PRINCETON, New Jersey — A new two-year study of low-income New York City students in grades 3 to 6 who received vouchers to attend
private schools shows no significant difference in test scores between the scholarship and the control group. The report just released from Mathematica Policy Research, Inc., finds that students who were offered
scholarships to attend private schools as part of one of the nation’s largest private voucher programs performed about the same on standardized reading and mathematics tests as students who were not offered
scholarships.
A detailed examination of the results by racial/ethnic group and grade level reveals no gains for Latino students in any grade.
According to David Myers, a principal investigator of the study and a
senior fellow at Mathematica, the results provide new information that can be used in the debate about the value of education vouchers on students and parents.
“In terms of enrollment, this is the largest ongoing
evaluation of the current voucher experiments in the nation and shows results for the most diverse population of low-income students studied,” said Myers.
African Americans showed a significant increase at the 6th
grade, however, the anomoly could not be explained.
“One group stands out with consistent positive impacts, but this is inconsistent with what we find for the other groups offered vouchers. Because gains are so
concentrated in this single group, one needs to be very cautious in setting policy based on the overall modest impacts on test scores,” Meyers added.
Impacts on Test Scores
— Overall, students who were
offered a scholarship performed at about the same level as students in the control group (24 percent never used a scholarship when offered). When results for Latinos (46 percent of the
sample) are examined
separately, there is no impact for Latinos than the control group.
— When results for Latinos are examined separately, there is no impact at any grade level.
School Facilities and Climate; Parent Satisfaction
and Involvement
— Parents using vouchers express more satisfaction with the private schools.
— As reported by the parents, the schools attended by the scholarship students were smaller, as was class size.
—
The level of parent involvement was about the same for those with children in private schools and those with children in public schools.
— Parents of children attending private schools were less likely to report
property destruction, tardiness, missing classes, fighting, cheating, and racial conflict at their schools than parents whose children were in public schools.
— Students in private schools had more homework than
their public school counterparts.
— Parents using vouchers reported more communication from their schools than did parents in public schools.
— Private school children were more likely to receive religious
instruction outside of school.
Participation in the Scholarship Program
About 64 percent of the students who were offered scholarships in 1997 were using them two years later.
Students who used the
scholarships were generally similar to nonusers, but in some instances they were not. For example, baseline test scores were similar, households were equally likely to speak English as their main language, and mothers
were equally likely to have been born in the U.S.
However, nonusers were somewhat less likely to have received special education before the baseline test.
Scholarship users had mothers with somewhat higher
educational attainment and lived in families with higher incomes (about $2,700 higher, with an annual average income of about $10,400).
The most frequently cited obstacles preventing parents from sending their
children to the preferred school included cost, transportation problems, and lack of space at the school.
In February 1997, the School Choice Scholarships Foundation (SCSF) offered 1,300 scholarships to children from
low-income families attending New York City public schools. More than 20,000 students applied for the scholarships, worth up to $1,400 a year for three
years.
Recipients were selected by a lottery in May 1997 and
began attending private and parochial schools the next fall.
Mathematica's multiyear evaluation is based on a scientifically rigorous research design that takes advantage of the fact that the SCSF held a lottery.
Doing so allowed researchers to conduct a randomized experiment, in which students were randomly selected for a scholarship (treatment) group and a control group.
Mathematica has just completed collecting a fourth
round of data and will report on the results early in 2001.
The report, "School Choice in New York City After Two Years: An Evaluation of the School Choice Scholarships Program," by David Myers, Paul
Peterson, Daniel Mayer, Julia Chou, and William G. Howell, is available on
