State's reading success doubted

U.S. says test scores exaggerate pupil gains

May 15, 1999|By Mike Bowler | Mike Bowler,SUN STAFF

Maryland gains in last year's national reading test were inflated because of an increase in the number of special education pupils excluded from the testing pool, the U.S. Education Department said yesterday.

But federal and state officials said overall results of the National Assessment of Educational Progress (NAEP) were valid and that no scores or rankings will be officially changed.

Nine percent of Maryland fourth-graders were excluded from the NAEP test in 1998, up from 7 percent in 1994. Between the two tests, the state's average reading score increased from 210 to 215 on a scale of 500.

As a result, analysts at the National Center for Educational Statistics say, fourth-grade increases in Maryland -- and in Kentucky, another state where more fourth-graders were excluded than in 1994 -- are "average" rather than "statistically significant," the term they used when the test results were released in March. They declined to release adjusted scores.

"It's an interesting debate for statisticians," said Mark Moody, Maryland's state testing chief, "but our real concern is not a few points on a scale of 500. It's our level of performance.

"This is like an argument over how many angels fit on the head of a pin. It won't make any difference to us in how we interpret how well our students are doing in reading or what we have to do to improve performance."

On last year's NAEP reading test, 39 percent of Maryland fourth-graders scored below the "basic" level, defined as partial mastery of the knowledge and skills of good readers. Marilyn McMillen, chief statistician for the National Center for Educational Statistics, said the analysis was the first of several designed to determine whether increases in NAEP exclusions in many states are playing a role in improved scores.

Because states use NAEP scores as a source of bragging rights, the test's validity has come under scrutiny.

Pascal D. Forgione, U.S. commissioner of education statistics, said the analysis found a "moderate positive correlation" between score increases in most states and increases in the percentage of students excluded.

"We have been woken up on the issue that we must monitor this participation rate," Forgione said. "If it keeps going like it is, the increase could be statistically significant."

Moody noted that NAEP officials choose the schools and pupils who participate on a sample basis in each state, not state or local officials. The only pupils excluded by local educators have "seriously impaired cognitive functioning" or lack proficiency in English.

About 2,200 Maryland fourth-graders were in the NAEP sample last year, and about 200 were excluded from the sample. The state has about 65,000 fourth-graders.

"We don't have a darn thing to do with who takes the test or who is excluded," said Moody, "though we get blamed for manipulating it."

The federal analysis found that more than half of the 36 states where NAEP is administered excluded higher percentages of special education students last year, while five excluded more non-English-speaking students.

Connecticut, which had the highest scores in the nation, left out 6 percent in 1994 and 10 percent four years later.

The Associated Press contributed to this article.

Pub Date: 5/15/99

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.