Federal Charter Study Sparks Heated Debate
A vigorous debate over charter schools unfolded in the nation’s capital on Dec. 15 as officials released the results of a federal pilot study of how students in the independently run but publicly financed schools perform on the tests commonly known as “the nation’s report card.”
The official results from the 2003 National Assessment of Educational Progress of 4th graders in reading and mathematics painted a picture of charter school performance that was roughly equivalent to the one presented in a highly publicized report based on the same basic data that was released in August by the American Federation of Teachers. In both cases, charter school students were generally found to be trailing their peers in regular public schools—at least in math—although that gap disappears once the race and ethnicity of the students are taken into account.
One notable difference between the official report and the AFT analysis was the results for reading. While the AFT study found 4th graders trailing their counterparts in regular public schools by a statistically significant margin, the report released by the National Assessment Governing Board found no measurable difference in those reading results.
Reflecting the deeply polarized national debate over charter schools, the actual results of the report—which was prepared by the statistics arm of the U.S. Department of Education but released by the bipartisan, independent NAGB—were eclipsed by the pointed rhetoric of supporters and critics of charter schools.
Offering an admittedly pro-charter take on the results was the No. 2 official at the Education Department, outgoing Deputy Secretary Eugene W. Hickok.
“While the study points out some differences,” he said, “it also points out that charter schools are holding their own with regular public schools. “ A far different interpretation came from Bella Rosenberg, an assistant to the president of the AFT and a co-author of the August report issued by the union. She was invited by NAGB to present her views following the report’s release, along with Jeanne Allen, the president of the pro-charter Center for Education Reform, based in Washington.
Echoing Mr. Hickok’s remarks, Ms. Allen characterized the results as showing that charter schools, despite formidable political obstacles and funding disadvantages, are “in a statistical dead heat” with regular public schools.
Ms. Rosenberg, by contrast, said the results clearly showed that “most charter schools are doing worse than our much-maligned regular public schools. If charter schools are supposed to rescue children trapped in regular public schools, who will rescue children attending demonstrably worse charter schools?”
Positive or Negative Spin?
Offering additional data not included in the AFT report, the NAGB study—which examined data for about 6,500 charter school students and 376,000 regular public school students—found that charter school children taught by teachers with four years of experience or less performed significantly worse in both reading and math than students in regular public schools with similarly inexperienced teachers. And students at charter schools were far more likely to have such teachers, the study found.
The NAGB study also found that whether a teacher had a standard teaching license did not make a difference in student performance. It did find, however, that far more teachers in charter schools lacked regular teaching credentials.
In summarizing its findings from the special NAEP study, NAGB stressed that a higher proportion of students in charter schools were African-American and that a higher percentage of the charter schools were in central cities. On the whole, such students and schools tend to post lower scores on standardized tests.
Among test-takers in math, for example, 31 percent of charter school students were black, compared with 17 percent in regular public schools. And while half of those charter students were in urban schools, just 29 percent of the students in regular public schools were in central-city locations.
The governing board also stressed that the performance of white, black, and Hispanic 4th graders in charter schools was not “measurably different” from that of students in regular public schools with similar racial and ethnic compositions.
But Ms. Rosenberg suggested that the assessment governing board was putting an overly positive spin on the results, in part by downplaying the finding that students from low-income families in charter schools performed worse than such students in regular public schools. “I would have wished for a more neutral NAEP report,” she remarked.
Overall, results from the special study found that in math, 69 percent of 4th graders in charter schools attained at least a basic level of proficiency on NAEP, compared with 76 percent of students at regular public schools. NAGB determined that the gap between those percentages met its standard for a statistically reliable difference.
In reading, on the other hand, the governing board determined that the lower scores posted by charter school 4th graders did not denote a measurable difference. The study found that in reading, 58 percent of students in charter schools scored at least at the basic level, while 62 percent of students in regular public schools did.
Ms. Rosenberg said the advantage in reading proficiency among students in regular public schools did become statistically significant once special education students were excluded from the analysis.
Tracking Poverty’s Impact
The study also found that among students who qualified for federally subsidized lunches, charter school students scored lower both in reading and math than those enrolled in noncharter schools.
Students with disabilities were also found to score lower in both reading and math than those in regular schools. The same disparities did not occur, however, among students with limited English proficiency.
The report found that the 150 charter schools included in its nationally representative sample appeared to serve a slightly smaller percentage of students from low-income families, although NAGB did not deem those differences to be statistically significant. Charter schools also had slightly smaller percentages of students with disabilities among students tested in math, but not in reading.
Ms. Rosenberg contended that those statistics and others gave a counterpoint to assertions that charter schools generally serve the “most disadvantaged of the disadvantaged.”
But Mr. Hickok and Ms. Allen countered that charter schools often take children who were struggling in regular public schools, and therefore had disadvantages beyond those suggested by demographic data.
The NAGB report was released just a day after a study by Harvard University economics professor Caroline M. Hoxby, who examined charter school achievement not as measured by the federal NAEP, but by state tests.
Comparing charter schools with nearby “matched” schools, Ms. Hoxby found that students in charter schools nationwide were 5.2 percent more likely to be proficient on their state exams in reading and 3.2 percent more likely in math. She also found that charter students did better the longer their schools had been up and running. Ms. Hoxby had released preliminary results of her study in the early fall, shortly after the AFT report on the NAEP data. Charter critics poked holes in the study, even as charter supporters seized on it as positive news.
That pattern continued at the release of the NAEP study, as Ms. Allen pointed to Ms. Hoxby’s study as painting a clearer picture of charter performance, and Ms. Rosenberg found fault with its methodology.