Using the three paper format, this dissertation contributes to the literature evaluating school choice and school competition. This study highlights important gaps in our collective understanding of the impact of school choice policy. This dissertation contributes in multiple ways to the closing of important gaps related to the effect of school choice induced competition on student outcomes: 1) developing a consistent measure of competition grounded in theory and empirical evidence and 2) evaluating the systemic effects of competition, the effects of school competition on all students within an educational market regardless of the school attended. I address the first gap in three ways. I highlight the existing variation in competition measures and demonstrate that the presence of multiple competition measures in the extant literature is cause for concern. I do this by showing that you can infer positive, negative, and null impacts of competition on student outcomes simply by substituting the various measures of competition, operationalized for the Michigan context, into the same Fixed Effects regression model. Second, I lay out a process for selecting between measures of competition, with the goal of a continued conversation around improving our measures of competition. Third, I suggest a theoretically grounded, empirically refined measure of competition. These efforts contribute to the current school choice policy conversation by focusing attention and thought on the definition and measurement of competition, a key avenue through which choice is to improve the educational system. Improving on the measurement of competition should appeal to all interested in school choice as it is essential to any evaluation of the policy. Given the current debates and mixed results, the common ground of developing an accepted measure of competition would help develop a consistent literature base. I address the second gap in two ways. I first highlight the importance of bringing all students in an educational market under one framework – that of systemic effects – in addition to comparing the performance in a student’s residentially assigned district to charter schools or other public schools through inter-district choice. While evaluating the systemic effects of school choice induced competition allows us to address questions of policy relevance, such as what are the average impacts of competition on educational outcomes for all students residing within an educational market, there are no domestic studies of the systemic effects of competition. I then produce the first such domestic study for the state of Michigan. I create a unique five year panel dataset covering the school years 2008-09 through 2012-13 – drawing on the Michigan’s Center for Educational Performance and Information and the Common Core of Data – to evaluate the systemic effects of competition on the average and variation of student test score outcomes. The evidence suggests that there is not a single systemic effect of competition for all districts or contexts. The average impact and the impact on the variation of test scores differs across district context. Competition is associated with negative impacts in some cases and positive impacts in others. In other words, competition may not be a tide that lifts all boats nor does it lead to a narrowing of the gaps between the boats.In sum, this paper makes multiple contributions to the school competition literature. I underscore the need for more careful attention to be paid to the measurement of competition. I also demonstrate the value-added by evaluating the impact of competition on the entire system rather than just on its component parts. In this work I demonstrate that competition does not operate in a simple manner – introducing options yields competitive pressure yields positive improvements. There is evidence that competitive pressure has mixed impacts dependent on context. This has implications for both the design of policy as well as whether or not school choice will yield improvements for all students.
Read
