Study Finds No Clear Edge for Charter Schools
Students who won lotteries to attend charter middle schools performed, on average, no better in mathematics and reading than their peers who lost out in the random admissions process and enrolled in nearby regular public schools, according to a national study released today.
The federally commissioned study, involving 2,330 students who applied to 36 charter middle schools in 15 states, represents the first large-scale randomized trial of the effectiveness of charter schools across several states and rural, suburban, and urban locales. The charter schools in the sample conducted random lotteries for admissions, so that only chance determined who attended.
The study, conducted by Mathematica Policy Research, of Princeton, N.J., also concludes that the lottery winners did no better, on average, than the lottery losers on non-academic outcomes such as behavior and attendance.
The findings on academic performance echo, in part, those of researchers at Stanford University, whose 2009 nonrandomized, multistate study of charter schools sparked fierce debate when they concluded that, in general, most charters were producing similar or worse achievement results for students than traditional public schools were. ("Study Casts Doubt on Charter School Results," June 15, 2009.)
The Mathematica authors add context to the new study’s findings, however, by exploring when charters seem to work best, and for which students. They found, for instance, that the charter middle schools serving the most economically disadvantaged students—especially those in urban areas—were more successful than their counterparts serving higher-achieving, more affluent students in producing gains in mathematics.
That finding is similar to that of a recent study of 22 middle schools operated by the Knowledge Is Power Program, or KIPP, the nation’s largest charter-management organization. ("KIPP Middle Schools Found to Spur Learning Gains," June 22, 2010.) That study, also conducted by Mathematica, found that the KIPP students—most of whom were also poor and were members of ethnic- and racial-minority groups—outperformed their peers in regular public schools.
For other student subgroups—those defined by race, ethnicity, and gender—the new study found that the charters made no significant impacts on achievement.
People involved in the new study cautioned, though, against drawing any sweeping conclusions from it about the overall impact of charter schools on student achievement.
“There’s a wide variation in both math and reading impacts, as well as other measures,” said Marsha Silverberg, who oversaw the study as a project officer at the Institute of Education Sciences, the research arm of the U.S. Department of Education.
“Generally, we found that these charter schools were more effective for more low-income, lower-achieving students,” Ms. Silverberg said, “and less effective for higher-income, higher-achieving students.
“We are not suggesting that charter schools serving lower-income and lower-achieving students would always be more effective,” she said, “but that they were more effective than the traditional public schools around them.”
The most positive overall impact that all of the charter schools in the study produced, was on the satisfaction levels expressed by parents and students. Parents whose children had won lotteries to attend charters were 33 percent more likely to say the schools were excellent than parents whose children lost the lotteries and attended regular public schools.
The new study comes out on the same day that charter leaders and advocates will be gathering for an annual national conference in Chicago and follows closely on the heels of the KIPP study, which was released last week.
Debate over the impact of the nation’s 5,000 charter schools, which are publicly financed but largely autonomous, has been heightened over the past year with calls from President Barack Obama and U.S. Secretary of Education Arne Duncan to expand and replicate charters that have succeeded in raising achievement among the poorest students.
That debate grew more heated after the 2009 release of the multistate study from Stanford, which was produced by the Center for Research on Education Outcomes, or CREDO. ("Scholars Spar Over Research Methods Used to Evaluate Charters," October 14, 2009.) It was followed by findings of another Stanford researcher, Caroline M. Hoxby, who concluded the opposite. In her study, which also used lotteries to randomly assign students to either control or experimental schools, Ms. Hoxby found that charter schools in New York City were having positive impacts on student achievement.
“This is not going to settle the debate,” Jeffrey R. Henig, an education professor at Teachers College, Columbia University, said of the new study. “There are a lot of folks on both sides who are pretty committed to keeping the debate alive on these terms, but I think on the overall balance scale, this study adds weight to the side that is suggesting that simply talking about charters versus noncharters is a distraction. There needs to be much more nuance.”
The new study included only charter middle schools that had been in operation for at least two years, which Mr. Henig said would undermine past arguments from charter advocates who have said conclusions about the schools’ effectiveness shouldn’t be drawn until they have had a chance to mature.
Margaret E. Raymond, the director of CREDO and the lead author of the study that it released last year, said the results of this new study are “not surprising.”
“This is another layer of evidence that points to the wide variations in the charter school community and highlights, once again, that policy and context really matter,” she said. “I think what we really have to do now is get under the hood and find out more about how charters differ and why they differ. We are really just at the front gates of a large body of research that needs to be done.”
Researchers involved in the new study declined to disclose the states and communities where the studied schools are located, citing rules about anonymity for individual schools in federal research.
On average, though, the charter middle schools in the study enrolled a lower percentage of students who are eligible for free and reduced-price school meals than charters nationally, and served smaller percentages of students scoring below proficiency levels on state exams than their national peers.
Also, the percentage of African-American students who attended the charter middle schools in the study were smaller when compared with charters nationally.