Do Voucher Students Attain Higher Levels of Education? Evidence From the Milwaukee Parental Choice Program

• In Event: Moscone Center West Roundtable Session 9
  In Roundtable Session: Making Sense out of Recent Mixed Results From Private School Voucher Evaluations

Abstract

Objective. Educational attainment is highly predictive of future success in life (e.g. Belfield & Levin, 2007; Lleras-Muney, 2005). We conduct a quasi-experimental analysis of the effects of participation in the oldest private school voucher program in the U.S. on various attainment benchmarks.
Theoretical Framework. Access to private schools through tuition vouchers could increase attainment for participants because private schools focus more heavily than public schools on shaping the non-cognitive habits and behaviors of students, such as grit, conscientiousness, and personal responsibility (Nichols, 2017). Vouchers also might boost attainment because private schools enroll more advantaged peers than public schools (Epple & Romano, 1998; Hsieh & Urquiola, 2006). Vouchers might have negative effects on attainment due to the disruption caused by school-switching and the fact that private high schools are less likely that public high schools to have guidance counselors and other student supports (Wolf et al., 2010).
Methods. In 2006, researchers drew a representative sample of students in the Milwaukee Parental Choice Program in grades 3-8, plus the entire census of 9th grade voucher students, and matched them 1:1 to Milwaukee Public School (MPS) students on grade, neighborhood, baseline test scores, and other student background characteristics. Twelve years later, the rates of high school graduation and college outcomes for the two groups were compared, adjusting for student background factors as well as family factors (when available) using probit statistical regression techniques.
Data. Data came from the administrative records of the private school voucher program, MPS, parent telephone surveys, and the National Student Clearinghouse on College Enrollments. Data on student baseline characteristics and outcomes were available for 100% of both the matched 9th grade sample (1,601 observations) and the matched 3rd-8th grade sample (3,852 observations). Family background data were available for 54% of the 9th grade sample (863 observations).
Results. Controlling for both student and parent factors, voucher students in the 9th grade sample were 7 percentage points more likely to graduate from high school, 6 percentage points more likely to ever enroll in a four-year college, and 7 percentage points more likely to complete a full year of college than their matched MPS peers. College graduation rates were statistically similar for the two groups. Controlling only for student background factors, voucher students in the 3rd-8th grade sample were 5 percentage points more likely to ever enroll in a four-year college, 5 percentage points more likely to complete a full year of college, and 3 percentage points more likely to graduate from college than their matched MPS peers.
Scientific or scholarly significance. This study demonstrates both the promise and limitations of private school vouchers as a policy to promote educational attainment. Although some of the findings are positive, not all of them are robust across samples and alternative statistical models. More research on the topic is needed.

Authors

• John F. Witte, University of Wisconsin - Madison

• Brian Kisida, University of Missouri

• Patrick J. Wolf, University of Arkansas