Liz Cohen and Kevin Huffman make the case for tutoring
Hi folks. I hope you have fun and/or meaningful plans for the long weekend. Speaking of which, we’ll publish SCHOOLED on Wednesday instead of Tuesday next week because of the holiday.
Today, Liz Cohen and Kevin Huffman push back on my assertion that tutoring is too hard to get right. And Matt Yglesias pushes back on the argument (from me, David Brooks, and others) that red states are kicking blue states’ butts when it comes to education reform. Plus the week’s other great takes from Morgan Polikoff, Andy Rotherham, Dan Buck, Mike McShane, and many more.
Sign up to receive this newsletter in your inbox on Tuesday and Friday mornings. SCHOOLED is free, but a few linked articles may be paywalled by other publications.
On Monday I teed up a debate on tutoring, drawing from Liz Cohen’s great new book on the subject. Among other points, I argued that tutoring is “too hard to get right, too focused on fixing instructional problems after they’ve happened, and too little bang for too much buck”—especially compared to other reforms, such as boosting pay for teachers in high-need schools. Today we’ll hear responses from Liz, as well as Kevin Huffman; next week it will be Matt Kraft’s, Dan Goldhaber’s, and Matt Barnum’s turn. I have to admit: They all make a pretty strong argument that I’m being too pessimistic about tutoring’s potential.
Liz Cohen
As I said on the podcast we recorded earlier this week, I’m not actually interested in whether all kids get tutoring. What I’ve always cared most about is whether all kids are learning and how we can improve public education systems to improve student outcomes, especially for low-income kids. I continue to believe tutoring is an important tool in this work. Here are three big reasons I’m sticking with tutoring as a solution that should be on the table for states, districts, and schools in 2026 and beyond, even in the midst of very real resource constraints.
1. Schools and school systems need a strategy if they are going to truly change. Conveniently, tutoring works best when part of a strategy.
The districts and states doing the best work with tutoring are the ones using tutoring strategically as a way to support specific groups of students who are performing below grade level.
The District of Columbia is a great example of a place where teacher salaries are quite high, and where tutoring is being used as part of a strategic overhaul of math instruction that includes new curriculum, teacher training, and more. Two years ago, students in D.C.’s math tutoring program gained an average of two months of learning over 15–20 weeks of tutoring. But last year, students in the program gained three months of learning. D.C. is improving their tutoring and improving their math instruction, all while paying teachers quite well.
I would love to see a school system where no student needs additional support. In the real world, at least today, that’s not the case. So if districts or states are seeking to answer the question, “How do we support students below grade level with an approach that has the highest likelihood of success while also focusing on supporting teachers to use great materials so that kids actually learn?” tutoring is a good answer. In Baltimore City Public Schools, 72 percent of kindergarten students and 64 percent of first grade students who get tutored moved at least one level up on the early literacy assessment used by the district. In Guilford County, North Carolina, tutored students outpaced the state average for growth in kindergarten and first grade. These kinds of results seem like they would also help teachers as more students in the classroom are truly prepared for grade-level work. That might help teachers like their jobs more, regardless of pay.
2. Tutoring is a back-to-basics play.
School, especially the early grades, should be focused on reading and math, with a side dish of writing and general social skills. Tutoring gets kids more time on the exact core academics that matter most. Kids benefit from more time on task, more time reading aloud to a trusted adult, or more time practicing tricky math skills. Teachers, even amazing highly-paid teachers, can’t give 30 kids the individual opportunity to showcase their practice and get specific feedback, especially in a classroom where half or more of kids are below grade-level. Pair this with the knowledge that demonstrating competence is key to building motivation, and we begin to make the case that tutoring is a back-to-basics academic effort with the added benefit of increasing student motivation towards learning.
