As managerial problems become increasingly complex and decisions are made more frequently by groups than by individuals, the study of group ethical decision making becomes more important both to academic researchers and practicing managers. However, empirical studies investigating aspects of group effects on ethical behavior are almost nonexistent (Ford and Richardson, 1994; Jones, 1991; Randall and Gibson, 1990). Furthermore, while multilevel theorizing (Dansereau, 1989; Rousseau, 1985; Yammarino and Naughton, 1992) is a potential significant development in organizational research, most empirical studies investigating ethical behavior have focused only on decisions made at an individual level of analysis. Considering the attention of both the scholarly and business worlds toward the importance of group activities within business organizations, an important question is how is ethical behavior affected within a group environment? Is there a difference between ethical decisions made at an individual-level of analysis and those made at a group level of analysis?
The purpose of this study is to investigate the effects of certain variables that have been most used in studies of ethical decision making at an individual-level of analysis when decisions are made in a group setting. Reviews of the ethical decision-making literature indicate that a large number of studies have investigated the effects of performance, environmental change, age, and the differences between group type (college students and professional managers) on the ethicalness of decision making using an individual-level of analysis (Ford and Richardson, 1994; Jones, 1991; Randall and Gibson, 1990). The present study examines the effects of these four variables on ethical decision making at the group level of analysis. The first section of this article describes the type of groups that were studied, the second section presents the research hypotheses, and the third section details the research methodology and data analysis. The fourth section describes the results, and the paper concludes with a discussion of implications for both managers and researchers.
DEFINITION OF A WORK GROUP
A work group has been defined as two or more persons who interact with one another such that each person influences and is influenced by each other person (Shaw, 1981) and that members have differentiated roles (Goodman, 1986). McGrath (1984) stated that when group members interact, they behave in some recognized relationship to one another and this phenomenon can be labeled group process. In this study a number of work groups composed of professional managers and college students participating in a management simulation exercise were analyzed. In performing the requirements of the simulation exercise, participants assumed certain roles such as finance manager, marketing manager, manufacturing manager, and research and development manager and interacted with one another in a recognized relationship.
Hypotheses were developed from studies using an individual level of analysis. Our intent is to extend these findings to a group level of analysis. What follows is the development of specific hypotheses for this study.
Hosmer (1987) argued that current strategic planning systems used by many firms emphasize improvement in competitive positions. According to Hosmer, these systems define poorer performing divisions as having a potential for liquidation or sale. When a division finds itself in this position, there is a great incentive for its management to take short-term actions to save itself - actions which might be unethical or are directly contrary to the moral standards (either explicit or implied) of their organizations. Derry (1991) stated that individual managers in poorly performing firms will have a tendency to make unethical decisions if top management has communicated that predetermined performance levels must be achieved at "any cost. …