Harvard University
In your balanced report on our study of school vouchers and college enrollment (“Study: Vouchers Linked to College-Going for Black Students,” Aug. 29, 2012; includes correction, Aug. 30, 2012), Christopher Lubienski is paraphrased in the corrected article as saying that "[t]he study shows that being offered a voucher increased the college-enrollment rate [of African-American students] by 7.1 percentage points, suggesting that the 8.7-percentage-point gain from actually using the voucher to attend private school caused only a 1.6 -percentage-point increase more than being offered a voucher.” Mr. Lubienski’s comment is incorrect because it implies that the effect of a voucher offer does not include the effect of using the voucher, when in fact it does.
The 7.1-percentage-point increase is the effect of being in the treatment group (offered a voucher) as compared with being in the control group (not offered a voucher). This effect includes the impact of using the voucher among students who used it (as 81 percent of African-American students did).
When we adjust the 7.1-percentage-point impact to reflect the fact that 81 percent, not 100 percent, of African-American students used the voucher, we calculate that the impact of using the voucher was 8.7 percentage points. Because the impact of the “offer” includes the impact of using the voucher, it doesn’t make sense to compare the 7.1- and 8.7-percentage-point estimates in the way that Mr. Lubienski does.
Matthew M. Chingos
Fellow, Brown Center on Education Policy
Brookings Institution
Washington, D.C.
Paul E. Peterson
Professor of Government
Director, Program on Education Policy and Governance
Harvard University
Cambridge, Mass.