Notes: Coefficients reported with robust standard errors (clustered at the school level). The unit of analysis is the teacher. All regressions include area fixed effects to account for randomization that occurred within CPS areas. Teacher characteristics include years of experience in CPS (experience squared), race/ethnicity, gender, teacher’s prior evaluation rating, certification status (alternative, National Board), educational attainment (bachelors, masters, or doctorate), and undergraduate major. School characteristics include racial and gender composition of the students, proportion of students with special education status, the proportion of students with free/reduced-price lunch, and prior school achievement that is the average achievement level at the school in the year prior to the pilot (test administered in Spring 2008) and is standardized within the sample of 93 schools. To account for teachers missing prior evaluation ratings, we include a separate indicator variable for teachers missing that rating. Coefficients statistically significant at the *10 percent, **5 percent, and ***1 percent levels.