Education savings accounts, tax-credit scholarships, vouchers, charter schools, home schooling, tutoring, course choice, dual degrees, and microschools are transforming K–12. In “Talking Choice,” Ashley Berner and I try to make sense of the shifting landscape. Ashley directs Johns Hopkins’ Institute for Education Policy and is a leading authority on “educational pluralism.” Whatever your take on educational choice, we seek to foster a more constructive conversation about what it means for students, families, and educators. Today, we discuss whether state officials should require testing of private schools that participate in choice programs.
–Rick
Rick: Ashley, a big question when it comes to school choice is whether schools that participate in voucher programs or education savings accounts should be required to administer state tests. Even stickier is whether these schools should be subject to state accountability systems for their results. Those who support mandatory testing argue that it makes sense for schools that receive public funds to be accountable for student outcomes. Those opposed argue that choice schools are already accountable to the parents who choose them and that testing mandates would compromise autonomy, constrain curricula, and homogenize instruction.
In many ways, the debate mirrors familiar battles over test-based accountability within traditional public schools. There are questions about whether any testing should be for accountability or simply for transparency, as well as about whether testing will encourage “teaching to the test” or other harmful practices. The answers depend, in part, on how invasive the tests would be and how much they might impact the scope and sequence of instruction.
But, in the context of private school choice, there are some added wrinkles. For instance, most private schools enroll students who are not using public funds. How would any requirements impact them? Should all students in a school that takes public funds be tested? Moreover, with ESAs, some providers are offering specific services rather than operating entire schools—raising new uncertainties about all of this.
So, there’s a lot there. I’d love to get your take on all of this. But it might be useful to begin with first principles: How should we think about the relationship between test-based accountability and parental choice?
Ashley: You hit the nail on the head: “accountability” divides the field on both sides of the political aisle. Now, I don’t like the state tests we use in the U.S. because they’re focused on skills rather than mastery of knowledge. But before we get into a discussion about the type of tests, let’s start with the reasons why we should care about outcomes in the first place.
The first principle, for me, is that education is not a private good; it’s a public good. Unlike your choices about which jeans to buy, it matters to me that your child knows how to read; unlike my decisions about whether to rent or to buy a car, it matters to you that my children understand the branches of government and where Mexico is on the map; whether a high school graduate goes off to college, the military, or the workforce, it matters to all of us that she has made her choice intentionally and with adequate preparation for any of them. Why else do we tax individuals for other people’s children?
As I wrote in a 2024 book, “At no point has our citizenry declared that education is merely a private pursuit to which public funds must be devoted nor, to my knowledge, has any other democracy made this claim and enshrined it into policy. Such clarity is important because to assess the success of any public policy, we have to be clear about its original aim.”
That tees us up for a second point, which is that there is no good way to know whether we’re hitting our goals other than to look at academic outcomes, and test scores are an important part of that equation.
Rick: You’re right that the “public good” question looms large in discussions of educational choice. Indeed, there are some choice opponents who argue that only publicly operated schools can really serve those shared interests. Now, I don’t buy that claim. After all, one can credibly make a case that charter schools and private schools may actually be doing a better job than traditional district schools of promoting academic achievement or civic education.
On the pro-choice side, there are those who reject this whole conversation as a distraction. Diehard “no-testers” hold that parents are best positioned to decide what their kids need and that schooling should therefore be a familial decision. Period. They see state-mandated testing requirements as burdensome and unnecessary instances of bureaucratic creep. As far as accountability, they argue that schools of choice are already accountable to the families they serve—that they have to satisfy parents or they’ll close their doors.
While I think this camp makes some valid points, I also think their argument is overstated. When families use taxpayer funds to attend a school, taxpayers have a legitimate interest in how those dollars are spent. At a minimum, this entails policing waste and fraud. It also means asking whether public funds are delivering meaningful educational benefits. In higher education, for instance, there are concerns that colleges entice students to use taxpayer-provided loans to pay for degrees of dubious value. Both Democrats and Republicans have concluded this is a big problem. The same potentially applies here. That said, this principle can be a recipe for runaway red tape, so I’m torn as to exactly what it should mean in practice.
What should policymakers or educators make of all this when it comes to program requirements? You mentioned your concerns about the kinds of tests we tend to use in the U.S. Given that, what kinds of testing would you endorse?
Ashley: I think that “school choice” should be accompanied by assessment choice. In the short term, I would love for all choice participants—not necessarily the schools serving them—to take a nationally normed assessment and report the results to the state. The state could then publish aggregate scores when a particular program or school network hits a predetermined threshold of participants.
In the longer term, state leaders have a remarkable opportunity: They can pilot innovative assessments that are knowledge-rich rather than merely skills-based. Why does this matter? In the humanities, once students know how to read, the typical standards or skills are simply inadequate. Background knowledge matters more. The research basis for knowledge over disaggregated skills is so compelling that drilling on “find the main idea” or “compare and contrast” questions instead of providing a systematic approach to history, geography, important literature, and the universal questions about the human experience is borderline unethical. My colleague David Steiner set this out beautifully in this article from September.
High-performing systems around the world—in the United Kingdom, Alberta, and elsewhere—already do this to good effect. And it has been done in the States—even recently. Just look at former state superintendent John White’s success in Louisiana: He not only incentivized high-quality curriculum, but he also supported curriculum-aligned assessments, described here and here, that districts could opt into. This novel assessment was a game-changer for teachers, who were able to focus on content instead of leaping out to do skills-based test prep. (Our team at JHU worked on the project with many others.)
More than a dozen states are already pressing hard on high-quality instructional materials. They could deepen the curriculum play by designing state tests in a way that aligns with commonly used materials in the state’s district, charter, private, and home schools and that actually require something of students instead of chronically underchallenging them.
So sure, I get the trepidation in some quarters about assessments. But what if we had better tests? A variety of tests? Less frequent tests? What if we focused on building students’ knowledge no matter where they went to school?
As E.D. Hirsch noted, citizens in a heterogeneous country need specific reference points in common, even if they disagree about what these reference points mean. A menu of meaningful, knowledge-rich tests could help.