The book from 2025 that has most influenced my thinking about education isn’t a policy or political book, but a book called Instructional Illusions by three learning scientists. In remarkably concise, clear writing they explain exactly why learning doesn’t happen in so many classrooms. One takeaway I had from that book is that activities like quality tutoring—that require the student to actually do something real with immediate feedback from a trusted adult—helps avoid many of the illusions that plague classroom settings, like the engagement illusion that lets teachers believe busy students are learning or the discovery illusion that lets teachers believe children need to discover answers for themselves.
3. Tutoring is a wonderful opportunity for growing civil service opportunities and creating an American culture of learning.
From college students looking for part-time jobs to older Americans looking to spend their time meaningfully, tutoring is an opportunity to engage many more Americans in education. We’re seeing this play out. Hundreds of thousands of Americans across all ages have raised their hands to work as tutors or mentors in the past five years.
In turn, that might help create more of a culture of learning that our country sorely needs. Education shouldn’t purely be transactional. It’s not all about the degree. America is better off if young people care about learning, about exploring the world and all its possibilities, about finding out what excites them and motivates them. So the more adults, both young and old, play a role in a child’s education, the more a child learns that learning is itself a valuable part of life, one that remains meaningful throughout adulthood. Is this wishful thinking? Absolutely! I wish it nonetheless.
Kevin Huffman
I’ve been deep in the tutoring space for four years now. Accelerate funds promising programs, research, and policy development in an effort to scale practices that work. We have tried to approach the research with true neutrality, and have not been cheerleaders in the space. After four years, funding more than two dozen randomized control trials and dozens of other studies and analyses, and working deeply in 11 states, here is what I can say with great certainty:
High-dosage tutoring helps students gain a few extra months of learning in core subject areas when the students receive the sessions they are supposed to receive.
What I find heartening in the space is that all of the things that you delineate—from Liz’s excellent book—as critical questions are all: 1) resolvable and 2) peripheral to the core question of dosage. Should we ensure tutors are well selected and trained? Yes. Should we align the sessions with core curricula? Yes. But, across a wide range of domains, what we find again and again is that tutoring advances outcomes when it is delivered a sufficient number of times. The other pieces are basically just determinants of the degree of magnitude.
The challenge that we keep seeing—which Jill Barshay covered in Hechinger, analyzing the University of Chicago’s large scale study—is that districts and schools repeatedly fail to deliver the sessions. Then you wind up with very modest impacts.
That’s a failure of implementation and accountability. I would argue that tutoring doesn’t fall into the category of “it’s too hard so we just can’t scale it.” It’s really not that hard to have tracking systems and accountability for delivery. As with many of the walk-backs in education over the last decade, we just often decide that it’s too hard to do things like say, “if you get money to do this thing, then we require proof that you did the thing.”
I could go on, but to me, high-dosage tutoring is the embodiment of the truly absurd public conversation we have in the ed policy space that goes something like this:
Here’s a thing that works!
Hmmm, we have to do some stuff to actually implement it.
[A bunch of places take the money but just don’t do it.]
Wow—it doesn’t look like the thing worked.
Look—shiny thing over there!
I’ve lived through this in much tougher bare-knuckle arenas like teacher evaluation. (Hey, I wonder what two places—in the hundreds and hundreds of data points—had the largest NAEP gains in a cycle ever? Oh, the two places that leaned in on teacher evaluation? I’m sure it’s a coincidence.)
By comparison, tutoring is a layup. People like it (it polls higher among parents than any other education policy). Its evidence base is incredibly strong. Implementation can focus on one simple baseline proposition—ensuring kids get their sessions. And it builds on, rather than replacing, existing instruction.
We make things way too hard sometimes. Let’s get kids to school, ensure they have excellent instructional materials, train teachers on how to use them, have accountability for ensuring students actually advance on track, ensure any student that falls behind grade level immediately begins to receive tutoring, and track and report all of the above. I promise you: It will work.
A year ago, I argued that blue states are falling behind their red peers when it comes to education reform—on policies including charter schools, accountability, and the science of reading. David Brooks, among others, picked up on the argument, saying that red states were “kicking the butts” of blue ones. And Karen Vaites and others got folks to pay attention to the Southern Surge, showcasing the recent progress in Mississippi, Louisiana, Alabama, and Tennessee.
