Cherry-Picking Charter School Research

As the Legislature debates lifting the arbitrary cap on the number of charter public schools that state universities can authorize, those who oppose expanding parental choice are arguing that charters do not perform well on average. Their arguments often focus on a small slice of cherry-picked research that runs counter to the much larger body of evidence showing good results from charters.

Specifically, they often cite a 2009 Stanford study that suggests only 17 percent of charters out-performed conventional schools, 46 percent did about the same and 37 percent did worse. Although no Michigan charters were included, this study has become a mainstay for defenders of the conventional school status quo here, including competition-fearing superintendents and school boards.

But this is just one of many charter school studies. Researchers at the University of California at San Diego recently analyzed more than 30 that met rigorous methodological requirements. This “meta-analysis” found that on average charters have statistically significant positive effects on elementary students in both reading and math. The same held true for middle school math students. In middle school reading and overall high school achievement, no significant effects were found from charters in either direction.

These meta-findings included the results of the aforementioned Stanford study. Interestingly, when the authors ran the numbers with that single negative finding excluded, charters were shown to generate statistically significant positive results across all grade levels. This suggests that the Stanford study is indeed an outlier.

Some opponents ignore the research altogether, and disparage charters by pointing out that a disproportionate number of them are performing below the state average. Such a result is perfectly consistent with the reality that most Michigan charters operate in areas where students have historically performed below state averages, and nearly two-thirds of charter students come from households whose low incomes qualify them for free lunch programs.

Ignoring this critical context nears to outright dishonesty. According to the Michigan Association of Public School Academies, a more apples-to-apples comparison to conventional districts whose students share similar socio-economic characteristics to those in charters shows Michigan charters outperforming districts on standardized tests 89 percent of the time.

Debates over the effectiveness of charter schools in improving standardized test scores for students should never revolve around either a single study or snapshot statistics.

Permission to reprint this blog post in whole or in part is hereby granted, provided that the author (or authors) and the Mackinac Center for Public Policy are properly cited.