The purpose of this study was to examine the effects of criteria-referenced formative assessment (CRFA) on students' achievement in theater arts. The role of type of task in differentiating the treatment-achievement relationship was explored. The analytical sample included 520 fifth-grade students from 13 schools in New York City. Schools were randomly assigned to the treatment or control group. Selection bias in treatment assignment at student level was adjusted with propensity score analysis. CRFA had a positive effect on students' achievement on performance tasks (d = 0.25), but no significant effect on students' performance on the analytical constructed response tasks or the theater vocabulary multiple-choice items.