Researchers investigating expectancy effects in academic as well as motor skills contexts have consistently found differences in instructors' behavior towards high-and low-expectancy children. However, certain methodological problems which have recently been identified may limit the interpretation of such differential behaviors as evidence of instructor bias. The present study was conducted to examine expectancy effects in the athletic setting by directly testing three of these methodological issues. The instructional behaviors of five junior high softball coaches were recorded separately for practice and game situations using the Coaching Behavior Assessment System. Multivariate statistical analyses of coaching behaviors revealed that low-expectancy athletes received more praise for success and more general and corrective instruction in game situations than did high-expectancy athletes. Although these results demonstrate the existence of differential patterns of coaching behavior, the direction of the obtained differences does not support the predictions implicit in the self-fulfilling prophecy model. In addition, results indicated that the methodological issues under consideration do influence the accuracy with which coaching behavior is assessed and interpreted.
This investigation was conducted in partial fulfillment of the requirements for the Ph.D. degree under the direction of Daniel Gould at Michigan State University.
Requests for reprints should be sent to Thelma Sternberg Horn, Department of Human Kinetics, University of Wisconsin, Milwaukee, WI 53201