Does School District Leadership Matter? The Impact of School District Superintendents on Student Achievement

• In Event: 3429 - Section on Sociology of Education Refereed Roundtables (1 hour)
  In Refereed Roundtable 60min: 21. K-12 Teachers and Administration

Abstract

Despite the highly decentralized governance structure of the American educational system, very little research has examined the impact of school district management on student outcomes. This paper addresses this gap by providing the first estimate of the impact of U.S. school district superintendents (the “CEOs” of school districts) on students’ educational performance. I construct a longitudinal dataset that provides the employment histories for all school district superintendents in 28 U.S. states from 2009-2015, which I match to 3rd - 8th grade district and grade level student achievement and covariate data from the Stanford Educational Data Archive (SEDA). I exploit superintendent switching between districts to analyze differences in student outcomes under the same and different superintendents. I find that “superintendent effects” (defined as a high-quality superintendent’s effect on a new district) range from 0.04-0.07 SD on ELA scores. These results challenge existing theories of educational organizations as “loosely coupled” systems, where centralized administrative decisions have limited influence over classroom practices. In contrast, I find that school district leadership has a sizeable effect on student outcomes, roughly half the effect size of a high-quality teacher on achievement.

Author

• Greer Mellon, Columbia University