The Hechinger Report is a national nonprofit newsroom that reports on one topic: education. Sign up for our weekly newsletters to get stories like this delivered directly to your inbox.

NAEP scores over time
The reading skills of the nation’s eighth-graders rose slightly, but their math skills did not. Fourth-graders didn’t improve in 2017 in either subject. Credit: Jackie Mader/The Hechinger Report

The average performance of the nation’s fourth- and eighth-graders mostly held steady in math and reading from 2015 to 2017, now marking a decade of stalled educational progress, according to the results of a test released Tuesday. The one exception was eighth-grade reading, with the average score rising by one point between 2015 and 2017.

Since the biennial test, called the National Assessment of Educational Progress or NAEP, was first administered in the early 1990s, student achievement, particularly in math, steadily improved until the late 2000s, then flatlined. Reading scores also stagnated. In 2015, scores dipped in math among both fourth- and eighth-graders, and these math scores did not bounce back with the 2017 test. Average students’ scores remain well below what test overseers consider to be “proficient” for each grade level. In reading, 37 percent of fourth-graders and 36 percent of eighth-graders were proficient. In math, 40 percent of fourth-graders and 33 percent of eighth-graders hit this threshold.

“I’m pleased that eighth-grade reading scores improved slightly but remain disappointed that only about one-third of America’s fourth- and eighth-grade students read at the NAEP Proficient level,” said former Michigan Governor, John Engler, interim president of Michigan State University and chair of the National Assessment Governing Board that oversees NAEP, in a written statement. “We are seeing troubling gaps between the highest- and lowest-performing students. We must do better for all children.”

Related: Teaching kids not to be scared of math might help them achieve

Test administrators do not provide explanations for student performance, but scholars and policymakers have been arguing over the root causes of stagnation for years. Some blame the 2008 recession and the resulting drop in public funding for schools and increase in poverty among students. Others point to demographic shifts in the United States. The 2017 test was the first time that white students dropped below 50 percent of fourth-grade test takers. Hispanics now account for 26 percent of the fourth-grade population, up from 19 percent 10 years ago. Disproportionately poor, and sometimes not speaking English at home, Hispanics tend to score considerably lower than white students.

Policy changes may play a role too. Annual testing of students became a federal requirement after 2001, and that sometimes affected instruction. Then, nearly all U.S. states adopted new, more demanding academic standards in the 2010s, and there have been widespread reports about the difficulties in changing instruction in the classroom. (More than 40 states embraced the “Common Core” while additional states adopted similar standards.)

Story continues below chart

Florida bright spot

Florida produced notable gains in 2017, and it was the only state in the nation to make improvements in both math and reading.  (Only fourth-grade reading was flat in Florida, but it well exceeds the national average.) Education Secretary Betsy DeVos has frequently praised the state as a model of “school choice” with a taxpayer-funded voucher program for students to attend private school and many charter school options. The NAEP scores showed stellar gains within the traditional public school system. In math, the average white fourth-grader in Florida exactly matched the 255 score of the average white fourth-grader in Massachusetts, the top performing state in the nation with a wealthier and more highly educated populace. More impressively, Florida had smaller gaps between students of color and whites. Florida’s Hispanic fourth-graders outscored Massachusetts’: 242 vs. 234 in math.  Hispanics make up a third of Florida’s fourth-graders and fewer than a fifth of Massachusetts’. In reading too, Florida’s Hispanic population outscored Massachusetts’.

Two metropolitan regions, Miami and Jacksonville, helped drive the gains for Florida. Miami-Dade County, the fourth largest school district in the nation, and Duval County, the 20th largest, were two of only four large, urban districts to improve in fourth-grade math. (Alberto Carvalho, the superintendent of Miami-Dade County, made headlines in February 2018 when he accepted, and then rejected, an offer to become New York City’s schools chancellor.)

The Urban Institute, a Washington D.C.-based research organization, calculated “demographically adjusted” scores for each state Tuesday, showing how each state would stack up if it educated a similar mix of students with the same racial and economic backgrounds. Florida topped the chart in fourth-grade achievement; Massachusets was still strongest in eighth-grade.

Related: Is it time to update NAEP?

Concerns over computers

Fourth-grade test scores in both reading and math slid in five states: Alaska, Louisiana, New Hampshire, South Carolina and Vermont.  Last month, the Louisiana state superintendent of education, John White, complained that NAEP’s switch to digital-based testing penalized his state.  The 2017 NAEP was the first time a majority of students across the nation took the test on a computer, instead of with pencil and paper. Test administrators brought their own tablets and internet routers to testing sites. But even when the technology is working perfectly, it can be an adjustment for students to navigate a computer screen, perform drag-and-drop activities, do scratch work in math or underline words in a reading passage.

