secondary students were in racially homogeneous schools. Students attending religious schools
in the voucher program were less likely than public school students to be in racially
homogeneous schools (41 percent and 30 percent). In non-religious schools, elementary students
were more likely to be in racially homogeneous schools (73 percent). There was only one non-
religious private secondary school in the data set, and it was not racially homogeneous. Voucher
students as a whole were less likely than public school students to be in racially homogeneous
schools; the data tables indicate that 50 percent of elementary students and 29 percent of
secondary students were in racially homogeneous schools.
The only previous study of the Cleveland voucher program was conducted by Greene.
Examining elementary and middle schools, he found that 19 percent of voucher recipients
attended private schools that fell within 10 percentage points of the racial composition of the
metropolitan area, compared to 5 percent of Cleveland public school students. He also found that
61 percent of public school students attended racially homogeneous schools (more than 90
percent white or 90 percent minority), compared to half of voucher recipients.
Greene and Marcus Winters of the University of Arkansas examined the new voucher
program in Washington D.C. They find that in public schools the percentage of students who are
white differs from the percent white of the metro area by an average of 40 points, compared to
34 points for private schools participating in the voucher program. They also find that 85 percent
of public school students attend racially homogeneous schools (more than 90 percent white or 90
percent minority), compared to 47 percent of students in participating private schools. When the
definition of “racially homogeneous” is made stricter, such that schools need to be 95 percent
white or 95 percent minority to qualify, the gap widens. While 84 percent of public school
students attend racially homogeneous schools by this definition, 43 percent of students in
participating private schools do so.
9
Convention