Group Dynamics: Theory, Research, and Practice 1998, Vol. 2, No. 3,168-184 Copyright 1998 by the Educational Publishing Foundation 1089-2699/98/S3.0O Inspecting the Emperor's Clothes: Evidence That Random Selection of Leaders Can Enhance Group Performance S. Alexander Haslam, Craig McGarty, Patricia M. Brown, Rachael A, Eggins, Brenda E. Morrison, and Katherine J. Reynolds Australian National University This article tests the hypothesis that group performance might be superior when leaders are randomly rather than systematically selected. In Experiment 1 groups with ran- domly selected leaders performed a survival task better than groups whose leaders were systematically selected. This effect was replicated in Experiment 2: Groups with a random leader also performed better than groups with no appointed leader and followers adhered more strongly to the group decision. In Experiment 3, naive participants' experimental expectations confirmed the counterintuitive nature of these findings. Results suggest that systematically selected leaders can undermine group goals and group maintenance. The possibility that this occurs because leaders assert their personal superiority at the expense of shared social identity is discussed. A large part of the research conducted by organizational psychologists over the last 40 or so years has been oriented toward the practical goal of seeking to improve individuals' perfor- mance in the workplace. At an initial stage in this research process, considerable effort went into trying to improve the selection procedures used to recruit employees, with a view to identifying personnel best suited to particular positions. This is particularly true with positions of authority and leadership. The underlying logic here, which can be traced back to some of the very first writings on management (e.g., Taylor's, 1911, second principle of scientific management; see also Locke, 1982), is that the process of scientifically endeavoring to identify the best leaders should produce dividends for the organization as a whole (e.g., in terms of group harmony and productivity). That this logic is implicit in management S. Alexander Haslam, Craig McGarty, Patricia M. Brown, Rachael A. Eggins, Brenda E. Morrison, and Katherine J. Reynolds, Department of Psychology, Australian National University, Canberra, Australian Capital Territory, Australia. We are very grateful to Russell Spears for detailed comments on an earlier version of this article. We also thank Kip Williams for his exceptionally thorough and insightful editorial work, which included suggesting the design of the third experiment. Correspondence concerning this article should be ad- dressed to S. Alexander Haslam, Department of Psychology, Australian National University, Canberra, Australian Capital Territory 0200, Australia. Electronic mail may be sent to Alex.Haslam @ anu.edu.au. theory and practice is suggested by the fact that although a number of studies have sought to compare the efficacy of different selection procedures (e.g., Levinson, 1994; Walters, Miller, & Ree, 1993), to the best of our knowledge, none has sought to examine whether the process of leadership selection is always beneficial in itself. Although they make a slightly different point, Emler and Hogan (1991) ap- pealed to common sense rather than empirical evidence when they stated, "you cannot ran- domly allocate leadership responsibility and expect the interests of justice or society to be well-served" (p. 86). In fact though, there are a number of good reasons for believing that under certain condi- tions the process of systematic leadership selection might be deleterious to group perfor- mance. In this vein, pioneering empirical studies in the 1950s and 1960s suggested that group productivity was improved to the extent that leaders showed consideration for their group (Cartwright & Zander, 1960; Shartle, 1956) and might diminish to the extent that leaders sought only to satisfy their own objectives (e.g., self-promotion; Fouriezos, Hutt, & Guetzkow, 1950). Yet clearly in the process of striving to become a leader individuals may display considerable concern for self-advancement and a commensurate lack of consideration for their fellow group members (e.g., Kipnis, 1972; Mulder, 1977). Work conducted around this time also suggested that groups (in particular 168 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.