This paper introduces an empirical strategy to estimate dynamic treatment effects in randomized trials that provide treatment in multiple stages and in which various noncompliance problems arise such as attrition and selective transitions between treatment and control groups. Our approach is applied to the highly influential four year randomized…

Reviews Tennessee's Project Student/Teacher Achievement Ratio (STAR) and puts the results in the context of other nonexperimental evidence about class size. Nonexperimental evidence does not support any achievement gains attributable to class-size reduction, and empirical evidence from STAR has design and implementation flaws that cast doubt on…