Compared with their counterparts in traditional public schools, Pennsylvania charter school students are making similar progress in reading but faring worse in math, according to a Stanford University study supplying the latest data in the years-long debate over charter performance.
But within the findings released Tuesday by the Center for Research and Education Outcomes (CREDO) are significant variations — in particular with virtual charter schools, which enroll about 25 percent of Pennsylvania charter-school students, and have been found to post poor results.
The study said cyber charter results were “overwhelmingly negative," dragging down the performance of the charter sector as a whole.
“Any potential benefits of online schooling such as student mobility and flexibility in curriculum are drowned out by the negative impacts on academic growth of students enrolled in such schools,” the report said.
Cyber charter leaders pushed back, saying they serve many students who haven’t fared well in brick-and-mortar schools.
“When trying to push a square cyber student through a round methodology, it simply doesn’t work. Factors outside of test score data come fully into play when they come to the cyber environment,” said Michael Conti, CEO of Agora Cyber Charter. “We are successful because we don’t just see a child as another student behind a desk.”
The study also found variations based on geography. Urban charter school students — who formed a majority of the study — performed better in reading than traditional public school peers, gaining an additional 35 days of learning on average, while charter school students in suburbs, towns, and rural areas all fared worse than peers.
Mark Gleason, executive director of the Philadelphia School Partnership, a nonprofit that funds charters and other schools in the city, said only about 5,000 of Philadelphia’s 70,000 charter-school students were enrolled in cyber charters. He noted that the study found positive results in reading for black and Latino students, and students in poverty enrolled in brick-and-mortar charter schools.
“We should be acknowledging that charter schools have a long way to go, but also noting that for certain at-risk students, they provide opportunities that certainly don’t exist in a traditional school system that assigns students to schools based on where they live,” Gleason said.
But the study may fuel calls for changes to charter-school funding — a topic of continued debate in Pennsylvania. Gov. Tom Wolf intends to pursue “legislative reforms” to improve charter performance, a Department of Education spokesperson said.
School districts, many facing budget pressures as state funding fails to keep pace with rising expenses, pay charters based on enrolled students.
“The abysmal student performance in cyber charter schools and mixed performance, at best, in brick-and-mortar charter schools create questions about the value taxpayers receive for spending more than $1.6 billion each year on student tuition payments to charter schools,” said Susan Spicka, executive director of Education Voters PA, which advocates for funding for traditional public schools.
The study considered student performance on state standardized tests from 2013-14 through 2016-17. It compared charter school students with “virtual twins” in traditional public schools that sent students to those charters, based on demographics and past student test scores.
Overall, CREDO researchers said the findings reflected “little to no progress” for Pennsylvania’s charter-school sector since 2013, when the state’s charters performed worse than the typical charter nationally, Raymond said.
Yet some Pennsylvania charter schools are performing very well, researchers said, with 45 percent doing significantly better in reading than traditional public schools sending students to those charters. One-third of charters were not significantly different than those schools in reading, and 23 percent were significantly worse.
For math, about one-third of charters performed significantly better; one-third not significantly different; and one-third worse.
“You’ve got a real problem at the low end that really needs to be taken care of,” CREDO director Margaret Raymond said. She noted that the schools doing significantly worse weren’t only cyber charters.
But the cyber schools’ poor performance had a “dramatic” effect on the overall charter performance numbers, Raymond said. CREDO equates academic growth to days of learning. Compared with traditional public school peers, a Pennsylvania student enrolled in an online charter loses the annual equivalent of 106 days of learning in reading and 118 days in math, the study said.
In contrast, students attending brick-and-mortar charter schools gain the equivalent of 24 additional days of learning in reading, and fare similarly to traditional public school students in math, the study found.
Ana Meyers, executive director of the Pennsylvania Coalition for Public Charter Schools, said families that choose cyber charters “are not only focused on test scores, but are also looking to enable their children to learn in a flexible environment" where they feel safe and can work at their own pace.
While Pennsylvania’s charter law tasks school districts with authorizing charter schools, the state Department of Education authorizes cyber charters. Spokesman Rick Levis said the study “affirms the actions the [department] has taken to strengthen oversight and accountability of cyber and brick-and-mortar charter schools.”
The department has been slow to renew cyber charters, which are granted as five-year operating agreements. As of March, eight of Pennsylvania’s 15 cyber schools were operating under expired charters. The current number wasn’t immediately clear Tuesday.
Correction: Due to an error in the CREDO report, a previous version of this story misstated the number of additional learning days the study found brick-and-mortar charter students gained in reading.