• Jeff tries to explain how government works to readers of Science.

Knowing More -- And Less -- About Science

10 May 2012 10:00 am


Test results. The latest data on what U.S. middle school students know about science raises many questions.

U.S. eighth graders did slightly better last year on a national science test than did their counterparts in 2009. But what that result says about the state of science in U.S. schools is open to debate.

A 2-point rise to 152 (on a scale from 0 to 300) is part of what Jack Buckley, head of the National Center for Education Statistics, calls the "uniformly positive" results from the 2011 Science National Assessment of Educational Progress (NAEP) at Grade 8. The pronounced racial gap in scores narrowed by a small but significant amount, says Buckley, from 36 points to 35 points for white students compared with black students, and from 30 to 27 for white students compared with Hispanic students. And all three groups did better. At the same time, he notes that the gap in scores between boys and girls grew from 4 to 5 points.

However, some science educators strongly disagree with Buckley's self-declared "optimism" that things are moving in the right direction. "It's pretty hard to get excited about these results," says Gerald Wheeler, interim executive director of the National Science Teachers Association. "It's like when a student who is flunking every subject finally comes home with a D."

For starters, there's the fact that 70% of middle school students possess no more than a "basic" understanding of what they need to know to do well in the subject, and only 2% perform at an "advanced" level. "I'll leave it to the politicians to decide how alarmed we should be that only 2% of our students are performing at an advanced level," says Buckley.

Speaking as an advocate for teachers and students, Wheeler is plenty alarmed. "The numbers may be moving in the right direction," he says. "But they are far, far below what we need to prepare a 21st century workforce. We've backed off on our commitment to science in the classroom and to providing adequate resources for teachers."

Wheeler acknowledges that there are "pockets of excellence"—seven states's scores topped 160, led by North Dakota's 164. But he notes that only 16 states improved on their 2009 performance. And some of the latest scores are truly abysmal: Students in the District of Columbia, which did not participate in 2009, posted an average score of 111. Mississippi has the lowest score of any state, at 137, although that number is up 5 points from 2009.

The NAEP results offer an intriguing hint of what providing more resources could accomplish. Students whose teachers report that they engage in hands-on science experiments "almost every day" scored 16 points higher than those who "never or hardly ever" do such activities. That finding "does not show causality," says Buckley. The students' performance, he notes, could also be due to attending a school with more resources, better teachers, and a stronger overall science program.

As is almost always true for U.S. education, what seems to matter most in a student's performance is his or her socioeconomic status. Poorer students, as measured by those eligible for free or reduced-priced meals, earned an average score of 137, while those not eligible averaged 164. Put another way, some 75% of students with scores in the lowest quartile (below 131) were eligible for the free meals, while only 21% of the poorer students scored in the top quartile (above 176).

The same disparity is true about race, which is often a proxy for economic status. Although white students comprised 55% of the representative sample of 122,000 middle school students who took portions of the test, they make up 76% of those scoring above 176. Only 4% of black students performed at that level, despite their 15% share of the total participant pool. Hispanics also underperformed, with 10% of them in the top quartile despite comprising 21% share of the total pool. The racial makeup of those with the worst scores is also skewed-35% Hispanic, 31% black, and 27% white.

The science component of NAEP is normally given every 4 years, in rotation with math and reading. Its schedule was altered so that NAEP results could be synched with the latest version of the Trends in International Mathematics and Science Study's international test of what fourth and eighth graders know about math and science, which is due out in December.

Posted In: