View all Articles
Commentary By Jay P. Greene

Choice Battle Needs A Real Truce

Education, Cities, Economics Pre K-12, Finance

Anneliese Dickman’s critique of my study on graduation rates in Milwaukee’s public schools and voucher program (“How about a truce in the school choice propaganda war?” Nov. 14) is remarkably devoid of substantive criticisms for a person attempting to position herself and her organization as “dedicated to non-partisan analysis” that “places a premium on objective research.”

She has only one substantive criticism of the results to offer: My method for estimating graduation rates does not properly reflect the transfer of students into and out of the schools studied.

While my method does not follow individual students from 9th grade to graduation or dropping out, that does not mean that it fails to account for student transfers. By tracking a cohort of students over time, my method can produce a reasonable estimate of graduation rates as long as students transferring out are roughly equal to students transferring in.

For Dickman’s re-estimate of the choice program’s graduation rate of 41% to be accurate, there would have to have been a huge imbalance of in-transfers relative to out-transfers during the high school grades. There is no evidence to support such a claim nor does she attempt to present any.

In addition, I did not dismiss the possibility that my estimate was biased by the transfer of students as casually as Ms. Dickman suggests.

Instead, I described this concern in my report as “the primary source of uncertainty in these estimates.” One of the ways I addressed this concern was to check the precision of my estimate by using an alternative method developed by the Harvard Civil Rights Project and Urban Institute to see if it would produce different results.

It did not. The Harvard/Urban Institute method produced results that were nearly identical to my own.

One might also have expected Dickman’s criticism of estimating graduation rates by tracking cohorts rather than individual students to have been muted given that my method has been widely cited and endorsed by ideologically diverse organizations, including the Education Trust, Education Week’s Quality Counts and the Gates Foundation. Unless Dickman believes that these organizations, as well as the Harvard Civil Rights Project and the Urban Institute, have joined the conspiracy to engage in “propaganda” supporting Milwaukee’s voucher program, the evidence clearly shows that students in Milwaukee’s voucher program graduate high school at a much higher rate than students in the Milwaukee Public Schools.

I fully acknowledge here, as I did in the report, that my study of graduation rates in Milwaukee “does not provide a definitive evaluation of the city’s choice program.” Instead, as I wrote in the report, “a more definite evaluation would require longitudinal data for individual students and employ a random-assignment research design to control for the selection of students into the choice program.”

In short, I believe that Dickman and I are in complete agreement about the need for a state-sponsored evaluation along the lines passed by the Wisconsin Legislature but vetoed by Gov. Jim Doyle.

Given that our recommendations are identical and her one substantive criticism is mistaken, it is unclear what motivated Dickman to launch her attack, particularly with its gratuitously nasty tone.

If she wishes to reduce the level of ideological acrimony in public debate of Milwaukee’s voucher program, she might consider looking in the mirror.