A recent article in the Huffington Post badly distorts the findings of new study of Michigan’s charter public schools by Stanford University’s Center for Research on Education Outcomes, and in the process blames charter public schools for serving too many poor kids.
Highlighted in the article is an observation contained in the study that 80 percent of charter public schools in Michigan performed below the state average in reading and 84 percent were below average in math.
While accurate, this is one of the study’s least meaningful statistics, because it is does not consider the impact poverty has on standardized test scores. Using this statistic to claim charter public schools in Michigan are of poor quality blames the victim.
Based on the well-established relationship between test scores and student poverty, one should expect most Michigan’s charter public schools to score below the state average, since they serve a higher portion of poor students. The CREDO study says 70 percent of charter public school students qualify for a free or reduced-price lunch compared to 43 percent in conventional public schools.
Further, this statistic also does not consider learning gains made over time by students in charter public schools. The CREDO study measured that too and found that 82 percent of Michigan charter schools had positive average growth in reading and 72 percent did the same in math.
Other results from the study that measure learning growth over time and account for the impact of poverty are almost entirely positive. For instance, 98 percent of charter public schools did the same or better than conventional public schools in reading, and in math the same is true for 94 percent of charter public schools. Of the 56 subgroups of students and schools the CREDO study analyzed, 52 showed positive learning gains for charter public school students.
To be fair, the article does briefly reference one such finding — charter school students averaged making academic gains in reading and math equivalent to about two months of learning compared to their demographically similar peers in conventional schools.
But the piece quickly tries to downplay these findings by calling out the study’s “limitations.” For example, it cautions that the study did not “measure most high school students.” Actually, it measured none and for good reason: Michigan only tests high school students in the 11th grade — annual learning gains were impossible to measure.
Despite this, CREDO’s study is still the largest and most comprehensive one ever done of Michigan’s charter public schools. It analyzed the individual performance of more than 85,000 different charter public school students over a five-year period.
The article also warns that the study did not use the most recent data available and implies that charter schools are likely doing worse since Michigan removed its artificial cap on the number of charters. It also claims that this legislation “deregulated” charter public schools and removed accountability requirements.
This criticism is speculative, at best. The law that lifted the cap on charters added dozens of new reporting requirements for charter public schools and did not go into effect until March of 2012. No data are available to measure charter public school performance since Michigan lifted the cap. Only about 30 new charters have opened since then, about 10 percent the number of schools CREDO studied. Even if every one of these new schools were terrible, it wouldn’t change the overwhelmingly positive results from the new CREDO study.
To be sure, all social science research has inevitable limitations. Downplaying the results of the most rigorous study of Michigan’s charter public schools to date due to its natural limitations is unfair. Not all charter public schools in Michigan are working miracles, but there’s no “cautionary tale” to tell here. In fact, other states might do well to emulate Michigan, as in CREDO’s own words: “These findings position Michigan among the highest performing charter school states CREDO has studied to date.”
Michael Van Beek is director of education policy at the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland, Mich. Permission to reprint in whole or in part is hereby granted, provided that the author and the Center are properly cited.