The paradox of standardized testing

Standardized achievement tests are deeply flawed, and test-based accountability has been terribly damaging to public education. Even so, such tests have their uses.


At the beginning of the school year, my kids came home with results from last spring’s state standardized tests. I was eager to see how they did, but I was also torn, just as I used to be every July when, as a superintendent, I waited anxiously for the data file from the state telling me how the district and schools had done that year.

Every day I tell my kids as they’re leaving for school to work hard and have fun. I’ve told them about the history of standardized testing, the limitations of the tests, the limits they’ve placed on creativity in the classroom, their inherent bias, and their connection to ill-formed policies that are based on a deep misunderstanding of what improves schools. I tell my kids that standardized tests don’t mean anything in the long run and that they’ll never have to take such tests once they start working (unless it’s for a certification of some kind). I tell my kids that I value them as people, not scores.

But even so, when I took a look at their scores, I breathed a sigh of relief that my kids had done as well as other kids in their school.


As a superintendent, and prior to that as a central office leader designing accountability systems, I experienced exactly the same contradiction. During the school year, I would make it a priority to discuss the limitations and flaws of standardized testing with teachers, school and central office leaders, the board, elected officials, parents, students, and the community. But then, when our test results came in, I would dive into them. On one hand, I decried No Child Left Behind’s narrow focus on scores; on the other hand, I saw that I could use the tests to help me design improvements in teaching and learning through an equity lens.

I saw two important benefits in particular. First, test-based accountability forced schools and districts to learn how to use data in new and powerful ways. Second, and more important, the test results made it abundantly clear how poorly our schools were serving black, Hispanic, and low-income students, English language learners, and students with special needs. Thanks to standardized testing, we finally had proof that the system wasn’t serving all students at a high level.

Standardized testing can help show that the system is broken, but in a sense, standardized testing is the system.

However, that only brought me to another paradox: Standardized testing can help show that the system is broken, but in a sense, standardized testing is the system. Since the Progressive Era of the early 20th century, public education has been dominated by what the historian Raymond Callahan (1962) described as a “cult of efficiency,” an almost religious faith in the power of rational, “scientific” management to tame the complexities of life in schools (also see Rose, 2016; Mehta, 2015). At a time when enrollments were skyrocketing and the country was struggling to absorb millions of immigrants, administrators sought to standardize every part of education, from budgeting, hiring, and record-keeping to lesson planning, teaching, and sorting students into separate classes and tracks. And of course, while the sorting mechanisms were said to be “objective,” it was assumed from the start that they would favor white students. It’s not an overstatement to say that, from the beginning, standardized tests were instruments of white supremacy.

So how can a school system leader who understands the sordid history of standardized testing use the results of those tests to push for changes intended to rectify the vestiges of the past?

It all makes my head spin, but I know that there’s no way forward in American education without embracing our contradictions. Frankly, annual state standardized tests aren’t going anywhere, and as our annual PDK poll reveals, the nation’s parents understand that testing is part of the system. It’s just that they want much more than that from our schools, whether career-related classes, art and music, or the development of interpersonal skills.

ESSA and its opportunities

Every year, when I was a superintendent, local politicians would pounce on any disappointing test results that stood out, reveling in the opportunity to beat me up and denigrate the school system. I would respond by explaining that the annual state reports included more than a thousand data points — we reported scores in two subjects, at seven grade levels, for seven demographic groups, at each school. No matter how well our system was performing, we were never going to see all of our indicators move in the same direction at the same time. So the real question should have been: Which of those data points can be most helpful? More important, what else would we need to know in order to identify useful levers we can use to improve schools? What other complementary data would allow for a deeper dive into students’ knowledge and skills, letting us know what they need to be successful? At best, I would tell people, the data can help you ask better questions.

Under the Every Student Succeeds Act, states and districts do have opportunities to collect and use data that go well beyond the traditional standardized tests. In particular, many school and system leaders are experimenting with complementary indicators that focus on social-emotional learning and career-readiness, topics that resonate with parents and can help place standardized test scores in context. Student and staff surveys, for example,
can provide important information about a school’s emotional climate — including classroom engagement, teacher morale, the prevalence of bullying, and other issues — which not only sheds light on why test scores might be going up or down but also raises questions about our priorities for school improvement. For instance, what should we do if we find that a school’s test scores are high but the climate is joyless and oppressive? And how should we evaluate a school that has boosted its math and reading scores but hasn’t moved the needle on specific skills that matter to career-readiness (including the ability to solve complex problems, communicate well, and work effectively in groups)? Does that school deserve a blue ribbon, or does it need help?

