Why Do School Districts Protect the Test Pile?

We need to explore the resistance to test reduction

For nearly a decade, we at the Center for Assessment have given school districts the same advice about over-testing: Take an inventory of the assessments you’re giving, name the purpose of each test, find the redundancies, and then let go of what isn’t earning its place.

What we don’t write about is what usually happens next. The review ends. The team works its way through a spreadsheet with forty rows (one for every test). A test or two gets trimmed. And the big lineup, for the most part, survives.

The resistance to dropping tests isn’t universal; our work with two California districts, Bakersfield and Coachella Valley, shows how an audit can help districts focus and streamline their testing programs. I’ve argued for strategic abandonment of commercial interims. The guidebook we wrote with the National Academy of Education walks districts through the process. And we aren’t the only organization that has created inventory tools and processes and works with states and districts on this.

Education First’s recent research puts numbers on the “test pile”—students in high-testing districts take as many as 88 standardized assessments before high school, roughly seven per grade, with no statistically significant relationship between testing volume and proficiency.

None of that is news to anyone who has sat through a district assessment review. In a landmark study in 2015, the Council of the Great City Schools identified district-required tests as significant contributors to the overall testing burden on students and teachers. What’s been nagging at me is why the review, done well, with good people, so rarely leads to subtraction.

I realize I’ve been treating this as an information problem: If districts just have good information, they’ll eliminate the redundant tests. But it doesn’t always work that way.

People Resist Subtraction

In 2021, a team led by Leidy Klotz at the University of Virginia published a study in Nature with a straightforward title: “People systematically overlook subtractive changes.” Klotz’s team gave participants a Lego structure to stabilize, a grid pattern to make symmetrical, an essay to improve, an overstuffed travel itinerary to fix. In every case, most people added something. Removing a piece was often the better solution and almost always the cheaper one, but most participants never considered it.

The important word is considered. Subtraction wasn’t weighed and rejected; it never came up as an option. When the researchers cued people—“you can add or remove pieces”—subtraction rates jumped. In other words, the practice of simply cueing people helped people to subtract.  

Klotz’s book, Subtract: The Untapped Science of Less, extends this to institutions, and anyone who works in a district will recognize the picture he paints. Organizations have many processes for adopting things and almost none for removing them. Adding a program is visible and makes an organization’s work appear to have strategic forward momentum and progress. Removing or de-implementing programs can make it look like nothing happened, or, worse, that something went wrong.

Auditing Assessments as the Cue, Not the Cure

Here is where this discussion gets uncomfortable for me. An assessment audit is precisely the intervention Klotz’s research would suggest. It puts subtraction on the table. It says, you are allowed to remove pieces. If over-testing were mainly a failure to consider subtraction, the audit should fix it.

But in my experience, the audit alone doesn’t fix the problem of too many tests. Districts complete the inventory, see the redundancy, and still keep most of the tests. That tells me other constraints and challenges are at play. Our tools are well built to identify redundancy, misuse, underuse, or misalignment. But they don’t seem to help districts subtract in meaningful ways.

Suspending Tests During COVID: What We Learned

If you want evidence that information doesn’t equal action, look at the COVID pandemic, which forcibly subtracted from our heap of tests. In spring 2020, states waived their required tests and most districts suspended their interims and benchmarks. It was an involuntary, system-wide experiment in less. Klotz tells a similar story about San Francisco’s Embarcadero Freeway, which collapsed in the 1989 Loma Prieta earthquake; Only afterward did the city discover it didn’t need the freeway.

COVID gave districts the same chance to discover what they didn’t need. And within a year, the mound of tests was back, with a fresh layer of learning-loss diagnostics on top. We saw less, and we restored more. Whatever is holding the tests in place, it’s stronger than the experience of doing without them.

Naming the Barriers to Test Reduction

I’m working on a paper that dives into what’s actually getting in the way of subtracting. But here’s one suggestion for district and school leaders in the meantime.

The next time an assessment review ends with a spreadsheet that highlights recommended cuts in your test lineup, and little real change is made, recognize that you’ve got a barrier that’s getting in the way of translating the information you sought—because you wisely recognized a problem—into meaningful action.

Something stopped the cuts and changes. Recognizing that fact, and understanding that it means the work isn’t done, can lead you toward a deeper exploration of what is getting in the way. The audit is just the first step in the process.

For those of us who build audit tools, the lesson is harder. We’ve been handing districts a very good cue and calling it a solution. The next thing we design must reach past the inventory into the decision itself. If you’re a district leader who has tried to eliminate a test and found you couldn’t, I’d like to hear what got in the way. Email me at cevans@nciea.org.