Card.
Is Trump Gutting Ed Research a New Beginning or Just ‘Slashing & Burning’?
Now a non-resident senior fellow at the conservative American Enterprise Institute, he is cagey when pressed about returning to IES in a second Trump term. Schneider notes that the deep cuts have left no actual agency to run. “Who wants to go in there and head a 20-person unit?” But leading a revamped IES, he admits, would present “a wonderful challenge.”
At the end of the day, though, he says it remains an open question whether the next step in the Trump administration’s plans is rebuilding or neglect.
The conversation has been edited for length and clarity.
The 74: I wanted to start with something you said about Musk’s crew on a podcast a couple of weeks ago — namely that given your work to reform IES, you were “a little envious” they could “do it all in one day.” Let’s drill down. Is that what you had in mind when you were there? Or did this go further?
Schneider: [Laughs.] Let me try to figure out the best way of putting this. The issue, of course, is that IES was a 23-year-old institution. A lot of stuff got locked in. None of this is surprising: Institutions get locked down and they keep doing the same thing over and over again. I tried hard to change things. It was almost impossible. I tried to get Congress to create ARPA Ed — NCADE [National Center for Advanced Development in Education, an agency to develop and scale innovative, cutting-edge practices and tools]. A lot of people worked really hard, but we were never able to get it through Congress.
A DARPA for K–12? Omnibus Bill Includes Substantial New Funds for Education R&D
Whether Congress will ever do anything is a different question. But the fact of the matter is that even though people were in favor of it and we had a lot of political support, we still couldn’t get it across the finish line. Well, now the National Center for Education Research doesn’t exist anymore. There’s one person left there. So whether or not this is naive, we don’t need NCADE anymore. We should rebuild NCER to look like ARPA Ed. We don’t need any legislation for that, because it’s in the purview of the director to do that. That’s an amazing opportunity. We can just create a modern research funding organization with no need for congressional action.
My colleague Kevin Mahnken recently talked to Doug Harris from Tulane. He said IES is “being knocked over by these cuts.” I think beneath a lot of this is people like Doug worried that this administration is simply anti-science. It sounds like you are saying the opposite. Should this give people like Doug a little hope?
As of right now, we have no indication except every once in a while some words bubble out: “Oh, we are going to rebuild IES. We are looking for a future direction for IES.” But there is no concrete plan. So the proof is going to happen in the next several months. If the department says, “Yes, we are redoing NCER, we are redoing NCES,” which are the two biggest units that are most in need of repair, and they announce plans to rebuild them in a modern way, then we’re fine. But if nothing ever happens and we end up with three people at NCES and one person at NCER, we’ve got a problem.
IES Director Mark Schneider on Education Research and the Future of Schools
You have no sense one way or another?
You probably hear the same things I do. I have no concrete information about any plans to rebuild. I hear rumors. But until concrete plans are announced and actions are undertaken, then we should maintain a healthy skepticism. That said, I still believe that if this administration wants to modernize IES, they have an opportunity that no one’s ever had before — since 2002.
I have no concrete information about any plans to rebuild. I hear rumors. But until concrete plans are announced and actions are undertaken, then we should maintain a healthy skepticism.
Since they created it.
Congress created IES in 2002, and it was a brand new, innovative organization that radically changed the way education research was done. Well, 23 years later, that opportunity repeats itself. So my hope, and maybe this is naive, is that we grab that opportunity. We know a lot more about education research. We know a lot more about modern statistical data collections — and we learned a lot from NCES. For example, the lack of timeliness hurt them endlessly. So now it’s like an open field. Let’s build a better edifice now than what we had a year ago.
Let’s talk about the nuts and bolts of this vision. One of the first shocks to the system we got was in February when DOGE canceled all those federal contracts. And one of the hardest hit was AIR, where you spent 10 years. I wonder to what extent your views have been shaped by being an insider there. Is this a sector that needs a shock to the system?
There are at least two parts of that question. So there’s the question of the quality of the work. AIR does good work. I don’t think there’s any question. The big contract houses are capable of doing quite good work. However, people are really pissed off about the overhead rate [also known as indirect costs covering expenses] that these contract houses and universities charge. The overhead rate is just too high. When I was at IES after the pandemic, AIR got rid of their building on Thomas Jefferson St. [in affluent Georgetown] — a huge, expensive building. Many, many people ended up working at home, and the rest went with much less — and much cheaper — office space.
