Questions & Answers About Reauthorizing No Child Left Behind

The 114th Congress is attempting to pass a reauthorization of the Elementary and Secondary Education Act, known to many as its last iteration, the No Child Left Behind Act of 2001 (NCLB). The Republican-controlled House Education & Workforce Committee has already approved an overhaul of NCLB on a party-line vote, and the full House expects to take up the bill this week, with the Senate not far behind. This memo answers some of the most frequent questions posed about NCLB and its legacy for students and explains the policies that are under discussion to update, or overhaul, that law.

1. How did students fare under No Child Left Behind?

The No Child Left Behind Act of 2001 was not perfect—everyone agrees it has problems that should be addressed in the next reauthorization, and most people think we should move to a model that is significantly less prescriptive. But in many ways, it worked. Data from the National Assessment of Educational Progress (NAEP), the largest nationally representative longitudinal assessment in the country, shows that American students across the board have seen academic gains since NCLB’s enactment. The average 9 year old went from glacial reading improvement (+0.1 point/year pre-NCLB) to steady gains in the decade since (already up 9 points). In math, 9 year olds improved a scant 2 points over the entire decade of the 1990s; they've gained one point per year since NCLB. Gains were even more evident among students of color. African American 9 year olds improved by a whopping 20 points in reading, narrowing the black-white reading achievement gap by 12 points, and Hispanic 9 year olds improved by an impressive 15 points. In math, African American 9 year olds gained 15 points, and Hispanic students of the same age picked up an astounding 21.¹ To proclaim that NCLB failed is to completely ignore the data, which supports the exact opposite conclusion—without question, students across the U.S. saw gains in their academic achievement during the NCLB-era of annual assessments and accountability, and we began to close gaps that have persisted for decades.

2. Did NCLB lead to over-testing?

Under current law, the federal government requires students to take a statewide assessment in math and reading once a year in grades 3 through 8 and then one time in high school. Though parents and teachers in some areas of the country may be concerned about the amount of time students are spending preparing for and taking tests, the majority of those tests are put in place at the state and district levels, where local decision makers have chosen to layer on additional testing requirements (in some places, exorbitantly) on top of the federally-required assessments.² And because districts and states are responsible for the bulk of the tests students (and parents) are experiencing, eliminating the federal ones will not solve the problem of over-testing. Instead of getting rid of the annual statewide testing required by the federal government, and the accountability that goes with it, we should take a closer look at the additional tests states and districts are piling on top of those requirements and then streamline and eliminate those which are duplicative or low quality.

3. How do parents feel about standardized testing?

Though those who have concerns about testing often represent the loudest voices in the political debate, polling shows that parents actually approve of testing for the purposes of accountability and tracking student progress. Two-thirds of parents (66%) view standardized assessments favorably, and among African American and Hispanic parents, those numbers are significantly higher (75% and 79%), likely due to concern about the low expectations that have too often plagued students of color in the absence of common benchmarks.³ Parents value annual tests because they allow them to track their kids’ progress. Of course, we need to ensure the types of tests students are taking are measuring learning in the most effective ways possible, and many of the tests have improved significantly over the past decade toward that goal. But without annual tests and accountability for all students, as well as specific achievement breakdowns for low income students and students of color, millions of parents won’t know how their kids (or their kids’ schools) are doing until it’s too late.

4. Could requiring fewer tests at the federal level reduce test anxiety?

At first blush, it might appear that proposals to only test students once in each “grade span” (elementary, middle, and high school) rather than annually in grades 3 through 8 might offer a way to reduce testing anxiety. But this assumption breaks down when you consider that fewer tests actually means higher stakes, and higher stakes could mean increased anxiety for students. How would you feel if you knew a single test taken only by you and the few other students in your grade would determine how your entire school was judged? Not only could we see that single tested grade turn into a gap year reserved solely for test preparation, but few teachers would actually want to teach in the only high-stakes grade in a school. Grade span testing would also have the unintended consequence of eliminating our ability to measure the single most important indicator of success: whether every child is achieving at least a year’s worth of growth and learning every school year. With no way to compare a student’s score with one from the previous year, kids who have struggled in the past or started behind their peers could be labeled failures even if they are making huge gains and on track to catch up. And by measuring students against a static yardstick of proficiency instead of annual growth , the teachers who work with kids who need help the most could be punished, not rewarded, even if their students are rapidly gaining.

5. Can we still track kids with fewer tests?

Without annual statewide testing and the accountability that goes with it, we would lose the ability to monitor the growth of millions of students—particularly students who have been historically underserved, like students of color, students with disabilities, and students who speak English as a second language. Because privacy concerns limit states and districts from releasing or relying upon data from too small a number of students (often 30), far fewer schools would be required to report the test results for these disadvantaged groups under a grade span approach—because they would now only be able to do so if they had 30 students of that group in a single grade, not the whole school. That also means far fewer schools would be held responsible for making progress among those groups. For example, an analysis by Chad Aldeman from Bellwether Education Partners featured in The New York Times found that in D.C., “about half of the city’s 119 elementary schools with fifth graders taking math tests would not be held accountable for the progress of low-income or African American students, because there aren’t enough of them in that grade to constitute a reliable sample size.”⁴ In Richmond, Virginia, only 2 of 27 schools with fifth graders would be held accountable for Hispanic students and English language learners. Worse still, none of those schools would have to report scores for students with disabilities.⁵These are just a few examples of an undisputable fact—if we move away from annual statewide testing, millions of students would disappear from state progress reports, and their schools would no longer be responsible for showing whether these students were learning.

