For example, U.S. News & World Report said in this story, headlined “U.S. Fourth-Graders Lag Behind Other Countries in Reading”:
Reading comprehension among fourth-grade students in the U.S. has flatlined since 2001, allowing education systems in other countries whose students used to perform worse than those in the U.S. to catch up — and even surpass — the U.S. in an international ranking.
The Washington Post reported in this story, headlined: “U.S. schoolchildren tumble in international reading exam rankings, worrying educators”:
The United States tumbled in international rankings released Tuesday of reading skills among fourth-graders, raising warning flags about students’ ability to compete with international peers.The decline was especially precipitous for the lowest-performing students, a finding that suggests widening disparities in the U.S. education system.
PIRLS is one of several international tests that students from dozens of education systems around the world take regularly, and U.S. policy-makers always express concern about how well U.S. students do on them. Such concern ignores the fact that U.S. students have, as a whole, never done well on these exams — not in the history of international tests — and that there are big questions about how some of these are created and whether scores from different countries can actually be compared.
So if you are worried about the PIRLS information, please read this post. It will give you some new perspective. It was written by James Harvey, executive director of the National Superintendents Roundtable. You may also want to read a post he wrote titled “Ten things you need to know about international assessments.”
By James Harvey
Americans can be forgiven for missing this, but since 1996 when the U.S. Department of Education published the results of an international assessment of fourth-graders’ reading skills, the results have defied the expectation that American students would perform poorly. As the government dryly reported at the time: “U.S. students turned in a creditable performance vis-à-vis their peers in other nations.” Actually, they turned in a spectacular performance. Among 32 nations, the average performance of American students in fourth-grade reading placed them second. Finland took pride of place.
The public can be forgiven for missing this because results like these tend to be ignored by editors. Headlines indicating that American students are slipping, or doing worse than they did last year, find plenty of space on Page 1 and editorial commentary. The stellar results from fourth grade might as well be filed among the legal notices for all the attention they receive.
That may change any day now. The latest iteration of this international assessment of reading in the fourth grade (Progress on International Literacy Study, PIRLS) is likely to generate plenty of hand-wringing. Out of 58 education systems participating in the assessment, 12 report PIRLS scores higher than the U.S. scores, according to a pre-release briefing on the highlights issued by Peggy G. Carr, the impressive acting commissioner of the National Center for Education Statistics.
On one hand, that puts the U.S. average scores in the top quarter of the participating school systems. On the other, 12th place isn’t nearly as satisfying as second. And the slippage nicely fits the decline narrative with which editors approach school news.
Let’s leave aside all the questions that can be legitimately raised about these international assessments. And let’s ignore for now the fact that some of these comparisons are of apples and oranges. For example, all the fourth-grade students in the United States — rich and poor, documented and undocumented, able-bodied and struggling with disabilities — are being compared in some cases with students in cities, provinces or subpopulations of other nations — Hong Kong, Northern Ireland, Taipei and just the French-speaking population of Belgium.
But put aside the peculiarity of international assessments that compare the performance of 100 percent of the fourth-grade children in most nations with just some of their peers in others. Even accepting those shortcomings, in viewing the latest PIRLS results, there is a lot to like.
The first thing to point out is that the general pattern defined in the 1996 report holds. U.S. fourth-graders continue to stand out. In every PIRLS assessment — 2001, 2006, 2011 and 2016 — the average reading performance of American fourth-graders is consistently on, above or around the PIRLS “High” benchmark. A score of 550 or above on PIRLS’ 0-1,000-point scale marks the “High” benchmark. The U.S. average score in 2016 is 549, statistically no different from a score of 555.
That’s a remarkable accomplishment given the demographic profile of students in the United States — a highly diverse student population, 50 percent of them or more coming from low-income backgrounds (as measured by eligibility for free- and reduced-price meals).
Far from rending one’s garments about a 12th-place ranking, one should be encouraged that the average score of American fourth-graders in reading places them solidly in the top quarter of 58 “education systems.” Here it is worth noting that four of the top 12 are enclaves of privilege or relative privilege — Moscow, Northern Ireland, Hong Kong and Chinese Taipei.
Because the overall score of American students is so impressive (549) it is hardly surprising that on all four PIRLS sub-scales, U.S. performance outstrips the PIRLS mean of 500. PIRLS assesses reading for literary experience, acquiring and using information, retrieving and straightforward inferencing, and interpreting, integrating, and evaluating. The mean scores for U.S. students on these scales are 557, 543, 543, and 555, respectively.
It is encouraging, too, that the gap between males and females has narrowed. The advantage girls have over boys in overall reading ability at the fourth-grade level has declined from 18 points in 2001 to 8 points in the latest results. The smaller gap has not come at the expense of female participants. Girls’ scores have increased 2 points, while the average for boys has jumped by 12 points.
A new twist in the latest PIRLS assessment is the administration of the assessment via computer in a simulated Internet environment. Some 16 educational systems participated in this ePIRLS experiment. Only three can be said to have outperformed the United States (Singapore, Norway, and Ireland). This result can likely be linked to the wisdom of an earlier Congress, which insisted as part of the Telecommunications Act of 1996 that everyone deserved access to advanced telecommunications services at reasonable rates, while requiring telecommunications companies to provide their services to schools and libraries at a substantial discount.
Yet one especially troubling finding jumps out of Carr’s briefing: the shameful difference in reading achievement levels by income and race. Students in wealthier communities (those in schools with fewer than 10 percent of students eligible for free lunch services) produced average scores of 587. (Those students in schools with between 10-24.9 percent scored even higher: 592.) But students in schools with 75 percent or more eligible for free lunch services produced an average score of just 516.
Meanwhile, white and Asian students outperformed other students of color, with Asian students significantly outperforming white, black, and Hispanic students.
So the results do not indicate broad, sunlit uplands across the board. Clearly a lot of work remains to be done. But don’t miss the forest for the trees. The latest PIRLS results show that when it comes to reading performance, American fourth graders continue to shine.