As regular readers know, I have concerns about the General Assembly’s zeal for charter schools, which I consider a quick fix approach to education based largely on renaming schools rather than reforming them.
I like charter schools; I just don’t think they are the answer to under-performing schools. Nor are they a surrogate for the challenging work of improving teacher quality.
And here’s a new federal study — the first large-scale randomized trial of the effectiveness of charter schools in multiple states and types of communities — that underscores my concerns that lawmakers believe that turning schools into charters is all they need to do to improve education in Georgia. (Legislators claim that they have other reform ideas, but take a look at the significant legislation of the past four years.)
This study has been long awaited and will spark a lot of debate. Let’s start it here.
Many years ago, when charters were just appearing on the education horizon, I attended a program at UNC where a speaker predicted that the most likely benefit of charter schools would be higher parental satisfaction because parents chose the school and thus were more likely to see positive effects, even if the academics were not better. That led to a spirited discussion among the reporters over whether parental satisfaction, in the absence of measurable improvements in key areas, was enough to deem a reform successful.
Our view was that it was not. I still feel that way. There has to be more reason to support charters than parents feel good about them.
I recall one of the deans of Southern education reporting talking about how he found parents were often pleased with their school and their children’s teachers, even when the school was under performing.
I also want to note that the comment in this press statement — that while charters on average don’t outperform their traditional counterparts, there are wide variations among charter schools — is true of all schools.
From the Institute of Education Sciences of the U.S. DOE:
Charter middle schools popular enough to hold admission lotteries are, on average, no more successful than nearby traditional public schools in boosting student achievement, behavior, and school attendance according to a new evaluation released today by the Institute of Education Sciences. However, charter schools vary widely ‐‐ some are more effective and others less effective than nearby traditional public schools.
Those located in large urban areas and those serving disadvantaged students are the most successful. “This study adds to a growing body of evidence on this important policy issue,” said IES Director John Easton. “We examined academic progress, but we also dug deeper to try to understand more bout the variability of charter school outcomes and why some are more or less effective than traditional public schools.”
The Evaluation of the Impact of Charter Schools was conducted with 2,330 students who applied to 36 charter middle schools that held lotteries for admission. The study was directed by the National Center for Education Evaluation within IES and conducted by Mathematica Policy Research and its partners.
The study focuses on students who attended charter middle schools, usually grades 5 through 8, and who attended a school in operation two years or more. These schools had to be popular enough have a lottery because that allowed researchers to compare two similar groups of students, one offered admission to the charter they applied to and one not offered admission. On average, the participating charter middle schools served more advantaged students than other charter middle schools nationally.
The study charter schools had lower numbers of students eligible for free or reduced‐price lunch (44 percent to 62 percent nationally) and smaller percentages of students below proficiency on state assessments when they applied to the charter school (34 percent versus 49 percent in math). Additionally, fewer African‐American students attended the study charter schools than other charter schools nationally (16 percent versus 29 percent).
In each charter school, impacts were estimated by comparing average achievement outcomes among lottery winners with those of lottery losers over the 2 years following the lottery. Researchers compared performance on state math and reading tests, but because the tests varied by grade and from state to state, the scores were converted to a comparable scale.
Key findings from the evaluation include:
On average, study charter schools did not have a statistically significant impact on student achievement. However, the averages mask wide variation across the charter schools in how well their lottery winners performed relative to the lottery losers, who typically went back to their neighborhood schools.
Study charter schools did not significantly affect most other outcomes examined, except for parent and student satisfaction. These outcomes included absences, suspensions, and other measures of performance, as well as survey‐based measures of effort in school, student well‐being, behavior and attitudes, and parental involvement. However, lottery winners were 12 percentage points more likely and their parents were 33 percentage points more likely to say they were more satisfied with their schools than lottery losers.
Study charter schools were more effective for lower-income and lower achieving students and less effective for higher-income and higher-achieving students. On average, lottery winners with initial low test scores and lottery winners from low‐income families benefited academically from admission to charter schools (in math) while their more advantaged counterparts did not. However, there were no significant differences in charter school impacts for other student subgroups—such as those defined by race, ethnicity, and gender.
The variation in student achievement impacts among charter schools may be related to certain school characteristics. Charter schools in large urban areas, those serving more lower‐income or more lower‐achieving students produced positive impacts on student math scores relative to other nearby school options. Charter schools outside of large urban areas, those serving fewer low‐income students, and those serving higher‐achieving students had negative impacts on test scores. Less negative impacts were found in smaller charter schools and those more likely to use ability grouping.