Enter uber-wonk Matt Yglesias with a strong argument that “schools are getting worse in most red states.” He’s not wrong, of course, about the basic NAEP trends; outside of the four Southern Surge states (plus Matt’s current hometown of Washington, D.C.) there’s not much good news to see anywhere. (Though the (mostly red) states that opened schools sooner during Covid did appear to suffer less learning loss.)
But I’m personally doubtful that the “Southern Sag,” as he called the trends outside of the Surging Four, is going to last forever. Take Texas, for example. Yes, the Lone Star State’s test scores have been in the dumps like most everyone else’s, but with a slew of policy reforms in recent years—boosting funding, taking over Houston’s schools, reforming teacher pay—I’d be more willing to bet on achievement gains there than anywhere else.
Nor do I buy his argument that Republicans have given up on accountability in favor of school choice. Most red states, as I argued last year, are doing both. Though Matt is sadly correct that “schools don’t really compete on the basis of being highly effective at teaching students.” That’s true in higher ed and it’s true in K–12, too.
Still, kudos to Matt for urging Democrats to embrace education reform, including across the Potomac River:An interesting test case is coming down the line in Virginia. The state has a newly elected Democratic governor, Abigail Spanberger, who has been cultivating a moderate brand. A new Democratic trifecta is obviously not going to voucherize the Virginia public-school system or cut school spending to give everyone a tax break. What they ought to do is what you would expect Democrats to do with a public-school system: try to make it better so as to deliver better results, especially for high-need kids.
The federal government is no longer trying to make states hold schools accountable for performance, but state governments are still allowed to do this. And as it happens, Glenn Youngkin’s outgoing administration brought in two former Obama Education Department hands (including friend of the newsletter Chad Aldeman) to recommend changes to the state’s school accountability system, recommendations that were released in December while Youngkin was a lame duck.
Going forward with these recommendations would help new Governor Abigail Spanberger deliver some real results on education quality and build her résumé for national leadership.Yes!
Jason Riley joins the Mississippi Miracle fun. He runs through the major points of last weekend’s New York Times article on the Magnolia State, noting in particular the Times’s reporting that Mississippi “was able to muscle through some changes, in part, because it has weak teachers unions, which have traditionally resisted accountability linked to standardized testing.” And though his piece came out before Yglesias’s, I doubt he’d agree with Matt’s assertion that most red states are doing just as poorly as blue ones. As Jason concludes: “too many states are hampered by teachers unions, which pursue their own interests at children’s expense, and by an education establishment that thinks its social crusades are more important than basic skills.”
I’ve been arguing that the most important question for our movement in coming years is whether any Democratic presidential candidates run on education reform. Jorge Elorza is excited that Rahm Emanuel appears to be doing exactly that. Jorge writes:
At a recent town hall, Emanuel argued for an outcomes-driven agenda centered on extended school days, phonics-based literacy, and stronger guardrails on technology in the classroom. He’s making the case in places like Mississippi, pointing to the substantial attainment gains there as evidence that decline isn’t inevitable.
It’s still very early, and whether he ultimately runs in 2028 or not, I’ll continue to applaud Emanuel (and any Democrat willing to stake out a substantive lane on education) for forcing a conversation many in the party have long avoided. We’ll also keep making the case that the issue should be as salient for any aspiring policymaker as it is for students and families.Check out a bunch of smart predictions for 2026 from Andy Rotherham. He too is eager to see if education becomes a big issue in the 2028 presidential election, which will start in earnest after November’s mid-terms. Other things to watch for: (1) a lot of “variance,” in Andy’s words, from state to state—for example, in how they respond to the federal tax credit initiative in the One Big Beautiful Bill; (2) clashes over ed-tech and civic education; and (3) more pain as districts deal with their school enrollment crises. “It’ll be a challenging year,” Andy concludes. “Wishing you well.” (!!!)