Students commonly perform worse on computerized tests than they do on paper-and-pencil ones. The release of the 2017 NAEP scores was delayed several months so that statisticians could adjust the raw scores upwards and confirm that they were comparable to the results of prior pencil-and-paper tests. To accomplish this, NAEP still administered a paper-and-pencil test to 20 percent of test takers and then calculated the gap between the paper and the computer scores for different groups of students. “We have the full weight of psychometric scientific evidence….that we are comparing apples to apples,” said Peggy Carr, acting commissioner for the National Center for Education Statistics, in a briefing with reporters ahead of the score release. “The results will be based on student performance and not contaminated.”

Michael Petrilli, president of the Thomas B. Fordham Institute, a center-right education policy think tank, wrote in a blog post last week, “if scores drop among low-income and low-performing students—the kids least likely to be comfortable with digital devices, especially in the fourth grade—that could signal that something went awry.” Indeed, math and reading scores decreased for lower performing fourth-graders, both for the bottom 10 percent and the bottom 25 percent. For individual states, fourth-grade tests scores were more likely to disappoint. Ten states saw declines in math. Another nine in reading. Worsening reading performance in fourth grade, especially among low-performing students, was also detected in a 2016 international exam which was administered entirely by pencil and paper.

The NAEP is the only national exam and it allows scholars and policymakers to compare students in one region of the country with another. Every state has its own annual assessments, but they vary in difficulty and how they are scored. Fewer than 600,000 fourth- and eighth-grade students take the NAEP, but they are carefully selected, as in an opinion poll, to represent the actual geographic, income and racial differences in the nation. Twenty-seven large school districts, representing roughly 10 percent of the nation’s students, volunteer for extra testing so that they can have NAEP scores for their cities, in addition to their states.

Related: Third indication U.S. educational system is deteriorating

Achievement gaps widening

The decline among low-performing students means that achievement gaps are widening. In fourth-grade reading, for example, the gap between the top 10 percent and the bottom 10 percent of students widened by four points. In fourth-grade math, the gap widened by six points.

For eighth graders, the strongest students are advancing. The only reason national reading scores increased was that students in the top 10 percent and top 25 percent scored higher. Scores for average and below average students were flat. The gap between the top 10 percent and the bottom 10 percent widened by three points. In math, the gains by top students offset worsening scores by weaker students. The national average was left unchanged, but the achievement gap between the top 10 percent and the bottom 10 percent widened by six points.

Notably, these widening achievement gaps between the strongest and weakest students aren’t reflected in the racial or income data. The achievement gaps between different races were unchanged since 2015. There was also no increase in achievement gaps between students who are poor enough to qualify for free or reduced-price lunch and those who don’t.

California cities improve, Chicago and D.C. stall

California’s fourth-grade scores were rather flat statewide, but three large cities were standouts. San Diego, Fresno and Los Angeles lead the nation’s cities in showing the biggest gains in fourth-grade reading scores. San Diego and Fresno were also among the top three urban school districts showing the biggest gains in fourth-grade math.

Meanwhile, Chicago students, who had shown above-average gains in 2015, didn’t continue to improve in 2017. Last year, the New York Times celebrated Chicago as the most effective school district in America, based on how much students had been advancing each year from 2009 to 2015. Washington D.C. was also another high flyer in years past. After many years of impressive gains, educational progress stalled in 2017. Students remain well below the national average.

The Hechinger Report provides in-depth, fact-based, unbiased reporting on education that is free to all readers. But that doesn't mean it's free to produce. Our work keeps educators and the public informed about pressing issues at schools and on campuses throughout the country. We tell the whole story, even when the details are inconvenient. Help us keep doing that.

Join us today.

Letters to the Editor

1 Letter

At The Hechinger Report, we publish thoughtful letters from readers that contribute to the ongoing discussion about the education topics we cover. Please read our guidelines for more information. We will not consider letters that do not contain a full name and valid email address. You may submit news tips or ideas here without a full name, but not letters.

By submitting your name, you grant us permission to publish it with your letter. We will never publish your email address. You must fill out all fields to submit a letter.

  1. That test scores have stabilized and stopped going up is the best enducational news in a decade. The only justification for the pervasive policy of trying to increase test scores is that future national economic success will go to the nations with the highest scores. This is a myth, as I showed in 2007 (Phi Delta Kappan, Oct.) in an undisputed analysis showing that future economic success goes to low scoring nations, not high scoring. Consider the first international test from 1964. The lowest scoring nations were Sweden & USA. Highest scoring were Israel and Japan. Pick any current national economic outcome measure you want and you will find that low scoring Sweden and USA blew the high testers out of the ball game.

    NAEP, like all massive testing programs, is at best a monstrous waste of money. At worst, and very likely, NEAP and all massive testing program, have done severe and perhaps irreparable harm to the nation. It is long past time to stop this stupidity and end testing.

Submit a letter

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