The effect of bias instructions on the judgments and interobserver agreements of naive behavior observers

Behavioral scientists have been interested in increasing their understanding of the variables affecting observational data. Concern with improving the reliability of these data has led to numerous investigations of systematic observation methods. One aspect of observational procedure which has begun to receive research and speculative attention has been the area of observer training. Since 1970, investigators have become increasingly interested in identifying variables that might influence the effectiveness of observer training procedures for improving the extent to which observers agree on their rating of behavior. Although empirical findings concerning variables relevant to observer training procedures have only begun to appear in the literature, investigators using systematic observation for data collection have become increasingly aware of the need for organized and specific training experiences for their observers. Suggestions for effective training procedures, based on empirical findings and intuition, have recently been made available (DeMaster, Reid, & Twentyman, 1977; Johnson & Bolstad, 1973; O'Leary & Kent, 1973; Romanczyk, Kent, Diament, & O'Leary, 1973; Wildman & Erickson, 1977; Wildman, Erickson, & Kent, 1975)