Untitled

In this article, the authors study how teacher performance improves with experience, using classroom observations ratings by school principals.The average teacher's performance improves three quarters of a standard deviation over the first 5 years of her career, with more than one quarter of a standard deviation improvement in just the first year.Understanding how performance improves with experience-the "returns to experience"-could inform teacher policy and management decisions about hiring, turnover, investments in employee training, and more.However, claims of having estimated "returns to experience" are causal claims.The authors describe the assumptions required to interpret changes in observation ratings over time as the causal effect of experience on performance.They compare two difference-in-differences strategies: the two-way fixed effects estimator common in the literature on teacher experience, and an alternative estimation strategy which avoids potential bias arising from effect heterogeneity.Using data from Tennessee and Washington, DC, the authors show empirical tests relevant to assessing the identifying assumptions and substantive threats-e.g., leniency bias, manipulation, changes in incentives or job assignmentsand find their estimates are robust to several threats.