EdSource's third evaluation of California charter schools largely affirms conclusions from its 2005 and 2006 studies (e.g. strong performance of middle schools, faster academic growth of classroom-based charters over virtual ones). But the distinguishing characteristic of the new analysis is the researchers' use of statistical regressions to delve deeper into observable trends. The report tackles charter school performance on two fronts: comparing charter and non-charter schools and analyzing differences in academic performance among types of charter schools (conversion, virtual, managed, etc.). All the usual data suspects are examined, including the state's Academic Performance Index, NCLB's Adequate Yearly Progress, and California's School Characteristics Index (SCI) (which scores schools based on composite information about English learner status, parental education, and student demographics). Yet, despite the wealth of data, results are mixed. Neither charter elementary nor high schools demonstrate big achievement differences from their non-charter peers, whereas charter middle schools outperform non-charter middle schools by a statistically significant and durable margin. Further, schools overseen by a management organization outperform those run independently. While not earth-shattering, this study is a solid addition to charter school research. See for yourself here.