Staff turnover is an issue faced by public school districts throughout the nation, especially those districts and schools serving the poorest and the lowest performing students. Staff turnover at any level of the education hierarchy can threaten continuity for school and district staff. In the United States 15 percent of superintendents, 23 percent of principals, and 16 percent of teachers do not return to their placements from one school year to the next. This dissertation is comprised of three essays each of which explores a different perspective on public school district staff turnover. The first essay uses quantitative models, which include both principal and school fixed effects to show that teacher ratings of the school environment, known to be associated with teacher mobility decisions, depend on the principal, independent of other school and district contextual factors. The second essay, using administrative records data creates a panel data set of the employment history of high school principals in North Carolina and Ohio to show that their past job experience predicts their subsequent turnover risk. The third essay combines data from 26 interviews of district superintendents, central office staff, and principals with 18 years of public school employment history data from Ohio. Using mixed methods, it provides insights on how superintendent turnover impacts district principals and staff at the district's central office.