So after AIR shed that big office space. I called up CAM [Contracts and Acquisition Management], the contracts management people in the department, and I said, “You know, the overhead is based on many factors, but office space and parking are major components. They’ve reduced the cost of their offices — they didn’t get rid of them, but they downsized and went to cheaper places. So let’s renegotiate their overhead rate.” [He imitates CAM officers]: “No, no, no, no, no, no.” I couldn’t get CAM to even consider reducing the overhead rate to reflect the lower cost.
Just to be clear: You couldn’t get the department to lower the cost?
The department had to reopen the negotiations. And they would not. I’m not sure what the right word is. It’s laziness, corruption. This was wrong. Why were we paying such high overhead rates when their costs went down? And you know as well as I do that many, many, many organizations got rid of office space and reduced their costs. So why wasn’t CAM renegotiating overhead rates? I never was able to get a good answer.
But that’s a department problem, not an AIR or Mathematica problem.
Yes and no. Clearly the department had within its authority to reopen those negotiations. But the problem, of course, is that the agencies end up getting captured by the people they’re supposed to regulate, the people they’re supposed to monitor. [Test developer] ETS and NAEP are an even clearer example: How much money went to ETS to do things that they weren’t capable of doing?
Such as?
Such as building the platform for NAEP. The software. This is not firsthand, so it could be hearsay, but the people from DOGE looked at the platform that ETS had built for NAEP, and they just said, “What is this? This is not the way modern software is built.” And I believe that’s because we used ETS, which is a testing company, not a tech company, to build the platform. And again, that has to do with the overly close relationship between NCES and ETS.
‘Going for Blood’: After Mass Firing, Many Wonder How Ed Dept. Will Function
This is not just an Ed and contractor issue. We know this exists in other places besides education: the close relationship between the contractors and the agencies that are supposed to be supervising them. This is a well-known problem. The companies capture the government agency. [Editors Note: Asked to respond to Schneider’s comments, Christine Betaneli, an associate vice president at ETS, released the following statement Tuesday: “ETS delivers nearly 50 million tests every year across the US and around the world on robust technology platforms. We have consistently delivered innovations on NAEP suited to the specific requests of NCES. We’re incredibly proud of the unmatched quality we have provided to the American people in supporting the Nation’s Report Card. We will continue to innovate on behalf of America’s teachers, parents and children who rely on this critical data to improve access to quality education nationwide.”]
How do you prevent that from happening in the next iteration of this department? Is it just by bringing in totally new people? Is it by changing the contours of the contracts? Is it by doing things totally differently?
There are a couple of things. First of all, there’s a serious cultural issue. That’s clear. I will tell you another story, and this will give you some more depth to how bad this can be. When I first showed up at IES, we brought in [consulting firm] McKinsey & Co. to do an analysis of how to improve the efficiency and effectiveness of the organization. They went around and interviewed people — staffers, program officers — to try to get some idea of what was going on. And remember, this is an outside consulting company we hired. They interviewed one of the program officers who said to this outside consultant, “I’m never giving up this contract. You will have to pry it out of my dead hands.” I mean, that’s a stupid thing to say, but it’s also illegal. This is a long-term project officer who admitted to an outside person that they had been totally captured, totally in bed with the contract shop. That a veteran project officer would say this to an outside consultant says there’s something really, really wrong.
This exists in other places besides education: the close relationship between the contractors and the agencies that are supposed to be supervising them. The companies capture the government agency.
So the culture is a problem. Is the scope of what the department does a problem as well?
Yeah.
Is a breakup necessary to change the culture?
Do we need to break up IES and move all these pieces around? If the goal is to shrink the department, or make the department go away, then we have to find homes for these activities. But when I wrote that last summer, I was not envisioning the disappearance of 90% of the workforce.
I believe if we don’t get congressional approval to end the department, it’s going to be around. But I keep thinking about both NCES and NCER, the two largest units, and there’s now an open field. I’ve always had problems with NCES. As a major federal statistical agency, like many other federal statistical agencies, they just kept falling further and further behind. But we can now imagine, we can actually execute, rebuilding NCES as a modern, lean and mean statistical agency.
After Gutting Staff, Education Department Remains Quiet on Future of NAEP Test
For example, the state longitudinal data systems. I’ve written about a different vision of how to build that. The TUDAs [Trial Urban District Assessments, NAEP tests given in 27 urban school systems] are incredibly expensive. Nobody can tell me how much they cost. I’ve asked many, many times how much they are, and the fact of the matter is, we don’t need them anymore, because we have other ways of getting estimates for these large cities. I’m talking about Tom Kane and Sean Reardon [who have developed an Education Recovery Scorecard detailing achievement nationwide]. They compute the exact same statistics that TUDA does.