Morgan Polikoff, meanwhile, bemoans the move away from college admissions testing and toward rampant grade inflation in high schools, which led to the USCD scandals and has hollowed out standards nationwide. Morgan wants a renewed commitment to testing—and better assessments—plus writes that:
Grades also need to mean something again. This involves rolling back reforms that lower expectations for students. And in the context of AI, it also entails helping teachers figure out how to assess student learning in ways that are trustworthy and valid. Without these kinds of fundamental changes, neither classroom teachers nor universities will be able to make the kinds of decisions that will help ensure student success.
Speaking of assessments, AEI demographer Nicholas Eberstadt is skeptical of claims from PISA that China’s students are number one in the world in student achievement. After running through numerous anomalies in the data, Nick writes:
We are inclined to believe that the Confucian tradition places a premium on study and academic excellence. We are prepared to believe that overall academic achievement could thus appear more favorable for China than our own statistical analysis based only on a country’s socioeconomic indicators might predict. (That fits Vietnam.) But some of the results strain credulity. And it does not help that they are unverifiable.
Back here at home, Dan Buck bashes the “portrait of a graduate” movement:
The idea is simple: High school graduates should possess a set of essential skills beyond mere academics, decided by state education agencies or school districts. These portraits offer standards-setting frameworks that expand beyond traditional academic skills and knowledge to include affective competencies such as collaboration, critical thinking, social awareness, and “being an active citizen.”
The problem, Dan explains, is that:
These frameworks extend the scope of traditional standards from academic content and skills to personality traits, fundamental beliefs, and patterns of thought. Portraits of a Graduate change the question facing policymakers from “What should we teach our children?” to “What kind of people do we want to create?” In so doing, they extend the goals of schooling from traditional academics to values and character traits. That represents a seismic shift.
If so:
They deserve a level of scrutiny and debate that would make arguments over Common Core look like playground scuffles. More realistically, they’ll remain so vague and insipid that they create a permission structure for schools to prioritize most everything except academics and to excuse themselves when they fail at their responsibility to teach even basic literacy and numeracy.
Mike McShane writes up a new paper by Dany Shakeel, Misty Gallo, and Patrick Wolf that looks at recent achievement trends for public, charter, Catholic, and DoD schools. Perhaps the most interesting finding is that low achievers have done worse in Catholic schools, too, over the past decade or so. Which might indicate that phones and social media, rather than public policies, are to blame for the “bottom falling out” for public schools, too.
Mark Schneider digs into the latest version of the NEED Act, which would create a DARPA-like agency for education. He’s all for it but offers some constructive suggestions for making the proposal even stronger. Among them: Require IES to adopt best practices in contracting, and work to modernize the National Center for Education Research, perhaps along the lines of “the recently announced National Science Foundation Tech Labs initiative.” Of course, as Mark notes, none of this matters if the Trump OMB succeeds in killing off IES entirely.
“Just like during Great Recession-era job struggle, Gen Z graduates are pouring into education as Teach For America reports a 43 percent surge in new teachers.” —Emma Burleigh, Fortune
Future teachers at the University of Maine are debating the science of reading instead of learning the science of reading, according to Kristian Moravec of The Maine Monitor:
On a Wednesday night in November, more than a dozen UMaine students huddled in opposite corners of a classroom in Orono, writing pros and cons on giant Post-it notes to stick on the wall.
They were debating the use of books that encourage students to use context clues such as pictures to learn how to read—called leveled readers—versus books that build on phonics skills, called decodable texts.
Their task: to understand and argue which method of literacy instruction is better, simulating real-life debates happening within the “reading wars,” as one student described it.
More elementary schools are departmentalizing by subject, a model traditionally reserved for older kids. “We want to make sure that our teachers are experts in what they’re teaching so that they can do a good job of teaching it to the kids,” a Louisiana administrator tells Ariel Gilreath, who looks at the upsides—particularly teacher strengths and preferences—and where the model falls short. —The Hechinger Report
See ya next week,
Mike