School Choice as Education Reform: What Do We
Know? ERIC Digest.
by Goldhaber, Dan
School choice exists today in a variety of forms, from charter schools,
magnet schools, and district and state open enrollment plans to publicly
and privately financed voucher plans. Despite years of research and debate,
the question of whether school choice improves student outcomes persists.
Choice proponents suggest that injecting greater competition into the education
system can revolutionize education, while opponents argue that choice would
help only a select few students and hurt the many who are left behind.
These starkly different views belie a much murkier research picture that
suggests some forms of school choice may benefit some students under certain
conditions. Much of the rhetoric surrounding the choice issue also ignores
the crucial role that the specifics of school choice policies likely play
in determining their effects. This digest explores the issues surrounding
school choice and highlights some of the major research findings.
HOW CHOICE MIGHT IMPROVE K-12 EDUCATION
There are two arguments about why greater school choice would result
in better educational outcomes: (1) It could allow schools to better tailor
their programs to attract students with particular interests or learning
styles, thus providing a better match for students' unique educational
needs; and (2) it would break the public school educational monopoly and
force schools to compete for students in an educational marketplace in
which "good" schools would prosper and "bad" schools would improve or be
forced to shut down.
If the primary benefit of choice is the match between students and schools,
greater choice would be beneficial regardless of whether a school's resources
are directly connected to its student population. However, if the primary
benefit of choice is the creation of incentives designed to squeeze inefficiencies
out of the system, then the connection between student shifts and educational
resources may be essential so that there are financial consequences associated
with losing students; here competition from the private sector may be beneficial.
Although the theory behind the potential benefits of choice is relatively
straightforward, the educational marketplace is not directly parallel to
the private sector. Students and parents may choose schools for a variety
of reasons, and the vast majority of schools are not for profit. As a result,
the ultimate impact of choice depends on how parents and schools respond
to more schooling options and greater competition.
FINDINGS FROM THE RESEARCH
School choice, in its various forms, has been the focus of numerous
studies over the past two decades. Most studies find greater parental satisfaction
associated with choice. But studies also clearly show that less educated
parents with more modest means are less likely to exercise choice, which
raises concerns that choice systems could lead to less equity and greater
racial, ethnic, and socioeconomic stratification.
No definitive conclusions have emerged about the academic effects of
school choice on students, in part because much of the evidence on choice
is derived from non-experimental research designs where the participants
in the study have self-selected a school, and therefore may differ from
those not in the study. Even students in the study may differ from one
another in ways that are unobservable to the researcher. For example, we
cannot directly observe students' attitudes toward academics, though they
clearly play a role in explaining their achievement.
EFFICIENCY OF PUBLIC AND PRIVATE SCHOOLS
Numerous studies have examined differences in the outcomes of students
who attend public and private schools, enabling speculation about the effects
of a voucher policy on the K-12 education system. Private school students,
in general, outperform their public school counterparts on standardized
tests, and they are more likely to graduate from high school and attend
college. Significantly, positive private school effects have tended to
be found predominantly for minority students in urban settings. There is,
however, mixed evidence about whether this is an effect of the schools
they attend or a result of student factors, such as family background.
Even if private schools do not outperform public schools in terms of
test scores, they might still be considered more efficient if their costs
for educating students are sufficiently lower. It is true that private
school tuition, particularly for Catholic schools, is generally significantly
less than the amount spent on each pupil in the public sector. There are,
however, significant difficulties in accurately determining the cost of
educating private school students. The tuition charged does not reflect
subsidies from religious organizations or the in-kind contributions of
parents who are often expected to contribute to the school's maintenance.
The two sectors also serve very different student populations and provide
different services. Thus, drawing strong conclusions about the value of
choice based on comparisons between public and private schools is problematic.
Educational experiments can mitigate some of the problems in other types
of choice research. For example, voucher experiments have recently been
conducted in several large cities whereby low-income students wishing to
receive vouchers were randomly assigned to a treatment group that received
modest vouchers (around $1,500 per year) or to a control group that did
not. Evaluations showed that attending a private school had a statistically
significant large beneficial impact overall. However, there were no statistically
significant differences between public and private students in the test
score performance of non-African Americans students, and gains were not
found across all grades or subjects.
