Journal of Applied Psychology 1996, Vol. 81, No. 5, 483-497 Copyright 1996 by the American Psychological Association, Inc. 0021-90IO/96/S3.00 Relationship of Goals and Microlevel Work Processes to Performance on a Multipath Manual Task Giuseppe Audia Bocconi University Amy Kristof-Brown University of Maryland at College Park Kenneth G. Brown Michigan State University • Edwin A. Locke University of Maryland at College Park This laboratory experiment used real-time observation to examine the relationships among goals, work processes, and quantity and quality outcomes using a simple multipath assembly task. Trained observers were able to reliably and accurately docu- ment work processes used to perform the task, supporting the use of real-time observa- tion for documenting microlevel task strategies. Results show that (a) work processes were affected by goal content (quantity vs. quality); (b) goal form (gradually difficult vs. fixed and difficult) did not influence outcomes or processes used; (c) work processes and personal goals completely mediated the prediction of quantity outcomes and partially mediated quality outcomes; (d) process-only goals produced a greater number of process changes than outcome goals but led to poorer performance; and (e) outcome goals had a lagged effect on performance. The implications for goal-setting theory are discussed. Goal setting (Locke & Latham, 1990) is one of the best established theories in the field of management. Count- less lab and field studies have shown that conscious goals regulate task performance, principally when goal com- mitment is high and performance feedback is provided. Specific, difficult goals have consistently been shown to lead to higher levels of performance than "do-your-best" or easy goals on both quantity (Locke & Latham, 1990; Wood, Mento, & Locke, 1987) and quality outcomes (Erez, 1990; McCarthy, 1978; Terborg & Miller, 1978; Weingart & Weldon, 1991). Goal-setting theory states that these effects are achieved directly through increased attention, action, effort, and persistence and indirectly through the development of task strategies. Giuseppe Audia, Graduate School of Business, Bocconi Uni- versity, Milan Italy; Amy Kristof-Brown and Edwin A. Locke, College of Business and Management, University of Maryland at College Park; Kenneth G. Brown, Department of Psychology, Michigan State University. Giuseppe Audia is now at the De- partment of Organisational Behaviour, London Business School, London, England. We thank Judy Olian for comments on an earlier draft of this article and Cathy Durham, Suzanne Masterson, and June Poon for assistance in conducting the experiment. Correspondence concerning this article should be addressed to Edwin A. Locke, College of Business and Management, Uni- versity of Maryland, College Park, Maryland 20742. Electronic mail may be sent via Internet to elocke@bmgtmail.umd.edu. However, although the link between goals and direct, automatic mechanisms is well understood, the relation of goals, task strategies, and performance is an aspect of goal theory that merits further attention for two reasons. The first is that unlike direct mechanisms, the develop- ment of task strategies involves a conscious process of search and discovery rather than an automated reaction. This makes it more difficult to decipher the nature of the link among goals, strategies, and performance because many wrong-strategy options may be chosen (e.g., Barley & Shalley, 1991; Wood & Locke, 1990). The second rea- son is that very few studies have objectively measured the task strategies used by individuals or groups working on a task. Several research streams have documented the com- plexity of the relation among goals, task strategies, and performance, and the problems are yet to be solved. One group of studies has examined the effect of assigned or trained strategies on goal-relevant performance (e.g., Barley, Connolly, & Lee, 1989; Barley, Lee, & Lituchy, 1989; Barley & Perry, 1987; Locke, Frederick, Lee, & Bobko, 1984). Characteristically, these studies have shown that such training enhances performance if it is suitable to the task and used in conjunction with difficult, specific goals. In another group of studies, the effects of goal setting in conjunction with chosen (as opposed to assigned or trained) strategies have been investigated. These studies have shown that specific, challenging goals generally en- 483 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.