GROUP PERFORMANCE UNDER EXPbRIENCED AND INEXPERIENCED LEADERS: A VALIDATION EXPERIMENT

This study investigated the effect of experience and training on the performance of military officers in experimental leadership situations. As in a previously reported study conducted with Belgian naval personnel, group performance under trained «-»d experienced officers was not significantly better than performance under untrained recruits. Moreover, years of leadership experience as an officer was uncorrelated either with performance on any of the five different tasks or with group member satisfaction. Group Performance under Experienced and Inexperienced Leaders: A Validation Experiment 2 Fred E. Fiedler and Martin M. Chemers University of Illinois The military services, business and industry, and governmental agencies consider it taken for granted that leadership experience is essential for the development of leadership skills and effective leadership performance. Yet, an extensive search of the literature did not reveal one single study which addressed itself directly to this particular question. An experiment is here reported which attempted to determine the degree to which leadership performance on various laboratory tasks is related to previous leadership experience. The experiment represents a replication and attempted cross-validation of a study conducted by the senior author in cooperation with the Belgian naval forces in 1964 (Fiedler, 1966). While testing hypotheses relating leadership style to performance, this study also compared the performance of groups led by 48 petty officers and 48 groups led by 48 recruit sailors. The petty officers were career men with a minimum of two years of petty officer school, an average age of 29.5 years, and an average of ten years of leadership experience. Tlie recruits had had neither previous navy leadership experience nor training. Their average age was 20.2 years. The leaders as The research reported In this paper was supported by the Advanced Research Projects Agency, ARPA Order No. 454, under Office of Naval Research Contract Nr 177-472, Nonr 1834(36), (Fred E. Fiedler and Harry C. Triandis, Principal Investigators). 2 Now at the University of Delaware, Newark. 2 well as the groups were matched for intelligence, language ability in French and Dutch, and a leadership style score (Esteem for the Least Preferred Coworker, or LPC) which has been used extensively in our research program (See Fiedler, 1967). The groups were given four tasks, viz., (a) to write a recruiting letter urging young men to join the Belgian naval forces as a career, (b) to find the shortest route for a ship which had to touch at ten and (c) at twelve different ports, and (d) to teach men how to disassemble and reassen.ble a .45 cal. automatic pistol without use of language. These tasks were designed in cooperation with Belgian naval officers; they were considered fair and reasonable tasks by the participants. Performance on the four tasks was uncorrelated and we were, therefore, dealing with presumably independent measures of leadership performance. Despite the fact that the navy petty officers had gone through two years of petty officer candidate school, and had had an average of ten years of leadership experience in the Belgian navy, their performance as leaders as measured by their group's effectiveness was essentially identical to that of the new recruits. A number of possible explanations could be advanced for this lack of significant findings. First, of course, was the possibility that we were dealing with chance findings; second, that the petty officers' training might not have been sufficient; third, that the tasks were too short to be meaningful or not relevant. The negative findings of this study were sufficiently startling so that we decided to conduct a second study as soon as possible to check on the generality of the Belgian navy findings. The present paper reports the results of this study.