These evaluations did not include controls for the demographics or achievement
of the other students in the public and private schools, so it is possible
that what is perceived to be a private school effect is actually a student
peer effect. The estimated impact on this group that desires to attend
private schools is not necessarily the effect on the general student population,
and those schools that elect to participate in the experiment may not reflect
private schools in general. Small experiments also do not provide evidence
on the supply-side effects of a larger voucher program, but the quality
and type of participating private schools would, in part, determine the
overall impacts of expanded choice. Finally, many of the students offered
vouchers chose not to use them. This could affect research findings if
the characteristics of these students differ substantially from those who
use their vouchers.
Competition Between Schools
Another method used to assess the impact of choice and competition on
K-12 education is a comparison of student outcomes in localities with differing
amounts of competition, either between public schools and school districts,
or from private schools. Little definitive evidence has emerged about the
impact of charter and magnet schools on other public schools from which
they draw students, but some evidence does show that intra- and inter-district
choice plans affect student achievement. New York City's District 4, for
example, a high poverty district serving predominantly minority students,
has demonstrated dramatic improvements in achievement at least in part
resulting from a choice plan. Also, research on Minnesota and Massachusetts,
states allowing inter-district choice, suggests that districts losing large
numbers of students are likely to implement innovative programs designed
to attract students back.
More generally, a number of studies show that public school districts
that face greater competition from other districts in the same metropolitan
area have better student outcomes and are more efficient. Similarly, research
examining the performance of public schools with greater private school
competition tends to show, to various degrees, greater competition positively
impacts public school students. As with other non-experimental research,
these studies have to take account of a variety of statistical problems.
Failure to control adequately for unobserved differences in community preferences,
such as choice of locality, may lead to biased estimates of the effect
of competition on public schools.
Parental Selection of Schools
Studies generally show a positive relationship between the choice of
school and measures of school quality, implying that parents are making
decisions that are likely to benefit their children academically. However,
many of the cues that parents may use to identify good schools are strongly
correlated with characteristics of the students attending those schools,
such as their socioeconomic status, suggesting that greater choice could
result in greater segregation along racial/ethnic and socioeconomic dimensions.
But it is also true that the public school system is relatively segregated
today and has become more so as middle-class families have fled inner cities.
It is clear that not much is clear when it comes to the issue of choice
in education, but we can draw a few conclusions from examining the broad
array of findings:
* For most students, any benefits of public-private choice tend to be
* There is some evidence that private schools benefit minority students
in urban areas.
* Competition between public schools and districts appears to have small
beneficial effects on the efficiency of the schools competing.
* Parents are sensitive to measured school quality, but they also tend
to base schooling decisions, at least in part, on non-academic attributes
of a school.
Thus, there is an argument for increasing competition through enhanced
choice options, although we do not yet know the consequences of more expansive
choice policies. The existing evidence suggests that choice is unlikely
to be either the panacea that some advocates claim or the disaster that
opponents contend. Rather, its ultimate educational effects are likely
to be influenced by the specific components of choice plans: the requirements
placed on participating private schools, the size of the voucher, the eligible
students, and the financing mechanism.
Goldhaber, D.D. (1999). School choice: An examination of the empirical
evidence on achievement, parental decision making, and equity. Educational
Researcher, 28(9), 16-25. (EJ 600 548)
Goldhaber, D., & Eide, E. (2000). Methodological thoughts on measuring
the impact of competition in the educational work place. Unpublished working
paper, The Urban Institute, Washington, DC.
Greene, J.P. (1998). Civic values in public and private schools. In
P.E. Peterson & B.C. Hassel (Eds.), Learning from school choice. Washington,
DC: Brookings Institution Press. (ED 426 473)
Hoxby, C. M. (1998). Analyzing school choice reforms that use America's
traditional forms of parental choice. In P.E. Peterson & B.C. Hassel
(Eds.), Learning from school choice. Washington, DC: Brookings Institution
Press. (ED 426 473)
Levin, H.M. (1998). Educational vouchers: Effectiveness, choice, and
costs. Journal of Policy Analysis and Management, 17(3), 373-392.
McEwan, P.J. (2000, Summer). The potential impact of large-scale voucher
programs. Review of Educational Research, 70(2), 103-49.
Sander, W. (1999, Fall). Private schools and public school achievement.
Journal of Human Resources, 34(4), 697-709. (EJ 596 621)
Sander, W., & Krautman, A.C. (1995), Catholic schools, dropout rates
and educational attainment. Economic Inquiry, 33(2), 217-233.