It’s easy to see why school choice is popular with parents: It enables them to make the educational arrangements that best convey the values, knowledge, and skills they wish their children to acquire. And if they’re dissatisfied, they can move their children elsewhere. They don’t have to settle for their “assigned” school.

But what happens when we don’t trust parents to do that?

Even some who are sympathetic to school choice worry that accountability to parents is insufficient. They believe that bureaucrats should ensure parents are making good choices by curating the options available in school-choice programs. They envision using standardized test results to identify schools that are doing a poor job, and making those subpar schools ineligible for choice programs.

This use of test results as a form of consumer protection is based on the assumption that schools with lower test scores are “bad schools” and that parents are harming the long-term prospects of their children if they select those schools over ones with higher scores.

But test scores are strongly influenced by the students who are enrolled at a school. Schools that target disadvantaged populations tend to have lower test scores, even if they are doing a good job of helping their students overcome those disadvantages and make progress.

>>> The Clumsy Crusade Against School Choice

The solution, some argue, is to focus on growth in test scores, not on the level of scores. Schools that increase scores are highly regarded, while the ones where students make smaller gains or even declines are frowned-upon places that should not be considered an option for parents.

But there’s a deeper problem. Even using test-score gains as the indicator of school quality presupposes that the change in scores actually indicates whether schools are more effective at preparing young people for success later in life. Unfortunately, the ability of schools to increase test scores is a very poor predictor of that.

More than 20 studies have rigorously identified changes in test scores that schools make and tracked those students over time to see how they develop. The most recent was published this year in the prestigious American Economic Review. The results are consistent with the previous ones. Students who attended schools that grew test scores are not necessarily the ones who fared better long term.

The study’s authors examined charter schools in Massachusetts. They confirmed earlier research that found charter schools in Boston built on the “no excuses” model of high expectations, strict discipline, and narrow focus on traditional academic subjects produced large test-score gains for their students. Meanwhile, charter schools outside Boston that typically deviate from the same approach often reduced student test scores.

A bureaucrat equipped with these test-score results might conclude that the Massachusetts charter schools outside Boston were bad schools and should be shut down or made ineligible to enroll new students. Parents wishing to send their children to these schools, bureaucrats could argue, are just ignorant of the damage those choices would do, requiring officials to step in and protect children against them.

But the study tracked the students to see whether they earned a college degree. The Boston charter schools that had large test-score gains did boost college completion by 4.6 percentage points.

>>> From Mass Deception to Meaningful Accountability: A Brighter Future for K–12 Education

However, the “nonurban charter schools—the same schools that do not increase test scores—increase four-year college enrollment and Bachelor of Arts attainment by 9.7 and 11.4 percentage points, respectively.”

The charter schools that performed poorly on standardized tests were much better at preparing students for later success.

Why would test-score gains not necessarily predict the schools where students had better outcomes in the long run? Test scores might be artificially boosted if schools narrowly focus on the topics covered by the tests, but later success may depend on learning a broader set of knowledge. Non-academic aspects of school, like sports or clubs, that are not captured by tests may convey important skills and motivation for later success.

Parents are much better positioned to see the broader picture and pick the right schools for their own children—not distant bureaucrats armed only with test scores. If Massachusetts education officials had closed charter schools outside Boston, or forced them to imitate the same “no excuses” model adopted by Boston charters, they might have boosted test scores but halved the college completion benefit for those students.

Bureaucrats relying on test scores can’t distinguish the good from the bad, thus rendering them incapable of holding schools accountable for results. Parents know best which schools are working for their own children and are key to meaningful school accountability.