Schools are expected to help shape the character skills needed to live a good life and contribute to society. In theory, programs that help families send their children to a private school could improve character skills through exposure to peers and school cultures that discourage risky behaviors. Do private school choice programs actually help achieve this goal? Our just-released evaluation of the longest-running modern school voucher program in the United States—the Milwaukee Parental Choice Program (MPCP)—suggests that they do.
The study updates and extends our previous work on the same program with new data, including information on students' criminal activity and involvement in paternity disputes as young adults. We gathered these data in fall 2018, when the 2,178 students we have been following in a longitudinal evaluation of the MPCP were roughly 25 to 28 years old. Half of these students were using vouchers to attend a private school in 8th or 9th grade in 2006; the rest are a carefully matched sample of students who attended Milwaukee Public Schools. Comparing the outcomes of these two groups reveals how exposure to the MPCP at that time affected students' later outcomes, regardless of how long students initially using vouchers remained enrolled in private schools.
Our results confirm that the program reduces the incidence of both criminal activity and paternity disputes for young adults. Specifically, we find that participating in the program in 8th or 9th grade is associated with a statistically significant reduction of around 53 percent in drug-related convictions, 86 percent in property damage convictions, and 38 percent in paternity disputes.
Not surprisingly, the program's effects on some outcomes differ for males and females. Males participating in the MPCP experienced a 53 percent reduction in drug-related offenses and an 87 percent reduction in property damage offenses when compared to their public-school peers. Females exposed to the program, on the other hand, experienced little or no reduction in convictions for those crimes. Because young males are at greater risk than young females to commit such crimes, the MPCP had greater scope to affect their behavior. The program's effects on paternity suits were more similar by gender, with a 42 percent reduction for males and a 34 percent reduction for females.
The character-building effects of the school voucher program also varied with students' initial math ability. For two outcomes—thefts and traffic offenses—exposure to the MPCP did more to reduce negative outcomes for students with high initial math scores than for students with lower scores. In contrast, students with lower initial math ability experienced a larger reduction in paternity suits due to their participation in the MPCP than students with higher initial math ability.
Of course, much more research on this topic is needed. While two random assignment studies find that winning a charter school lottery reduces crime for male students, no random assignment studies link private school choice to adult crime or paternity disputes. Even so, our new results suggest that the Milwaukee voucher program improves low-income urban students