Ever since the School Choice Movement got momentum in the early 1990s, its proponents have claimed that charters and vouchers would “save poor kids from failing schools.” Their metric, of course, was scores on standardized test scores, and they welcomed No Child Left Behind and its successor Race to the Top. They were certain that choice schools—free to select their students, free to kick out students, free from bureaucracy, free from unions, free to pay differential pay to teachers—would prove their value by generating sky-high test scores.

There are some charter schools that get high scores, but most don’t. Most studies find that some charters get high scores, some get the same scores as nearby schools, and some are far worse than the so-called “failing schools.”

Recent voucher studies have converged on the finding that students who use vouchers actually lose ground as compared to their peers who won a voucher but didn’t use it. The more optimistic say that the voucher students make up the lost ground in 3-4 years, but they don’t take into account the attrition of the weakest students from the voucher schools.

A new paper by three school choice advocates concludes that test scores are not the best measure of success (whoa! Who knew?). Other long-term impacts, they say, matter more, like graduation rates. Why are they moving the goal posts? Voucher programs show no academic gains, but they do show higher graduation rates, so that’s what really matters. There is a trick here, however. Every voucher program has a high rate of attrition, which pro-choice researchers ignore or downplay. The “higher graduation rates” in evaluations of voucher programs in Milwaukee and D.C. do not acknowledge the high number of kids who started ninth grade and didn’t make it to the end of twelfth grade.

Patrick Wolf of the Department of Educational Reform at the University of Arkansas (funded primarily by the Walton Family Foundation) conducted the official evaluations of both Milwaukee and the District of Columbia. In his initial report about Milwaukee, he wrote that the attrition rate was 75%, but decided that was an error and revised the attrition rate to 56%. Either number is huge. Huge and huger. 

The survivors had a higher graduation rate than the students in the Milwaukee Public Schools, which included the kids who dropped out of the voucher schools.

Wolf’s D.C. evaluation does not break out the attrition rate, but it is likely to be significant. William Mathis of the National Education Policy Center reviewed Wolf’s Congressionally mandated evaluation of the D.C. voucher program but could not determine with certainty how many students had dropped out before graduating, but it appears to be nearly three-quarters.

All of this is background to Secretary Betsy DeVos’ nonchalant response to the latest [2017] negative evaluation of the D.C voucher program.  She never expected vouchers to raise test scores, she says. And it doesn’t matter.