Student Achievement

U.S. States Fare Well on Global Tests, But Not at ‘Advanced’ Tier

By Andrew Ujifusa — December 12, 2012 7 min read
  • Save to favorites
  • Print

International test results in reading, math, and science show individual U.S. states performed relatively well, but they had a very small share of top-flight students compared with those of high-performing countries.

In this year’s reports from the Progress in International Reading Literacy Study and the Trends in International Mathematics and Science Study, states were broken out as individual “education systems” that can be compared with other international results, although not all of the same states had their performances broken out on all the tests.

In 2011, PIRLS was administered to 4th graders, and TIMSS was administered to both 4th and 8th graders. PIRLS was first administered in 2001, and in 2011, 53 “education systems” participated (including entire countries’ public school systems, but also specific school systems within countries, like Hong Kong or Flemish Belgian schools.) TIMSS was first administered in 1995, and in 2011, 57 countries or “education systems” participated in the 4th grade tests, and 56 countries participated in the 8th grade tests. However, for TIMSS in 2007, the only states to be broken out as separate systems were Massachusetts and Minnesota, and that number increased significantly in 2011 as you will see.

Florida came up a big winner on PIRLS, in terms of average scores (more on that in a bit). On TIMSS, Massachusetts, mirroring its strong performance on the National Assessment of Educational Progress, did very well in both science and math, compared with the United States and other high-performing countries and education systems. However, the good news largely evaporates when you look at the percentages of students that scored “advanced,” “high,” “intermediate,” or “low” on the tests, the formal benchmark names for performance levels. In terms of the share of students scoring at the top level, or advanced, other high-performing countries leave many U.S. states eating dust.

For the PIRLS test administered to 4th grade students in 2011, results from Florida’s public schools are included in comparisons between international education systems because it was counted as a separate “national” education system. So Florida’s results are compared with U.S. results as a whole throughout the report released by NCES. It’s worth noting that those U.S. results do include the same Florida results, so to a certain extent, Florida’s isolated score is being compared against itself.

Anyway, Florida has reason to be pleased with how it stacks up internationally in terms of its average score. Its public education system earned an average PIRLS score of 569, which was well above the U.S. national average of 556. The other four “nations” to top the U.S. average were Hong Kong (the highest-scoring with a 571 average), followed by Russia (568), Finland (568), and Singapore (567).

During education policy and achievement discussions, all those education systems, except perhaps Russia’s, come up frequently when people point to countries that outperform American students or could serve as helpful policy role models. So to the extent one takes stock in international assessments like PIRLS, Florida has to be pleased that all by itself it can go punch-for-punch with traditional academic powerhouses like Hong Kong, which of course isn’t strictly a “nation” itself. (China as an entire nation did not participate in PIRLS or TIMSS.)

What about those various achievement levels? In this case, Florida’s students stack up pretty well. In Singapore, 24 percent of students were deemed “advanced,” compared with 22 percent in Florida. No other country or education system had better than 19 percent of its students scoring advanced. So by this measure as well, the Sunshine State can put on a happy face.

Now let’s look at TIMSS in math and science. This test has more breadth, both because it measures 4th graders and 8th graders, and also because more states agreed to be broken out as individual “education systems” in order to be compared with other countries. (As with PIRLS, some countries like China did not participate as an entire nation.) Florida and North Carolina acted as individual education systems in the 4th grade, and Alabama, California, Colorado, Connecticut, Florida, Indiana, Massachusetts, Minnesota, and North Carolina acted as such in the 8th grade.

The TIMSS math breakdown runs as follows:

• North Carolina (554) beat the U.S. average score (541) in 4th grade math (which in turn was higher than the average TIMSS score). The state’s education system was among 12 internationally that beat the U.S. average. But again, remember that the scores from the Tar Heel State are part of that U.S. average. In addition, Singapore’s 4th graders, the world’s top performers, on average did far better (606) than North Carolina’s. Florida technically achieved a higher score (545) than the U.S. average, but it was considered not “measurably different” from the U.S. average.

• In 8th grade math, Massachusetts (561), Minnesota (545), North Carolina (537), and Indiana (522) all had higher average scores than the United States (509). Massachusetts on its own had a higher score than all but five education systems (South Korea, Singapore, Taiwan, and Hong Kong), with South Korea (613) coming out on top. Colorado (518), Connecticut (518), and Florida (513) also achieved technically higher scores than the U.S. average, but again, these were considered not “measurably different” from the U.S. average. California (493) and Alabama (466) were below average U.S. performance.

• Now for those achievement levels, which spell trouble. On the 4th grade test, 43 percent of Singapore’s students scored advanced, compared with a mere 16 percent of North Carolina’s students and 14 percent of Florida’s. In the 8th grade, it’s no better: 49 percent of Taiwan’s students scored advanced, compared with 19 percent in Massachusetts and 14 percent in North Carolina.

• One important discovery from the mathematics statistics is that from 4th to 8th grade, average U.S. performance in math declined by 32 points. North Carolina’s score dropped by 17 points, while Florida’s dropped by 32 points. (During the 2007 TIMSS administration, there was a similar 21-point drop for all U.S. students from 4th to 8th grade, from 529 to 508.)

Now for the TIMSS science breakdown:

• In 4th grade, neither of the two states broken out separately can claim special bragging rights on the world stage. The U.S. average score was 544, and although Florida (545) technically scored higher, again, TIMSS doesn’t consider that measurably different from the U.S. as a whole. North Carolina was well behind the United States (528). South Korea led the pack (587) followed by Singapore (583).

• In 8th grade, by contrast, a few states earned bragging rights. Compared with the average U.S. score of 525, Massachusetts (567), Minnesota (553), and Colorado (542) all did better. So did Indiana, North Carolina, Connecticut, and Florida, but their scores weren’t considered measurably different from the U.S. score. Only Singapore (590) outperformed Massachusetts flying solo as a state.

• Once again, other countries have a much bigger share of top-shelf students in science. In the 4th grade, Singapore, in second place among average overall scores, had 33 percent of advanced students (the highest share), compared with just 14 percent in Florida and 12 percent in North Carolina. In the 8th grade, Singapore once again had the biggest share of top-flight students (40 percent), leaving the best state, Massachusetts (24 percent), and second-best-state, Minnesota (16 percent), well behind.

• As in math, U.S. performance dropped from 4th grade to 8th grade, in the case of science by 19 points. Florida’s score dropped by 15 points between the two grades, while North Carolina’s slipped by 6 points. As you can see from the 4th and 8th grade results for Singapore, drops between the grades aren’t automatic.

From a state perspective, here’s a phenomenon that will likely surprise virtually no one: huge gaps in scores based on students’ socioeconomic status. Let’s take Connecticut in 8th grade science on TIMSS as an example. In the student group where less than 10 percent of students were eligible for free or reduced-price lunch, students’ average score was 581, 49 points better than the state’s overall average score. By contrast, in the student group where 75 percent or more of students qualified for free or reduced-price lunch, the average score was 420, a whopping 112 points below the state’s overall average. So in purely statistical terms, being relatively poor had a bigger impact on students’ scores than did being relatively rich, at least compared with the state average.

A version of this news article first appeared in the State EdWatch blog.