Data and Code for "School Boards and Education Production: Evidence from Randomized Ballot Order"

We examine the causal influence of educators elected to the school board on local education production. The key empirical challenge is that school board composition is endogenously determined through the electoral process. To overcome this, we develop a novel research design that leverages California's randomized assignment of the order that candidate names appear on election ballots. We find that an additional educator elected to the school board reduces charter schooling and increases teacher salaries in the school district relative to other board members. We interpret these findings as consistent with educator board members shifting bargaining in favor of teachers' unions.

Time Period(s):  1996 – 2018

Universe:   Local school boards

Data Type(s):  administrative records data; observational data

Sampling:  Schools boards in California with contested elections for seats over time.

Data Source:  Numerous publicly-available education records from California as well as election and ballot order information. Please see paper for full details and citations.

Unit(s) of Observation:  School boards (distinct school district-election year combinations)

Geographic Unit:   School-grade-years

The following publications cite the data in this project.

• Shi, Ying, and John Singleton. “School Boards and Education Production: Evidence from Randomized Ballot Order.” American Economic Journal: Economic Policy, n.d.