6. Can we split the difference, still testing kids annually but only holding schools accountable for one test in each grade span?

Keeping annual statewide testing but not allowing the results of those tests to be used for anything wouldn’t solve any issues with over-testing, and it also would create more problems just as outlined above around proposals for testing once a grade span. Worse, requiring annual testing without having states report the progress of their students on those annual assessments seems like a waste of teachers’ and students’ instructional time, along with the federal funding that is helping to support it. Under this proposal, districts and states would likely continue to layer on additional tests, and just as with a grade span approach, the years when the tests actually count would be even higher stakes for students and teachers than under current law. And using the data for accountability only once per grade span would make it impossible to measure individual student growth, leaving a static “proficiency” marker as the only way to measure success. Demanding annual tests without annual accountability makes little sense—ultimately, it’s just bad for kids.

7. Will states continue to maintain a high bar for their students without federal oversight?

The evidence available points to a short answer: no. Before the passage of NCLB, 17 states didn’t even have an accountability system for their schools, and a measly 2 states took into account the performance of low income students or students of color in whatever system they did have in place. Even after NCLB was enacted, states have attempted to inflate their students’ progress by lowering the bar for proficiency on their state assessments.⁶ For example, by 2009, 35 states had set their 4th grade reading proficiency score at a level that would be considered “below basic” on the National Assessment of Education Progress—and not a single one set it at a level that would be considered actually “proficient” under that longstanding and reputable test.

In addition to lowering standards generally, there’s even more evidence that states will turn a blind eye to persistent income or race gaps in their schools. Given additional flexibility under the first round of NCLB waivers, a review by the Alliance for Excellent Education found that states took major steps back in tracking and ensuring progress among the students who need help the most. Under approved waivers, 9 states created accountability systems in which low graduation rates for specific groups of kids (i.e. students of color, students with disabilities, or English Language Learners) didn’t qualify a school as one of the “focus” or “priority” schools in need of improvement. In 5 states, low graduation rates among those specific groups didn’t even count toward the state’s accountability index at all—meaning a school could be awarded top marks compared to other schools in the state even if its graduation rates for students of color or students with disabilities were abysmal.⁷ If this is what states are doing with some level of federal accountability in place, one can only imagine what some might do without any oversight at all.

8. Can we let districts experiment and develop their own tests?

Current law already includes an allowance of flexibility for districts to opt out of the statewide tests required by the federal government and innovate by piloting their own assessments. But they can only do so if they can prove that those tests will yield comparable data to the ones employed at the statewide level. For example, New Hampshire is currently piloting their own district level tests, but these tests took years to develop and have been deemed comparable to the statewide assessment.⁸ It is crucial that this flexibility not be expanded to create an exception that swallows the rule. Though many people are critical of certain pieces of NCLB, nearly everyone agrees that it did make major progress by allowing parents, teachers, researchers, and policymakers to compare schools and measure achievement of students, including high-needs groups of students, across districts and states. We shouldn’t undercut that progress by allowing districts to opt out of statewide assessments without proving to the federal government that they will be able to produce the crucial data that makes it possible to measure the progress of that district’s students and compare it to those in other districts.

9. Don’t states and local communities know best how to run their education systems?

Yes, and that’s why states should have the flexibility to set their own goals depending on the needs of their schools, which can include targets that we know lead to college and career readiness like improving graduation rates among all students and high-needs groups, ensuring each student gains one grade level or more per year, and narrowing achievement gaps among student populations. But if states are setting their own goals, they should be able to meet them—and if they don’t, and their efforts to correct those problems fail over and over again, the federal government shouldn’t continue to let states simply pour $25 billion in federal taxpayer dollars into a broken system. We simply cannot return to a time when some students are sentenced to attend persistently failing schools with no path to progress in sight.

10. If Title I funding is meant to target high-need students, should it follow those students to whatever school they attend?

Both the House and Senate rewrites of NCLB include a “portability” provision which would allow funding from Title I to follow a specific student, rather than targeting that support at the schools that have the most high-needs students. This provision is rooted in a misnomer: although policymakers and analysts often refer to “per pupil spending” when discussing state- and district-level funding, state and district officials do not budget based on how much money is available to an individual child. Furthermore, research has consistently shown that it costs more money to educate a child in concentrated poverty than to educate the same child in a more affluent community—a fact which forms the basis for why Title I funds are needed in the first place. But the idea of allowing federal tax dollars to follow an individual child does not account for this critical difference. So while the intentions behind Title I “portability” might be good, it’s clear that in practice, it would redirect funding from the poorest schools to wealthier schools in each state. In fact, it would take away $700 million from students in school districts with a concentration of poverty above 25% and give $470 million to districts with low concentrations of poverty.⁹