We Don’t Need Another List Ranking Schools Based On Test Scores

Test-based rating systems have already distorted curriculum and instruction.


Long known for its college rankings, U.S. News and World Report is now ranking elementary and middle schools on the basis of test scores. That’s a terrible idea.

“What is wrong with you people?” a professor at the University of Virginia tweeted at the magazine’s editors. Former U.S. News editor James Fallows took a similar if more polite tack on NPR. “These rankings have already done enough to the higher education ecosystem that, with all the problems public elementary schools and high schools have, they don't need this as well,” he said.

U.S. News staffers say they hope parents will benefit from the “sophistication” of their elementary and middle school analysis, which attempts to account for student demographics. The methodology uses two criteria: a school’s reading and math state test scores, and also how well each school scored compared to other schools with similar proportions of economically disadvantaged students and underserved minorities. (Unlike the high school and college rankings, which are national, those for lower grade levels are ranked only by state and district, because different states give different tests.)

The objective is to avoid rewarding schools that simply enroll children of the affluent, who generally do better on standardized tests. U.S. News has made a similar effort with its college rankings, instituting a separate “social mobility” category. But its top-rated universities are elite institutions like Princeton, Columbia, Harvard, and MIT.

The elementary and middle school rankings aren’t much different, according to Michael Kurtz, associate professor of economics at Lycoming College in Pennsylvania. After scanning the list for his state, he told me, “It still basically looks like schools with lots of economically disadvantaged students are at the bottom of the list and those with lots of affluent students are at the top. Well, you just made a map of the wealth in the country.” An analysis by the education website Chalkbeat came to a similar conclusion.

When I looked at the rankings for Washington, D.C., where I live, I found a slightly more complicated picture. Among traditional public schools, those at the top did generally have low proportions of economically disadvantaged students. But the story was different for charter schools, which enroll about 45% of D.C. public school students. Among the top ten elementary schools, one was a charter serving a low-income population; among the top D.C. middle schools, at least four fit that description.

It’s possible those schools are helping their students beat the odds, but that’s hard to say unless we know how they fare in high school and beyond. The U.S. News rankings of high schools and colleges may be simplistic, as critics charge, but at least they don’t just rely on reading and math scores. The criteria also include things like graduation rates and AP scores (for high schools) or academic reputation (for colleges).

As Kurtz notes, test scores are only one aspect of many that parents should look for in choosing a school. What about fostering kids’ creativity, social skills, curiosity, he asks—or just helping them have fun? And he worries that if parents base decisions on U.S. News rankings, “schools may feel they have to focus more on test scores than they otherwise would.”

He’s right to be worried. Curriculum and instruction in elementary and middle schools have already been distorted by pressure to raise scores at those grade levels. (Federal law also requires testing once in high school, but those scores tend to receive less attention.) Many states and districts, including D.C., have their own publicly available rankings based on scores, as does the website GreatSchools. Some of these, unlike U.S. News, at least factor in test-score growth.

Because of the emphasis on scores, schools have marginalized or eliminated social studies and science from the curriculum to focus on reading and math—especially where test scores are low. And reading instruction has come to mimic test structure: short, disconnected passages followed by questions that purport to assess the ability to “find the main idea” or “make an inference.”

High-stakes testing didn’t create this instructional approach, but it did exacerbate it—and the U.S. News rankings, along with headlines about the “top ten elementary schools” in a state, could make things even worse. It’s not just that a steady diet of “finding the main idea” is boring. It’s also self-defeating.

Contrary to education orthodoxy, reading comprehension isn’t just a set of skills that can be taught in the abstract and applied to any text. It’s essentially an outgrowth of academic knowledge and vocabulary. And the subjects that have been sidelined in the quest for higher reading scores—social studies, science, and the arts—are the ones that have the most potential to boost that knowledge and vocabulary, if students are exposed to them through a coherent, content-rich curriculum.

Parents do need help in choosing schools. And surveys have shown that what parents care about most is “academic achievement.” That has come to mean high scores on standardized tests, but it doesn’t have to—and especially in the case of reading tests, that measure can be misleading. High scores at lower grade levels don’t necessarily translate into the ability to do college-level or even high-school level work.

Passages on reading tests at lower grade levels don’t assume much academic knowledge or vocabulary, so it can look like children are getting better at “finding the main idea.” But at higher grade levels the passages become increasingly complex. And if their schools haven’t been systematically building academic knowledge, students may no longer understand what they’re expected to read—on tests and in the classroom. Reading tests also fail to measure students’ writing ability, which is crucial in high school, college—and often, in life.

What can policymakers, governmental agencies, and publications do to provide parents with a better guide to school quality? Kurtz and others have proposed a Consumer Reports-style list of attributes. I’d say that should include a key factor that rarely gets mentioned: curriculum. Are young children being taught to decipher words through systematic instruction in phonics, as supported by scientific evidence? Are kids listening to and discussing complex text that builds academic knowledge and vocabulary, beginning in kindergarten, guided by one of several recently developed elementary literacy curricula?

It’s surprisingly hard to get that kind of information, but that has to change. We need to understand that what gets taught is vitally important. It would also be nice to have a system of school inspection, on the English model, that could help parents determine whether a curriculum is being delivered effectively.

Ideally, we would administer a different kind of reading test, with passages grounded in the content of the curriculum, or at least in broader state academic standards for social studies and science. That would help level the playing field for students less able to acquire academic knowledge outside school and give teachers an incentive to focus on teaching content rather than illusory “skills.”

For at least the past twenty years, we’ve looked to reading and math scores as the measure of school quality. During that period, we’ve seen stagnant or declining scores and a widening gap between high- and low-scorers. Some argue the problem is not enough accountability—that is, not enough emphasis on holding schools accountable for student test scores. Those observers might view the new U.S. News rankings as a welcome development.

But the evidence indicates we’ve put too much emphasis on accountability, at least in its current form. If we don’t wean ourselves from our dependence on standardized tests, we’ll only continue to doom millions of students to a version of education that squelches their natural curiosity and their true potential.