Moderators

eric.ed.gov har udgivet: In randomized control trials, such as the one the authors conducted in Alabama, moderators are potentially an important source of variation in treatment effects. Whether moderators used in the RCT’s analysis are pre-existing characteristics of students, their teachers, or the school or other organizational context, they can point toward differences in the way the program (treatment) is implemented, the way teachers interact with different kinds of students when implementing the program, or how effects are operating at levels in the organization above the level where the differences were measured. The authors appreciate Cronbach’s (1975) metaphor of the “hall of mirrors” to point to the endless possibilities for higher-order interactions complicating the explanation of what was found. But the goal of this paper is to illustrate a disciplined… Continue Reading →