Interest has rapidly increased in studies that randomly assign classrooms or schools to interventions. When well implemented, such studies eliminate selection bias, providing strong evidence about the impact of the interventions. However, unless expected impacts are large, the number of units to be randomized needs to be quite large to achieve…

Describes an in-class exercise using students' discussion of their answers on a multiple choice test to demonstrate the powerful effects and pervasiveness of social influence. After a brief multiple choice test, students discuss their answers with their neighbors and then answer again. Similar response patterns emerge among the participants. (MJP)