This dissertation includes three essays on school choice. In Chapter 1 we consider the priority-based affirmative action policy in school choice. We weaken the responsiveness to affirmative action policy by requiring at least one minority student to be weakly better off when their priorities are improved. We find that under both the student-proposing deferred acceptance mechanism (hereafter DA) and the top trading cycles mechanism (hereafter TTC) at least one minority student becomes weakly better off when their priorities are improved. We also strengthen the responsiveness to affirmative action policy by requiring all minority students to be weakly better off when their priorities are improved. We find that there is no non-wasteful, individually rational, mutually best and strategy-proof mechanism that is strictly responsive to affirmative action policy. We then find a sufficiency condition by setting restrictions on the priority improvements for DA to be strictly responsive to affirmative action policy. Our experiments using data from Wake County magnet-school application show evidences that the fraction of minority students that are hurt after the affirmative action policy is very low under both DA and TTC mechanisms. We also find that the affirmative action policy helps a higher fraction of minority students while also hurts a higher fraction of minority students under TTC than DA. But the difference between the fraction of minority made better off and the fraction of minority students made worse off is larger under TTC than DA. In Chapter 2 we study the precedence order of seats in school choice. Many school districts use DA with reserve policies. Because of the reserve policy the precedence order of reserved or targeted seats needs to be defined in order to run the mechanism. Recently, concerns have been raised that the precedence order of seats can affect students' assignments. Earlier studies find that when the precedence order position of a targeted seat of a school is swapped with the position of a subsequent open seat, the number of reserve-eligible students assigned to the school at which the precedence order has been changed is weakly increased. But the aggregate effect of such precedence order change on the students' assignments across all schools is not clear. We find that under an assumption we can have clear predictions on the total number of assignments for reserve-eligible students, reserve-ineligible students and all students across all schools. We find a certain precedence order of seats can maximize the total number of assigned students across all schools. We also find evidences in the Wake County magnet school application that support our assumption and theoretical results. In Chapter 3 we study the use of DA and TTC, in the reallocation of students when a school district breaks up into smaller districts. Under DA with a carefully designed priority structure, we find that letting all students participate in the reallocation can Pareto dominate over only letting misallocated students participate. Under TTC a similar result can be found only under restrictive assumptions. By using administrative data from Wake County Public School System (WCPSS), we conduct a counterfactual break up analysis. Different ways to break up WCPSS are considered in the analysis. Empirically, DA with a three-tier priority structure that lets all students participate is the best modification in all break up cases with regards to factors such as utilitarian welfare and welfare equality. As an alternative, we also study the counterfactual in which WCPSS does not break up but instead fully uses a school choice mechanism. The results suggest that fully switching to a school choice mechanism is more beneficial than breaking up WCPSS into smaller districts. But if WCPSS chooses to break up in the future, our study suggests breaking WCPSS into two districts using a clustering technique. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]