Researchers Reprise Earlier Erroneous Crime-Cutting Voucher Verdict

Faced with recent research linking voucher receipt to decreased test scores, voucher advocates have been busily moving the goalposts. The most creative of these attempts is a new “working paper” from researchers from the University of Arkansas and the Cato Institute, which first notes that “[s]tandardized test scores…do not fully capture society’s goals for education” (p. 3) and then concludes that “[s]tudents who participated in the [Milwaukee voucher program] are less likely to commit drug and property crimes and experience paternity suits than their peers in [Milwaukee Public Schools], all else being equal” (p. 24).

Since most schools participating in choice programs are sectarian…and these religious schools teach students that God always and everywhere is watching and evaluating what they do, private schools of choice might be expected to improve the subsequent behavior of their charges (p. 6).

For those of us who remember the decades upon decades when voucher policies were sold a lifeboat to rescue students from public schools with low test scores (“failing public schools” was the preferred terminology), this shift in rationales is quite striking. The claim now is apparently that lower math scores in voucher schools are accompanied by a decrease in paternity suits, pickpockets and pharmaceutical peddling—an unusual trade-off but one that might be worthwhile.

Yet even that claim has not held up to scrutiny. An earlier version of this report was reviewed by Clive Belfield. Professor Belfield found that the report’s premise that vouchers might reduce crime, while plausible, could not be sustained due to the study’s unsupported causal assertions. Further, the magnitude of the correlations was so small that the study’s results could, in fact, have been used to substantiate the opposite interpretation. That is, the data and analyses suggest a reasonable finding that there is no meaningful relationship between vouchers and the measured outcomes.

Parents and students who actively engage in school choice differ educationally and economically from non-choosers. Active choosing is also often grounded in educational values and motivation. Repeating the errors from the earlier study, the experimental and control groups in the new study do not, and could not, address these differences—meaning that the comparison group of public school students is likely different in important ways from the voucher-receiving group. Given the small magnitude of the correlations, the sampling concerns, and the inadequate controls, we are left only with the suspect conclusion that the Milwaukee voucher program “reduces adult criminal convictions and paternity suits” 13 to 15 years after students attended, even if the intervention had no measured short-term benefits.

That’s pretty tenuous, but it will have to suffice until we are told that vouchers reduce acne or cure the common cold.