NAEP Brings Good News and Bad News

Published October 1, 2001

The release of the 2000 math scores from the National Assessment of Educational Progress (NAEP), sometimes called the Nation’s Report Card, produced a flurry of good-news headlines during the summer doldrums, especially in states where scores were up from the 1996 sampling.

But realists like U.S. Secretary of Education Rod Paige quickly discouraged the popping of champagne corks.

“Despite this improvement,” remarked Paige, “it is clear that with only a quarter of our fourth- and eighth-graders performing at or above proficient levels on this exam, we are not doing well enough.”

On the positive side, the NAEP sampling showed that fourth-graders’ average score of 228 (on a 500-point scale) was 4 points higher than the 1996 testing and 15 points higher than 1990. In addition, 70 percent were at or above the “basic” level, a bare-bones ability to do simple computation, which compares with just 64 percent in 1996.

For the eighth-graders, the average score was 275, a gain of 3 points over 1996 and 12 points over 1990. For the 12th-graders, the average score was 301, down 3 points from 1996 but still 7 points over 1990. Two-thirds of both eighth- and 12th-graders scored “basic” or higher, an improvement of 4 percentage points over 1996.

The National Center for Education Statistics (NCES), which oversees NAEP, and Dr. Paige also handed out kudos to North Carolina, Maryland, Georgia, and Texas for doing more than other states to narrow the racial gap in achievement.

“Basic” Doesn’t Mean “Adequate”

But a deeper look at NAEP scores showed the Secretary was right to temper the enthusiasm.

For starters, the performance of the nation’s 12th-graders did not improve. In fact, it actually slipped since 1996, continuing a pattern, detected by other assessments, of performance falling in relative terms the longer students stay in school. The high school seniors’ average score of 301 points was down 3 points from 1996. Moreover, only 65 percent were at or above “basic,” a decline from 69 percent in 1996.

It is also important to remember that “basic” is not synonymous with “adequate.” It means only partial mastery. The next highest level–“proficient”–stands for solid academic performance. And only 14 percent of the 12th-graders reached the proficient level, while a scant 2 percent attained the top category of “advanced.”

Indeed, fewer than a fourth of students in all three grades–23 percent for fourth-graders, 22 percent for eighth-graders, and 14 percent for 12th-graders–scored high enough to be judged “proficient.”

Achievement Gap Unchanged

When the scores were broken down for minority groups, the results were even more distressing. Despite the federal government’s expenditure of tens of billions of dollars annually in compensatory aid for the express purpose of narrowing the racial achievement gap, it hasn’t budged over the past decade.

The situation for fourth-graders is typical: White pupils have remained exactly 31 points ahead of black pupils over the past 10 years. Even though blacks and Hispanic students have gained over the past decade–by 16 points and 14 points, respectively–the much-lamented gap has not narrowed.

Class Size Rhetoric

As shaky as the NAEP data are as proof of substantial progress, some education bureaucrats placed the numbers in a heavy spin cycle. For instance, the California Department of Education (CDE) claimed a 1-point narrowing between the state’s fourth-grade NAEP math scores and the national average since 1996 proves the state’s initiative to reduce class sizes in the cause of “reform” is working.

But Dr. Alan Bonsteel, the San Francisco physician who heads California Parents for Educational Choice, pointed out that because NAEP is only given every four years, the “improvement” claimed by the CDE “amounts to only one-quarter of an exam score point per year, a statistically insignificant improvement.”

Added Bonsteel: “At this rate, it would take California 48 years to reach the national average in mathematics. That national average is, of course, in any event not competitive in international comparisons of science and mathematics educational achievement.”

Class Size Reality

California fourth-graders got the “benefit” of the state’s deliberate effort to reduce class sizes, while eighth-graders did not. But the fourth-graders in those small classes finished second-to-last among the 40 states taking NAEP, ahead of only Mississippi. Meanwhile, California eighth-graders topped Mississippi, Arkansas, Louisiana, and New Mexico.

Among the Golden State’s fourth-graders in those vaunted smaller classes, both the white/black and white/Hispanic test-score gaps were worse than the national average. By contrast, the white/minority test-score gap for the eighth-graders in larger classes was better–that is, narrower–than the national average.

Bonsteel explains how politically motivated crash programs to reduce class sizes–initiatives endorsed by Republican and Democratic candidates alike in races across the country–actually can harm disadvantaged children.

“What went wrong in California is that in the spring of 1997, class sizes in kindergarten through third grade were reduced almost overnight from 30 students to 20. The need for 50 percent more teachers at that level meant hiring standards plummeted. That was especially a problem in inner-city, mostly minority neighborhoods, which saw their best teachers leave for the suburbs. California thus made the Faustian bargain of smaller class sizes often taught by weaker teachers, a problem that was especially devastating in minority communities.

“The unplanned, haphazard, and top-down imposition of reduced class sizes in schools that were already at capacity also often meant converting libraries to classrooms or holding two classes in one classroom, with a curtain hung in the middle to separate the two. At least one school held a reduced-size class in a converted janitor’s closet.”

Disabled Students Excluded

Finally, NAEP data are in question because the NCES did not report rates of exclusion of so-called disabled students from the testing. On the basis of preliminary investigation, testing gadfly Richard Innes of Kentucky believes several states’ scores were artificially inflated because of increased exclusions of disabled students, just has happened with the 1998 NAEP reading results.

Innes regards the worst offender as oft-touted North Carolina, which excluded 9 percent more of its eighth-grade sample in 2000. That 9 percent rise in testing exclusions could just about wipe out North Carolina’s reported 12 point rise in scores, if this factor were fairly taken into account, according to Innes.

Robert Holland is a senior fellow at the Lexington Institute, a public policy think tank in Arlington, Virginia. His email address is [email protected].