Is Steubenville Cheating?

An Ohio investigation exposes a bigger problem: We don’t know what to do with successful school systems.

Last week, I wrote about Steubenville, a small Ohio school district that defies the laws of educational gravity.

Imagine my surprise when the Ohio Department of Education and Workforce announced that its testing vendor had found “significant anomalies” in 2026 results from all six Steubenville schools.¹ All six. An investigation is now underway.

Well, then. How about that?

To be clear, this is a flag for odd patterns, not a finding that Steubenville cheated. We still don’t know much, including how many student tests were involved or what triggered the concern. The district has said only that it is fully cooperating.

But it’s an awkward development for someone like myself who just spilled 2,400 words celebrating Steubenville’s success.

From here, there are many possible outcomes. Steubenville could be cleared. The anomalies might be innocent and/or trivial. In that case, we’ll be more certain that these decades of superlative performance were real. Alternatively, the investigation could find misdeeds, raising the possibility that Steubenville has been a fraud, and we’ll need to figure out how it went on so long without being uncovered.

Whatever the outcome, this scrutiny is overdue. Why did we have to wait for a testing vendor to identify irregularities before taking a closer look at the highest-performing low-income district in the United States?²

It suggests that we have a much broader problem than whatever did - or didn’t - happen in Steubenville. Stick around, we’ll come back to that.

Did Steubenville Do It?

I realize you are dying to know. Me too.

Public data can’t tell us whether a district broke the rules. The state has access to information we do not, including student response patterns. Let’s let them investigate. But we have always known that Steubenville is imperfect and that some of its trends are unusual. Here are just a few examples from 2025, the most recent year available:

• Its incredible elementary reading performance erodes by eighth grade. While 99.0 percent of Steubenville third graders were proficient, just 49.5 percent of eighth graders were, which is actually below the Ohio average. Math does not have the same issue - 85.1 percent of eighth graders were proficient compared with 47.6 percent statewide. Why?

• Pass rates on high school end-of-course exams are mixed. Steubenville was only slightly above the state average in government, 74.5 to 72.6 percent, and biology, 67.9 to 65.4 percent. But it was exceptional in Algebra I, U.S. history and geometry, with proficiency rates of 93.6, 91.0 and 89.8 percent, respectively. Why?

• ACT scores are not very impressive. Just eight percent of Steubenville seniors were classified as remediation-free, which requires a minimum score of 18 in English, 22 in reading, and 22 in math. The state district median is 13.8 percent. Given the very high performance of Steubenville students in younger grades, why aren’t more of them scoring higher on the college entrance exam?

These data points are not proof of shadiness - far from it. But they are patterns that should have been investigated before anyone, including me, became confident about Steubenville’s success.

If it turns out that Steubenville cheated, it would be a profound betrayal resulting in theft of services from children who were told they performed better than they really did, and therefore were denied supports they deserved. It would break my heart, honestly.

There are several reasons I remain skeptical that Steubenville is cooking the books. For one thing, the district has been consistently exceptional. The graph below shows Steubenville’s percentile among Ohio districts on the state’s annual performance index. (The gap in the line is COVID.) Going back to 2016, it has never been lower than the 85th percentile or higher than the 95th. Cheating schemes often produce wild fluctuations as perpetrators clumsily twist various dials in search of their desired target.³ If Steubenville is guilty, it has the dexterous touch of a cat burglar.

Next, all six Steubenville schools were flagged. We would need to believe that a cheating conspiracy can be maintained across multiple buildings without any educators, students, or parents crying foul. Not easy.

And finally, there’s an unclear motive. Nobody appears to be getting rich or famous from Steubenville’s reputation by going on the keynote speaking circuit. Why would the district mortgage its integrity for a scheme that is the polar opposite of the ethos visitors have observed? This is a place where staff members drive to a child’s house to check on him when he doesn’t show up. And then they’re going to falsify that student’s test results?

I want the results to hold up. We need hope. We need examples we can learn from. But time will tell. Steubenville, a nation turns its lonely eyes to you.

Excellence in Obscurity

Now, the larger problem: we have no idea what to do when school systems achieve outstanding results.

If Steubenville were a corporation in a market like big box retail or air travel and had lapped its competitors for decades, consultants would have crawled over hot glass shards to know its secrets.⁴

But in education policy circles, Steubenville is almost invisible. I can’t find a single peer-reviewed academic paper about Steubenville or its methods. Zero.

May I remind you that we have peer-reviewed papers asking whether kindergarten classrooms have too many decorations and whether using red ink causes teachers to grade more harshly?⁵

That’s not to say nobody noticed. Karin Chenoweth has been urging us to pay attention to Steubenville and similar districts for decades. She wrote a whole book called Schools that Succeed. Her approach was to start with the data, identify systems producing surprising results, then go there and investigate. But she is a reporter, not an institution with the authority and data access to verify the results independently.

What happened when she told us what she learned? Not much. Because our system doesn’t know what to do with that type of information.

We’ve Built a Failure Machine

American education is quite good at finding outliers. Unfortunately, only the ones at the bottom.

We have a bulky apparatus for comparison, which we use to rank schools. We place struggling districts on unpleasant lists. They submit improvement plans. This song and dance, repeated year after year, does nothing to advance our understanding of how to succeed.⁶

On the flip side, there’s no serious national effort to identify exceptional high-poverty districts.⁷ No independent process to verify results. No team dispatched to spend six months on the ground figuring out what local educators are doing differently. No follow-up five years later to see if success survived the arrival of a new superintendent or, say, a global respiratory pandemic.

Academic research isn’t much help, either. It typically asks whether intervention X produces outcome Y. Great, we’ve always wanted to know. But Steubenville has put up amazing results over an extended period of time. How? It couldn’t be a single cause.

