Jason Grissom

Demonstrates that principals give higher ratings to teachers on high-stakes personnel evaluations than those they supply confidentially to researchers, though both sets of ratings predict teachers' value added to student achievement. Finds that principals skew high-stakes ratings up more for some teachers (e.g. beginning teachers) than for others.

Shows that Black children are much less likely than White students to be assigned to gifted programs, even after conditioning on math and reading achievement, socioeconomic status variables, characteristics of the schools they attend, and other factors. Proposes that a major contributor to this gap is that high-achieving Black students are substantially less likely to be identified as gifted when taught by a White classroom teacher. 

Takes stock of what we know about multiple measures-based teacher evaluation systems, which have been introduced in the majority of states in recent years. 

Demonstrates that, because of accountability pressures, schools tend to place more effective teachers in tested grades and subjects. Discusses how, in elementary schools, this pattern means that less effective teachers become concentrated in the earliest grades, which harms early grades achievement. These negative effects carry forward into the tested grades that come later.

Finds that superintendent turnover is substantially higher in the largest school districts. Shows that main predictors of a "non-retirement exit" from the superintendent's office are whether the superintendent is "homegrown" (i.e. was promoted from within) and ratings of how well the school board functions; objective district performance measures do not predict turnover.