So that actually leads me to my next question: What is your vision for something like NAEP? Can a lot of it go away?
The Political War Over the Department of Education is Only Beginning
For me, the most important thing about NAEP is the state-by-state comparisons. They’re important because governors hold the keys to so much education reform, and they care about the comparisons. When NAEP came out several months ago, Louisiana, Mississippi, Florida, they all cared about these things. Virginia had a 50-minute release event with the governor, [Virginia Education Secretary] Aimee Guidera and the Superintendent of Schools [Lisa Coons]. That’s three of the heaviest hitters in the world of education, all lined up, talking about the importance of the state-by-state comparisons and what they were doing to address Virginia’s terrible performance on NAEP.
That alone is an amazing demonstration of the power of the state-by-state comparisons. Do we need a $185 million-a-year NAEP to generate the trend line and the state-by-state comparisons?
Part of the reason we don't have to keep doing this — I mean, it's a sad thing to say — is because the people in NCES who were committed to this are gone.
I’ll give you another example: There are at least three different sub-domains in NAEP math. I’ve never seen anybody talk about those sub-scores. How much does that cost? Why do we keep doing it? We just need to bring some sanity to what we’ve built over 50 years that have grown up over time, the cost of those things, the backwardness of many of those things, and say, “Hey, we don’t have to keep doing this.” And part of the reason we don’t have to keep doing this — I mean, it’s a sad thing to say — is because the people in NCES who were committed to this are gone.
I don’t want to leave that point without addressing institutional memory and knowledge. A lot of the people who are gone know how these things work. Getting rid of those people might have changed the culture, but it also might have hollowed out the agency’s ability to get the next NAEP report out. Does that keep you up at night?
Clearly, that’s the horns of a dilemma. But where is the time, where’s the energy, where are the people to rethink this stuff? Part of the problem was that there was not sufficient rethinking. The machine worked. It got out on time. Many problems were solved by just raising more money. I attended NAGB [National Assessment Governing Board] meetings for 10 years. Every time there was a budget presentation, inevitably, the budget was in the red. And so then we have to cut this, and we have to cut that. And it was never like, “What is it that we need to preserve, instead of going to Congress and asking for another $30 million and getting $10 million?” That wasn’t the thought process.
NAEP lost its leading edge. The demands of running the operation are real. But if you never stop to think about what you're doing, then you're going to end up behind.
So I went to OMB [the Office of Management and Budget], and asked them to take the appropriation that Congress gave for NAEP and put 10% aside in a separate fund for R&D. I asked Congress to do this, and then OMB, because there was no commitment by the leadership of NCES and NAEP to spend that kind of money on R&D. Instead, it was always, “We need this money for the operations. We need money to do this other task.” As a result, NAEP lost its leading edge. The demands of running the operation are real. But if you never stop to think about what you’re doing, then you’re going to end up behind.
But to many, the way these agencies were trimmed doesn’t seem any smarter. There was a lot of cutting with a chainsaw rather than a scalpel. My sense is that’s going to require a great deal of work just to bring back basic functions. Am I right?
The years that I was at IES, I didn’t even have a scalpel. I had a dull butter knife. There are so many quotes about this: “Breaking things is easy, rebuilding things is hard.” From Hamilton: “Winning is easy, governing is hard.” All of that is true. And it’s just so much easier to just say, “No, no, no, no, no,” than to start rebuilding. Mancur Olson, a brilliant economist, wrote a book called The Rise and Decline of Nations. He had his finger right on the pulse. Over time, what happens is that you start accumulating all these lobbyists and all these interest groups and all this stasis, and it just builds up and builds up and nothing can get done, because you end up with this incredible superstructure of groups and people who are totally vested in the status quo. And he says, every once in a while, you just need to just blow the shit up and rebuild.
If the Education Department Can’t Be Closed, at Least Fix It by Breaking It Up
There’s a lot of concern about the rebuilding. How do you calm people’s fears that there is no rebuilding coming?
Look, there’s nothing we can do right now except wait and lobby Congress and the department that the rebuild is important. And I hope they know that the rebuild is important. Again, you can just give things away: Give NCES to BLS [the Bureau of Labor and Statistics] etc., but some of that stuff is going to require congressional action. Good luck on that. In some ways, again, we have an open field. Let’s take the opportunity to build that back in a much more modern, efficient way.
