Achievement Place: the reliability of self-reporting and peer-reporting and their effects on behavior

Abstract

The reliability of the boys reporting their own behavior of their peers was measured in two experiments at Achievement Place, a community based, family style, behavior modification program for delinquents based on a token (point) economy. The results of these experiments indicated that; (a) the boys were not "naturally" reliable observers, (b) the reliability of peer-reporting could be improved by providing training on the behavioral definitions and by making points contingent on agreement between each boy's peer-report and an independent adult observers' report, (c) the reliability of self-reporting could be improved by making points contingent on agreement between the self-report and the trained peer's report, and (d) giving self-reports and peer-reports did not produce a systematic effect on the boys' room-cleaning behavior as measured by an independent observer.