If we treat test results as a starting point for further assessment, they can help steer us in the right directions.

Again, my point is that standardized testing is not just inevitable but also useful. I would love to tell my kids to reject those test results altogether (and if I don’t, then perhaps I’m doing them a disservice), and I’d also like to tell every superintendent and principal that they shouldn’t be consumed with their test results every spring because there are more important things to worry about. Only, I’m not ready to give up on the tests just yet. If we treat the results as a starting point for further assessment, they can help steer us in the right directions.

That said, the test-based accountability policies of the past generation have done enormous harm to public schools and students, and we should stop testing every student in grades 3-8. But on this issue, I suspect that the only way to design a better system is to be a radical pragmatist, accepting that we need to start with the system we have. Plus, now that my two older kids are in high school, I’m curious to see how they’ll do on their SATs.         K


Callahan, R. (1962). Education and the cult of efficiency: A study of the social forces that have shaped the administration of the public schools. Chicago, IL: University of Chicago Press.

Mehta, J. (2015). The allure of order: High hopes, dashed expectations, and the troubled quest to remake American schooling. Oxford, UK: Oxford University Press.

Rose, T. (2016). The end of average: How we succeed in a world that values sameness. New York, NY: Harper Collins.


Citation: Starr, J. (2017). The paradox of standardized testing. Phi Delta Kappan 99 (3), 72-73.


JOSHUA P. STARR (@JoshuaPStarr) is chief executive officer of PDK International, Arlington, Va. 


  • Nice work. I agree that the tests are harmful. The kids who do poorly are the ones who show up already behind due to lack of language in the home. These tend to be poor kids and blacks and Hispanics are more likely to be poor. If schools allowed every student to learn as fast as they can the achievement gaps would increase. Trying to reduce the gaps could mean slowing down the fast learner, which some schools do just fine. As for analyzing data from these tests I don’t think it’s worth much effort. If you believe as I do that the tests a crappy, then the data is also crappy. Analyzing crappy data is unlikely to serve any meaningful purpose.

  • Reed

    Testing for more testing? I was with you, especially eliminating testing of every student in grades 3-8; but then… “If we treat the results as a starting point for further assessment, they can help steer us in the right directions”. Not a fan.

  • I am a life member in Phi Delta Kappa International, with membership status approaching 30 years. With this background, I say thanks to Joshua P Starr for being a truth teller and reminding me of why I became a member of PDK quite a long time ago. A long time ago I read “Phi Delta Kappa International seeks to scientifically study education in the interest of humanity. I believe this article on standardized testing is balanced in perspective with a clear view on the way forward. I hope this article begins a national conversation that never ends.

  • Ron Horton

    Testing in two subjects is only one source of adequately determining a students intelligence. Schools are far to limiting in their pursuit of intelligence and they squeeze the creativity of our students right out of them. Students enter the educational process with a wide depth of self knowledge and creativity and we funnel it into educations small and narrowly focused structure. The structure being created by the university professors in their own image. More people with be educated in the future then have come trough the system to date and the system in place will not suffice.

  • Anonymous

    Not saying that there aren’t downsides, misuses, and needed improvements in standardized testing, but this is only a paradox if you exaggerate the harm of testing. Otherwise, well said.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

WP_User Object ( [data] => stdClass Object ( [ID] => 27 [user_login] => jpstarr [user_pass] => $P$Bcm3QCDcd/c2Ob.MLRQWe8mHTnMCEX. [user_nicename] => jpstarr [user_email] => [user_url] => [user_registered] => 2018-08-24 09:29:18 [user_activation_key] => [user_status] => 0 [display_name] => Joshua P. Starr [type] => wpuser ) [ID] => 27 [caps] => Array ( [author] => 1 ) [cap_key] => wp_capabilities [roles] => Array ( [0] => author ) [allcaps] => Array ( [upload_files] => 1 [edit_posts] => 1 [edit_published_posts] => 1 [publish_posts] => 1 [read] => 1 [level_2] => 1 [level_1] => 1 [level_0] => 1 [delete_posts] => 1 [delete_published_posts] => 1 [author] => 1 ) [filter] => ) 27 | 27