Editor’s Note: This post originally appeared on U.S. News and World Report’s Knowledge Bank blog.
We’re in the thick of testing season here in the United States. If you know any kids in public schools, then chances are they are about to take or have recently taken some standardized test that feeds into state accountability systems, as required by federal education policy.
In the years since the inception of this springtime ritual, nationally tied to the enactment of the No Child Left Behind Act in 2002, many have wrung their hands about testing slowly crowding out learning in our public schools. Since teachers are the primary means of both instruction and preparing students for these exams, much of the concern about testing’s adverse consequences has focused on whether and how teachers respond to these goals that appear at odds. Are they teaching to the test to the detriment of authentic instruction? And how do their choices affect our kids?
These are important questions and reflect real concerns among parents. The recent rise of the Opt Out movement has been fueled in part by these anxieties. And perhaps to counter the narrative that teachers may be at fault and to protest the encroachment of test-driven evaluation on teachers’ autonomy, teacher unions have also joined in condemning policy’s overemphasis on standardized tests. But I wonder whether these responses from the public and teachers may be driven more by perceived threats to teaching rather than actual documented responses. Allow me to explain why I feel teaching to the test isn’t as menacing as commonly believed.
First, I argue that we should actually embrace teaching to the test from a policy perspective, even though such a stance may be unpopular in the current debate. A classic economics paper likens test-based accountability for teachers (and their students) to speed traps monitoring drivers: it’s impossible to monitor all drivers on all roads all the time, but strategically placed speed traps keep traffic closer to safe driving speeds that rein in what may otherwise become speedy or reckless driving by some individuals. In other words, curbing individual teachers’ autonomy for the good of all students in the system is, in fact, an important and intended function of tests. Naturally, this is why many groups representing disadvantaged and minority students are stridently in favor of standardized tests, as they feel tests provide a safety against what could otherwise be a tendency to leave low-performing students behind.
Second, concerns about nefarious teaching to the test should be inversely related to the quality of tests. If tests are designed to rigorously assess learning on good standards, these are precisely the tests we would hope are influencing teachers’ practices. Yet, not all state tests are created equal—some are very demanding and represent high standards, while others assess students against a low bar (these tests implicitly have a greater risk as they divert teachers’ focus towards low standards that fail to challenge most students). So concerns about teaching to the test should theoretically be mostly limited to states with low-quality assessments. The good news is that the new Common Core aligned assessments have recently been shown to improve upon even the best of the prior generation of state assessments; the bad news is that states continue to drop out of using these higher quality assessments in favor of the old tests. The irony here is that states doing this are implicitly trading tests that have a good likelihood of promoting genuine learning with those where the wrong type of teaching to the test poses a more credible threat.
Third, the empirical evidence of teachers responding to tests suggests the responses are relatively small in magnitude. A common interpretation of the research literature on pay for performance is that standard incentives for improving student learning don’t actually change teachers’ behaviors all that much. In other words, even when we want to promote more test-based achievement, it’s hard to get teachers to change their behavior (where teaching to the test would be one of several expected responses). It’s not clear why teachers are so unresponsive, but their small responses to individual bonuses implies they are unlikely coerced by even weaker distinctions in the specifics of tests.
There is perhaps reason for more concern in particular situations, like when teachers have many kids near proficiency thresholds who may get more attention at the expense of even lower-performing students. There are also well-publicized cheating scandals—essentially teaching to the test taken to a criminal extreme. While it may be tempting to see news of such scandals and generalize to all other districts and schools, I take the view that cheating among professional educators is clearly an exception rather than the norm in schools. And as technologies like computer-adaptive testing, erasure analysis and other checks to detect suspect behavior on assessments become more commonplace, gaming the system will be easier to detect. Thus, I speculate the prevalence of such scandals is likely to decline in the future.
From my vantage point, the threat of teaching to the test seems to be more hysteria rather than a real danger. This is not to say caution is not warranted, but the notion that standardized tests in general are corrupting our public schools does not square with what I know about the teaching profession.
The Brown Center Chalkboard launched in January 2013 as a weekly series of new analyses of policy, research, and practice relevant to U.S. education.
In July 2015, the Chalkboard was re-launched as a Brookings blog in order to offer more frequent, timely, and diverse content. Contributors to both the original paper series and current blog are committed to bringing evidence to bear on the debates around education policy in America.