What Would a Real Verification Process Look Like?

If we’re going to pay more attention to successful districts - and verify that they are real - someone needs to do the legwork. Who?

This is where the federal government should step up. The Department of Education has been cut to a fraction of its former size. I promise, I’m not here to re-litigate that issue. But a smaller federal role shouldn’t become an excuse for indifference about what’s happening in 13,000 districts nationwide.

The feds can help us see what’s happening - and more importantly, they have the resources to tell us why. This is the sort of thing agencies already do in other fields like housing and agriculture. We basically know why crop prices rise or fall. Why can’t we know why kids are struggling to read?

We can start by finding the oddballs.

Every family has weirdos.

Use Stanford’s EOP data set, NAEP results, state assessments and other sources to flag systems producing improbable results. Look not only for high average scores but rapid growth, tremendous subgroup performance or sustained excellence over time.

Then, most importantly, verify the results.

Review everything - demographics, enrollment, student mobility, testing participation, accommodations. Look at scale scores, not just proficiency rates. Compare performance against independent data and subsequent outcomes. Where results seem improbable, dig into things like testing times, answer changes, item patterns, and proctor assignments. Make sure it’s not an educational Ponzi scheme, a la Atlanta.

If only part of the story holds up, which part? A district could have strong instruction and sloppy test administration. It could produce genuine success, but only in some grades. Sainthood and fraud aren’t the only possibilities.

If it all checks out, dissect the operation from top to bottom. We need to understand curricula, union contracts, schedules, staffing, attendance, leadership development, and budgets. Interview the 20-year veterans who know everything. What makes the system work?

And finally, return over time. Some systems look amazing for a brief period of time before petering out. Others become true dynasties. We want the dynasties.

There might be dozens or hundreds of Steubenvilles worth understanding. Students in North Palos, Illinois are performing nearly four grade levels above the national average. This district has been improving rapidly for the past decade while the rest of the country has moved in the opposite direction. What’s going on? I have no idea. But I’d like to know.

Below, look at North Palos compared to some of the best-known suburbs in Chicagoland. None of them has a low-income rate higher than 20 percent. Sixty-three percent of North Palos students are low-income. And yet it has the highest achievement - in some cases by a huge margin. A compelling candidate for a deep dive, right?

You may have your own nominees for districts that deserve a closer look. Please add them in the comments. We’ll get a list going.

Don’t Foment Another Finland Mania

Even when we confirm the performance of oddballs, let’s hold off on canonizing them. We tried that with Finland.

After it hit the top of the international rankings in the early 2000s, we lurched from “this little nation is producing great results!” to “these must be the policies producing them, and every breathing wonk should take a junket to Helsinki and come home preaching the good news.” A complicated system (and history) became a too-tidy story about teacher preparation, trust, low-stakes testing and social supports. It was embarrassing when Finland proceeded to decline rapidly.⁸

Lesson learned. Positive outliers should lead to hypotheses, not clumsy prescriptions. We skipped a few steps.

In the real world, some successful systems will regress to the mean. Others will depend on fleeting gimmicks. Fine. We learn something by parsing that out, too.

I don’t know where the Steubenville investigation will end. Perhaps the district has a persuasive explanation that puts the whole thing to bed. Maybe the state will identify limited violations that do not call the long-term record into question. Or the findings could force a more painful reassessment and a “Say it ain’t so, Joe” moment.

No matter what happens, I know we need to search for success, not failure. When we find it, we should study the hell out of it.

And you shouldn’t need to hear about it 25 years down the road from a doofus with a leaf blower.

1. Shoutout to Ohio State’s Vlad Kogan for being the first to share the news with me. What exactly are testing anomalies? This article from Signal Cleveland lists four possibilities that were described in letters to districts:

• Answers that are changed from wrong to right at a higher-than-average rate.

• Student answers that don’t line up with their ability – for example, a low-performing student gets a lot of hard questions right.

• Student scores that are significantly higher than last year.

• Abnormally long or short test-taking times.

2. The state wasn’t even looking at Steubenville originally. After investigating a set of self-reported anomalies in Columbus, the state directed its vendor to do a broader review.

3. Examples from Philadelphia and Atlanta.

4. Toyota is a great illustration. MIT commissioned a five-year study of its manufacturing techniques in the 1980s. By the early 2000s, so many competitors had tried to follow the Toyota recipe that a new cottage industry arose asking why its success was so difficult to replicate. Harvard had a long-running obsession with Southwest Airlines.

5. We can go all day on this thread. Second-grade desk arrangements. Ok I have to stop. I’m sorry.

6. Evaluations of Obama-era investments suggest these approaches aren’t always useful in helping low-performing schools get better, either.

7. There is something called The National ESEA Distinguished Schools Program, which was created in 1996 as the National Title I Distinguished Schools Program. It is now operated by the National Association of ESEA State Program Administrators rather than directly by the U.S. Department of Education. Not an inspiring effort. State agencies may select up to two federally funded schools each year to be honored for exceptional performance/growth, closing achievement gaps, or service to special populations. Many states do not even participate - just 63 schools were recognized in 2025, down from 100 a decade ago. Tennessee and Virginia submitted schools that have selective admission processes, which can’t possibly be the spirit of the award. The public announcement is riddled with typos, e.g. the town of Fort Defiance, AZ is called “Fort Difiance” and an Everett, MA honoree is listed as “Webster School School.” This is our starting point. Nowhere to go but up. On the bright side, Ohio submitted a Steubenville school and its name is spelled right.

8. Finland’s long slide has continued, unfortunately. Chad Aldeman did a great piece on the release of PISA 2025 results last week.