Despite promotion by well-known economists and supporting economic theory, econometric analyses of voucher systems often find that they have been unsuccessful in improving traditional measures of educational success. This paper examines a possible explanation of this phenomenon by comparing the drivers of educational achievement and of school popularity by examining private school choice. The findings of this paper indicate that there is a disconnect between school success and school popularity, which adversely effects both the demand and supply-side benefits of voucher systems. Additionally, this paper reviews matching mechanisms that seek to efficiently match students with schools based on both student and school preferences.

As charter schools continue to proliferate across United States, their impact on the public education system is becoming an increasingly important public policy question. Charter school proponents argue that combined pressures of consumer choice and market competition will induce traditional public schools to respond by providing higher quality education and promoting innovation and equity. Skeptics worry that charter schools pose risks of segregating students by race and economic level, and reducing per-pupil resources available to traditional public schools. This dissertation provides a comprehensive evaluation of the effects of charter schools on regular public schools by addressing the following questions: 1) How do the charter schools affect the racial, ethnic and cosio-economic distribution, student-teacher ratios and achievement of traditional public schools? 2) How do the size and scope of competitive effects vary according to different measures of competition? Using two-period panel data from the National Center of Education Statistics (NCES) Common Core Data (CCD) for traditional public schools in Florida, New Jersey, Texas and Ohio, I compare changes in racial and ethnic distribution, student-teacher ratios and achievement in public schools that do and do not face competition. I use a variation of the difference-in-differences (DD) estimation strategy to study the effect of charter schools on the outcome measures. The findings from the study suggest that introduction of charter schools in the educational landscape has affected student distributions, and at least in some cases, student-teacher ratios and the performance of traditional public schools. Charter schools seem to contribute to declines in the share of non-Hispanic white students in traditional public schools in all four states. The results show variation in other outcome areas across states and competition measures. The findings highlight the importance of monitoring what will happen to non-choosers in traditional schools as well as the role of considering state context and empirical measures while generalizing from charter school studies.

This dissertation examines the effects of charter schools in Texas, using data from the Texas Education Agency for 190 charter schools and over 60,000 charter students. In Chapter II we examine charter effect test score gains for charter students. After controlling for individual student characteristics, we find that students in their first year in a charter school have large negative test score gains compared to when they were in traditional public school, and that charter schools that have been in operation for more than one year have higher average test score gains than new charter schools. Charter schools appear to have the most positive effects on African-American students. We find that the overall effect of being in a charter school for multiple years is that students have slightly lower average test score growth than when they were in a traditional public school. In Chapter III we examine the effect of charters on test score gains for students attending nearby traditional public schools. After controlling for campus and student characteristics, we find traditional public school districts and campuses that face greater competition from charter schools have higher average test score gains than other traditional public schools. This positive effect of charter competition is strongest for African-American and Hispanic students, and is focused entirely on students attending traditional public campuses in the bottom 50% of the initial campus average achievement distribution. In Chapter IV we examine the charter effect on the distribution of students by ability and race/ethnicity, as well as examining what factors are associated with a student choosing to move to a charter school. We find that students who move to charter schools tend to move to schools with a higher percentage of students of their same race/ethnicity, and that this gap is largest for African-American students. We also find that average math and reading test scores are lower than the statewide average at the traditional public schools that charter students leave, and that charter schools are attracting, on average, the lower-performing students from these lowperforming schools.

The Education Reform Act of 1988 introduced a policy of open enrolment into English secondary education that was designed to enhance the scope for parental choice of schools. In the resulting 'quasi market' for education, state school admissions authorities can no longer deny most expressed parental preferences, and the majority of state educational funding follows pupils to the secondary schools that they attend. Accompanying these policy reforms has been a longstanding concern that the new school attendance patterns resulting from the enhanced choice present within an open enrolment system would further polarize the social compositions of secondary schools in England. This thesis employs recently developed individual-level databases such as the Pupil Level Annual Schools' Census (PLASC), along with GIS mapping software, to investigate the role that choice of non-local schools played in the degree of social segregation in English secondary schools in 2002. A detailed analysis of the data reveals high rates of non-local school attendance across many areas of England in 2002 as large numbers of pupils from all backgrounds bypassed their local schools in favour of non-local alternatives. Although non-local school choice was exercised by both disadvantaged and more advantaged segments of the schooling population, pupils eligible for free school meals were less likely to attend higher performing non-local schools than their more advantaged counterparts. The disproportionate gains made from non-local school attendance by more advantaged secondary pupils within the marketplace helped to reinforce local school hierarchies already strongly associated with performance and social composition. As the individual level pupil data in PLASC illustrates, the exercise of non-local school choice in 2002 produced school compositions that were more segregated by socio-economic status than they otherwise would be under a system of local school catchments. Thus, rather than helping to diminish the social segregation of secondary pupils resulting from pronounced residential segregation levels, the availability of parental choice instead further stratified most English secondary schools by socio-economic status in 2002.

Thesis (Ed. D.)--University of Toronto, 2004. / Adviser: Joel Weiss.

Thesis (M.A.)--University of Toronto, 2006. / Source: Masters Abstracts International, Volume: 45-03, page: 1153.

Thesis (Ed. D.)--Illinois State University, 1979. / Title from title page screen, viewed Feb. 2, 2005. Dissertation Committee: Ronald Laymon (chair), Mary Ann Lynn, Dale Jackson, John McCarthy, Pat White. Includes bibliographical references (leaves 133-141) and abstract. Also available in print.

Thesis (Ph. D.)--University of Texas at Austin, 2002. / Vita. Includes bibliographical references. Available also from UMI Company.

Thesis (Ph. D.)--University of Missouri-Columbia, 2002. / Typescript. Vita. Includes bibliographical references (leaves 140-149). Also available on the Internet.

Thesis (M. Ed.)--University of Hong Kong, 1998. / Letter to the parents also in Chinese. Includes bibliographical references (leaves 125-139).