Urban 8th Graders Make Reading Gains on NAEP

Includes updates and/or revisions.

Eighth graders in large cities posted small gains in reading over the past two years, though urban 4th graders failed to show any improvement deemed statistically significant, according to national test data released today.

And while the 8th grade gains slightly outpaced the growth seen for the nation as a whole since 2007, both urban 4th and 8th graders still trail the country’s average student performance by considerable margins, based on the 2009 reading results for urban districts on the National Assessment of Educational Progress.

The analysis also includes specific results for a batch of districts that volunteered to participate. Only two of the 11 urban school systems that have taken part in that effort since 2007—Atlanta and Los Angeles—showed reading gains for 8th graders over the past two years that were statistically meaningful. If charter schools are removed from the comparison, however, the District of Columbia school system also would have statistically meaningful gains in 8th grade reading.

Meanwhile, four of the 11 districts saw 4th grade increases that were statistically significant: Boston, the District of Columbia, Houston, and New York City.

At a press conference in Atlanta to roll out the new NAEP results, Michael Casserly, the executive director of the Washington-based Council of the Great City Schools, said the results for urban 8th graders were “encouraging” but drew special attention to what he sees as particularly impressive gains over the long haul in 4th grade reading, which have moved up more rapidly than the national average.

Since 2002, the average scores of urban 4th graders have risen by 8 points, compared with a 3-point gain for students nationwide.

“We now are just 10 scale score points away from the national average,” he said. “We are not only improving, we are catching up.”

Daria L. Hall, the director of K-12 policy at the Education Trust, a Washington-based research and advocacy group, said there’s no question that the gains are “good news,” but she lamented that “we have districts like Detroit and Cleveland and Milwaukee and Los Angeles where 60 percent or more of students are ‘below basic’ in 4th grade literacy. That’s a tragedy for those communities, but it also has to be a wake-up call.”

The new findings come from the Trial Urban District Assessment, a specially collected set of test results from NAEP, known as “the nation’s report card.” The trial assessment provides a detailed look at the performance of students in cities with populations of 250,000 or more, as well as a separate look at representative samples of 4th and 8th graders in 18 urban districts that took part in 2009. Eleven of those districts participated in earlier assessment years and so have comparative data.

The report notes that large cities in general have a significantly different demographic makeup from the nation as a whole, with much larger proportions of minority students, those from low-income families, and English-language learners.

Digging Out of a Hole

Overall, the average reading score for 8th graders in big-city systems went from 250 to 252 over the past two years on a 500-point scale. Since 2002, the score has increased from 250 to 252, which is not considered statistically significant. For 4th grade, the average score for urban students went from 208 to 210; that growth also was not considered statistically significant. However, 4th grade scores grew from 202 in 2002 to 210 in 2009 for those students.

For U.S. students overall, the average 8th grade score was 262 in 2009, up from 261 two years earlier. In 4th grade, it remained flat, at 220.

Among the individual urban districts that took part in the trial assessment, the school systems in Houston and the District of Columbia tied for posting the strongest gains in 4th grade reading since 2007, as both climbed by 6 points, with average scores of 211 and 203, respectively.

At the 8th grade, the District of Columbia school system also saw a gain of 4 points, if charter schools were excluded from its 2007 score. In 2009 for the first time, any charter schools not counted as part of a district’s calculation for adequate yearly progress on the federal No Child Left Behind Act were excluded from the NAEP district analysis. As a result, none of the District of Columbia’s charter schools was factored in.

Arnold A. Goldstein, a program director in the assessment division of the National Center for Education Statistics, said the results for the District of Columbia system represented the only situation in which removing charter schools made a difference in showing a statistically significant change.

For its part, the Los Angeles school district had a 3-point gain in 8th grade. But the strongest improvement was seen by Atlanta, with its average score rising by 5 points, to 250. Atlanta’s results are up 14 points from 2002, more growth than for any other participating urban district. In the 4th grade, Atlanta has also seen gains of 14 points since 2002, though the 2-point increase from 2007 was not statistically significant.

At the NAEP press conference, Atlanta schools Superintendent Beverly L. Hall said that changes over time in her district have made a big difference.

“While our work is far from finished, our comprehensive reform agenda continues to deliver consistent and meaningful results in student achievement,” she said. Among the strategies she highlighted were “standards-based instruction,” using data to identify students’ “strengths and needs,” and intensive professional development and support for teachers.

“Our students are still below the national and state averages, but what the results do show is our students are digging out of a deep hole, and they’re doing so at a significantly fast rate.”

Gary W. Phillips, a vice president and a chief scientist at the American Institutes of Research, in Washington, said that as the NAEP data drill down to the level of individual school districts, it begins to highlight worrisome “disparities” in the U.S. educational system.

“We’re now getting into school systems where there was no one that was ‘advanced’ and almost no one is ‘proficient,’ like Cleveland and Detroit,” he said. “They’re at least four years behind their peers.” Advanced and proficient are two of the categories used to determine students’ mastery of the content.

Mr. Phillips said the NAEP results show that the typical 8th grader in Fresno, Calif., Houston, and Milwaukee is reading at the equivalent of the 4th grade proficiency level on NAEP.

“This is important data ... because it gives policymakers a better feel for what are the disparities and inquities,” he said.

Meanwhile, of the 18 districts that took part in 2009, Austin, Texas, and Miami-Dade County, Fla., had the highest average scores for 8th graders, at 261, and the Charlotte-Mecklenburg, N.C., district posted the strongest 4th grade average, at 225.

At the other end of the spectrum, the Detroit school system came in last in both the 8th grade, with an average score of 232, and the 4th grade, at 187.

“We are deeply and profoundly troubled by the performance in Detroit,” said the Council of the Great City Schools’ Mr. Casserly.

Tom Loveless, a senior scholar at the Brookings Institution, a Washington think tank, offered some cautions about making direct comparisons between the performance of individual districts on NAEP. For one, he said, the exclusion rates for English-language learners and students with disabilities differ considerably across the 18 districts.

Also, he noted that there appears to be a close relationship between the poverty rate of a district’s student population and its performance on NAEP.

In fact, the five districts with the highest NAEP reading scores for 8th graders in 2009 were also the five with the lowest student-poverty rates, as measured by free- and reduced-price lunch count, though they did not match up in rank order.

For instance, Charlotte-Mecklenburg, had the lowest poverty rate for 8th graders of all 18 participating districts, at 46 percent, and Austin’s poverty rate was tied for the second-lowest, at 54 percent.

“Just in terms of the rank ordering of the districts, it’s highly correlated with their free- and reduced lunch [counts],” Mr. Loveless said. “You have to take demographics into consideration.”