Hawthorne Control Procedures in Educational Experiments: A Reconsideration of Their Use and Effectiveness

This article reports on a descriptive analysis of research practices and a meta-analysis of effect sizes associated with control groups employed to address Haw-thorne effects in educational experiments. The descriptive analysis of 86 studies revealed a diversity of practices designed to control one of three artifact variables: special attention, activity related to the experimental task, or awareness of partici-pation in an experiment. The meta-analysis provided no evidence for a Hawthorne versus no-treatment control difference. Moreover, a detailed analysis of these studies by their control procedure, and subsequently by other moderator variables, also revealed no systematic trends to suggest a specific artifact source. A within-study analysis of the pattern of treatment/Hawthorne/control group effect sizes suggested that the artifact controls were of limited utility. Increased research is urged into other artifacts and alternative control procedures that the educational researcher should consider. Years ago Desmond Cook (1962) raised the specter of the Hawthorne effect, the methodological contaminant first observed in the industrial experiments known a