It seems like a lot of people, especially on the right, are talking in terms of the department reaching its sell-by date. But if you can change the culture and remove the barnacles, or whatever you want to call them, what’s the point of breaking it up?
As they say, that’s above my pay grade. [Laughs.] I’m writing a whole series of papers about what can be done and I think they’re all reasonable and in the realm of the possible. I have not had any contact with anybody in the department about any of my visions or plans. But there are ways to rebuild this so it looks like a modern science and statistics agency.
Would you like to lead it at some point?
If there’s anything there. [Laughs.] There’s nothing there. What do we have, 20 people left in that whole organization? NAGB is moving into LBJ [the Lyndon Baines Johnson Building, the department’s headquarters in Washington, D.C.] because it’s pretty much empty. And I assume IES will also end up moving into LBJ, but the fact of the matter is all of these places are ghost towns now.
I don't know what happened in between ‘You must be back in the office five days a week,’ and ‘By the way, you're fired.’ I know from friends it was not a fun place to be.
Before the pandemic, we were trying to argue, incredibly, about creating 10 more desks for the growth of IES. And then, of course, after the pandemic, there was nobody left. Nobody came back to the office. The place was pretty much empty already. And then the executive order had everybody coming back to the office — and then everybody got fired. I don’t know what happened in between “You must be back in the office five days a week,” and “By the way, you’re fired.” I don’t know what it was like. I know from friends it was not a fun place to be.
If I’m reading between the lines correctly here, it sounds like you’d like to lead IES, but you’d like someone to rebuild it first.
First of all, I’m not answering that question. Many people have asked me if I would go back. But I have a lovely life. I live an eight-minute walk from AEI. AEI is a very generous organization. They’ve never said no to any reasonable request I’ve had, so it’s extremely pleasant, extremely easy. But I spent six years at IES. I tried really hard to modernize the place. For someone who wants to create the next version of IES, there are incredible challenges, but the rewards of doing it would be amazing. But they’ve already eviscerated the unit. Who wants to go in there and head a 20-person unit? But if there is a taste to rebuild IES to look like a modern organization, that’s a wonderful challenge.
Could what you’re describing just as easily be done privately?
A lot of people are talking about that: How can philanthropy stand up and take over the role that IES used to have? Even the biggest foundations don’t have the kind of money IES had. IES spent over $100 million a year supporting education research, just from NCER. There’s no foundation that has that kind of money, and I’m not even sure if there’s a coalition of foundations that could come up with anywhere near that kind of money for research.
So there’s an indispensable role for research funded by IES or the Department of Education or some part of the federal government. But the return on that investment was not sufficient. I don’t know if part of the suspicion of IES was just a gut-level reaction to “Too big, too big, too big,” and how much of it was, “Hey, we have spent all these billions of dollars over the last 20 years and what have we got to show? We have declining NAEP scores. We don’t have any evidence of increasing achievement, etc, etc.” I’m not sure if the antagonism towards education research was because it wasn’t working or because it’s just that we were anti-science. I truly don’t know.
I don't know if part of the suspicion of IES was just a gut-level reaction to ‘Too big, too big, too big,’ and how much of it was, ‘Hey, we have spent all these billions of dollars over the last 20 years and what have we got to show?'
Years ago, we looked at how many math interventions have any evidence of success. It turned out to be about 15%. This was a very depressing number, until you start looking around at what the success rate is in any science: 10% of clinical trials work, 90% fail. And then of course, we’re learning that even among the 10% that work, there’s an incredible amount of dishonesty, lying, cheating. The “replication crisis” — we’ve glommed onto that term — says there’s a lot of stuff going on in the sciences that are not kosher. So at one level, the antipathy towards the Department of Education is, “This is not a function that the federal government should be involved in. This is all state and local.” O.K., I got that, and I believe a lot of that is true. But even in the most extreme form of federalism, there is a role for government support of research. There is a government role for statistics. And the question then is: How do we focus that to help states and local governments, parents, teachers, students achieve more? A lot of what happened was that that tight focal point just disappeared.
If I’m translating what you’re saying correctly, you don’t know how we got to this point in terms of the mechanisms of the cuts, and you don’t know what people were thinking. But in a way, you’re saying it’s not really important, because we needed to get this done.
We needed to get this done. A lot of what was done was incredibly important and was needed. There’s no question about it. But we’re going to come back to the same theme over and over again: For six years I had a butter knife, and then these guys show up in a day with a chainsaw and they cleared out all the detritus and all the underbrush. But what do we do now? That’s what matters.
English (US) ·