News from
PRINCETON UNIVERSITY
Communications and Publications
Stanhope Hall
Princeton, New Jersey 08544-5264
TEL 609/258-3600 FAX 609/258-1301

Release: June 19, 1995
Contact: Justin Harmon (609/258-5732)


New Research Finds Average Class
Size Contributes to Black-White
Test Scores Disparity

PRINCETON, N.J. -- Differences in class size are contributing to
the disparity between blacks' and whites' scores on standardized
achievements tests, concludes new research co-authored by
Princeton University economist Cecilia Rouse.

Most recent research has discounted class size as a possible
factor in African-Americans' lower test scores, finding little
difference in teacher-pupil ratios at predominantly white and
predominantly black schools. But judging teacher-pupil ratio an
inadequate proxy of school quality, Rouse and co-author Michael
Boozer, assistant professor of economics at Yale, looked instead
at data on average class size, obtained in part from their own
survey.

"The typical measure of class size -- the pupil-teacher ratio
-- is a misleading gauge of school quality when you're trying to
study black-white differences," says Rouse, an assistant professor
of economics who is also affiliated with Princeton's Woodrow
Wilson School of Public and International Affairs and with the
National Bureau of Economic Research. "While we see no difference
in teacher-pupil ratio, we do see a black-white difference in
actual class size."

The problem with teacher-pupil ratios, the authors explain, is
that they often mask enormous variations in class size that can
exist within schools. Special and remedial classes tend to have
smaller numbers of students; thus, at schools with a relatively
large number of special-education classes, regular students will
wind up in relatively large classes -- even if the overall
teacher-pupil ratio is about average. Predominantly black schools,
Rouse notes, tend to have more special-education classes, as well
as relatively large average class size.

This difference in class size, she and Boozer calculate,
accounts for about 15 percent of the black-white difference in
test scores.

The data for the study come from the National Longitudinal
Survey of 1988 and the researchers' own survey of a random sample
of 500 New Jersey teachers in June 1994.

"We started out puzzled by the incongruity between statistics
on teacher-pupil ratios and the anecdotal information we were
getting about inner-city schools," Rouse says. "The data show that
ratios are about the same in predominantly black schools as they
are in white schools. And yet, one constantly reads about the
large class sizes in inner-city schools. We were asking how we
could reconcile these different pictures.

"We think we know what's going on: In predominantly black
schools, there are more special-education classes, and they're
being compensated for with slightly larger regular classes," Rouse
adds. "Knowing what we know about student achievement and class
size, we think we've found a likely contributor to the difference
between black and white test scores and, perhaps more generally,
to the widening gap in this country between the economic status of
blacks and whites."

Now a working paper in the Industrial Relations Section of
Princeton's Economics Department, the research is awaiting
publication in an academic journal.

---