Rule Change Is Mulled for School Tests
This article is from the archive of The New York Sun before the launch of its new website in 2022. The Sun has neither altered nor updated such articles but will seek to correct any errors, mis-categorizations or other problems introduced during transfer.

Responding to mounting concerns about the validity of a test considered the gold standard for measuring American students, federal officials are pushing for a change in how the test is administered.
The revision would set a single standard for how to decide which students are excluded from testing and which receive special accommodations, such as extra time or permission to take the test in a small group.
Right now, policies for administering the National Assessment of Educational Progress, known as the nation’s report card, vary widely between states and school districts. In New York City, school officials this year gave extra time to at least 21% of students who took NAEP’s fourth-grade reading test, while nationally just 5% of test takers got extra time.
Arguing that the discrepancies compromise the test’s purpose, officials in the group that oversees NAEP are pushing to eliminate them before the next time the test is administered, in 2009.
One board member, James Lanich, said not having a reliable standard prevents states and researchers from drawing lessons from the NAEP results. Without knowing for sure which states are performing the best, lessons on which policies to pursue are harder to grasp, he said.
Studies have shown that excluding students can unfairly inflate test scores, though the effects of accommodations are unclear.
“We have to get our house in order,” Mr. Lanich, who is the president of the nonprofit advocacy group California Business for Education Excellence, said. “We don’t have time. We have to improve student academic achievement across the land very rapidly in order to remain competitive.”
Concern over states’ diverging accommodation policies are nearly as old as the accommodations themselves, first offered in 1996. But demands for some kind of solution seem right now to be reaching a peak.
The most recent NAEP reports carry prominent warnings about the danger of comparing state scores if testing policies diverge, and, to stress the point, future reports will eliminate numerical rankings listing states in order of their test scores.
The federal official charged with reporting the test, Mark Schneider, has committed to creating a top-notch statistical Band-Aid solution — complicated models that approximate how states would have fared if all had followed the same policies — by the next two years.
One member of NAEP’s oversight board, the National Assessment Governing Board, Alan Friedman, said he has made it a personal campaign to persuade states to take up a standard set of testing conditions voluntarily. After Mr. Friedman spoke with New York’s testing director, David Abrams, Mr. Abrams was appointed to a policy task force at the governing board.
Other board members say they would like to go further. Minutes from several recent meetings dedicated to the issue indicate the members called for uniform national rules on testing conditions.
In interviews yesterday, several members acknowledged that standardization would be difficult if not illegal to enforce in federalist America. The federal government only made states’ administering of the NAEP test mandatory five years ago, and only for two grades in two subjects. As a consolation, the new requirement came with a concession: individual students can opt out of the test.
Forcing students with disabilities to take a test in a certain way — without the extra help of an hour longer, say, or a smaller group setting — could also be impossible. By law, these decisions are made on an individual basis, during conversations between parents and schoolteachers that become written into a pact known as the child’s Individual Education Plan.
“Just think about that happening all over the country in all varieties of ways, and you talk about somehow standardizing that — it’s a rather monumental task even to think it through,” the chairman of the governing board, Darvin Winick, said.
An oft-cited model for how the hurdle could be overcome is a July 2005 meeting in Des Moines, Iowa, where 45 state governors gathered to sign a compact agreeing to measure their high school graduation rates by a single standard.
Working through that kind of a voluntary, state-driven path on testing policies, Mr. Lanich said, might be a way to avoid legal battles.
A Kentucky researcher who has studied NAEP, Richard Innes, said some states might not want to sign up, for fear of abandoning generous accommodation policies that could inflate their scores.
Mr. Innes said he is suspicious that New York City’s high rate of accommodations suggests this city might be opposed to new requirements, too.
Disputing that claim as ludicrous, a Department of Education spokesman, Andrew Jacob, cited a recent study by the National Center for Education Statistics, the federal organization headed by Mr. Schneider.
The study found that students who received accommodations in New York City posted smaller gains this year on average than those who didn’t. That suggests that accommodating more students, rather than excluding them, actually caused the city’s average scores to drop. But if the students had taken the tests without accommodations, they could have done even worse, causing the city’s average scores to drop even more.