Jump to:Page Content
CAMBRIDGE, MA -- Low-income African American students participating in a voucher program who attended private schools generally scored higher in math and reading after six to seven months than similar youngsters who remained in public schools in Dayton, Ohio and Washington, D. C. These findings come from two randomized tests of the effects of school vouchers.
"Many people think that equal opportunity will not be achieved until we eliminate differences in educational performance between blacks and whites. If the initial findings from Dayton and D. C. hold up over time, vouchers for students beginning in elementary school may help eliminate the black-white test-score gap," said Paul E. Peterson, director of the Program on Education Policy and Governance at Harvard University and a co-author of the evaluations. "These early results show the effects of only about six to seven months in private school," he added, "but they indicate that private schools serve African American students well--especially if the student enters private school during the early elementary years. For older students, the challenge is greater"
Dayton's African American students entering private school in grades 2-8 scored, on average, 7 national percentile points higher in math, and 5 points higher in reading, on the Iowa Test of Basic Skills than did similar black students in public schools. Test scores for public-school and private-school students who were not black did not differ significantly. "Today the gap between average black and white test scores is approximately 25 points," Peterson explained. "If the gap can be reduced by a fifth or more in one year, it gives hope that the difference can be eliminated altogether over the course of the other eleven years of schooling."
In Washington, D. C., black students entering private schools in grades 2 to 5 outperformed, on average, black public-school students by 6 points in math and 2 points in reading. Black voucher students who switched from public to private school in grades 6 to 8 scored 2 national percentile points higher in math but scored 8 points lower than their public-school peers in reading. As in Dayton, test scores for public-school and private-school students who were not black did not differ significantly.
Of those participating in the Dayton test-score evaluation, 75% of the students are African American, 23% are white, and 2% are of another ethnic background. In Washington, D. C., 95% of those participating are African American.
A lottery determined whether individual students received a voucher or became a member of the comparison group. "This provided a rare opportunity to conduct a high-quality randomized experiment of the kind generally required in medical research but seldom used in education," commented Patrick J. Wolf, professor at Georgetown University and co-author of the D. C. research.
The Dayton program is sponsored by Parents Advancing Choice in Education (PACE ), and the District of Columbia program is sponsored by the Washington Scholarship Fund (WSF). Both were locally-initiated scholarship programs that are now partners with the Children's Scholarship Fund, the nation's largest privately-funded scholarship program.
Most other results from the evaluations are quite similar for the two cities. Key additional findings include:
Middle-school students—In Washington, students moving from public to private schools in lower grades had an easier time adjusting to their new schools than did students in grades 6-8. Whereas younger students attending private schools are more likely than their public school peers to agree with the statement, "students are proud to attend my school", the opposite results are obtained for students in grades 6-8. A similar pattern of responses is observed when students are asked what "grade" they would give their school, whether they like their school "a lot", and whether students "get along well with teachers". Suspension rates reported by parents for younger students are similar in private and public schools (5% and 7%, respectively) but considerably higher in private school than public school for students grades 6-8 (20% as compared to 3%). These patterns are not observed in Dayton, however.
Skimming—In general, private schools do not appear to skim the "best and the brightest" of student applicants. Initial math scores of the Dayton students who took the scholarship were in fact lower than those who did not make use of the scholarship; initial reading scores did not differ significantly. In D. C., no differences in initial test scores were observed for students in grades 1-5 in D. C. However, students in grades 6-8 in D. C. who used a voucher to attend private schools had higher initial scores than those who declined the scholarship.
School Grade--Approximately 47% of Dayton private-school parents give their school an "A", as compared to 8% of public-school parents participating in the evaluation. (25% of all Dayton public school parents give their school an "A.") In the District of Columbia, 46% of the private-school parents say their school deserves an "A," as compared to 15% of public-school parents participating in the evaluation.
Parent Satisfaction--51% of Dayton private-school parents are very satisfied with their school's academic program, as compared to 9% of the public-school parents participating in the evaluation. (19% of all Dayton public-school parents say they are very satisfied with their school's academic program.) Washington results resemble Dayton's: 56% of the private-school parents are very satisfied with the academic quality of the school, as compared to 17% of D. C.’s public school parents participating in the evaluation.
Fighting--66% of the parents participating in the evaluation who had children in Dayton-area public schools say fighting at their school is a serious problem, compared with 16% of parents in private school. In D. C., the pattern is similar: 55% of public school parents say fighting is a serious problem, 25% of private-school parents.
Property Destruction--41% of the Dayton public-school parents report property destruction is a serious problem, compared to 8% of private-school parents. In D. C., the percentages are 37% for public-school parents, 17% for private school parents.
Parent-School Communications--Parents of students in Dayton private schools report closer communication with school officials. 92% of private-school parents say they receive notes from their children’s teachers, as compared to 76% of public-school parents. In D. C., the findings are similar: 94% and 77% for the two groups, respectively.
Mobility--students with vouchers are no more likely than public-school students to change schools in the middle of the year. Plans to remain in the same school for the next year were also similar for private and public-school students.
The voucher programs gave low-income families an opportunity to send children to the school of their choice. Scores from 408 Dayton students and 810 Washington students are included in the evaluation. The evaluations reported here are for students who had previously been attending public school. During the first year of the program, PACE offered Dayton lottery winners annual scholarships of up to $1,200 to help pay tuition at a private school for at least four years. WSF offered lottery winners annual scholarships of up to $1,700 for students entering kindergarten through eighth grade for at least three years. The lottery selecting scholarship winners was held in April, 1998. Follow-up testing and surveying were conducted in Spring, 1999.
The evaluations of these voucher programs are activities of the Harvard Program on Education Policy and Governance (PEPG), a program of the Taubman Center on State and Local Government, Kennedy School of Government, Harvard University. The evaluation was funded by private foundations and conducted independently of WSF, PACE, and all funding sources. The evaluation team thanks both organizations for their full co-operation with data collection. Authors of the D. C. report are Patrick J. Wolf, assistant professor, Georgetown Public Policy Institute and research associate, PEPG; William G. Howell, research associate, PEPG; and Paul E. Peterson, director, PEPG and the Shattuck Professor of Government, Harvard University. The authors of the Dayton report are William G. Howell and Paul E. Peterson.
Copies of the reports can be obtained by calling 617-496-5488 or requesting by email at firstname.lastname@example.org. Reports are also available at http://data.fas.harvard.edu/pepg/