What is group decision making?
Important societal, business, medical, legal, and personal decisions are often made by more than one person. Psychologists in the field of group decision making have attempted to describe the processes through which such decisions are made. The process by which a set of individual group members’ decisions becomes transformed into a single group decision can be described by a “social decision scheme.” Research by James H. Davis and his colleagues at the University of Illinois has shown the conditions under which groups use various decision-making rules, such as adopting the preference of the majority or that of the member who has the best answer.
The nature of the decision problem facing a group must be considered in an evaluation of the group’s decision process. Psychologist Ivan Steiner pioneered the analysis of the group’s task in his book Group Process and Productivity (1972). Steiner identified three characteristics of group tasks that should be considered in analyzing group decision making. The first is the ability to subdivide the task into subtasks that members can perform individually. For example, a group can plan a meal by making one member responsible for selecting a meat dish, another for choosing a vegetarian entrée, another for choosing desserts, and so forth. Other tasks, in which division of labor is not feasible, are said to be “unitary.” In general, the more important the decision, the more difficult it is to divide it among group members. Yet it is often precisely because a decision task is important that it is given to a group rather than an individual. Therefore, even if some aspect of the task (such as gathering information) can be done individually, final responsibility for the decision rests with a set of people.
Another variable in group tasks is the nature of the goal. In many cases, there is no “best” or “right” decision; the process is simply a matter of determining the group preference. Other decision tasks were called “optimizing” by Steiner because it was assumed that some optimal decision exists. The group’s task is to find it. Most important group-decision tasks are not only unitary but also optimizing. Finally, Steiner noted that the rules governing the group’s decision-making activities were a critical feature of the task. If the group members are not constrained to particular procedures, the task is “discretionary.”
Important insights about decision process and quality in unitary, optimizing, and discretionary tasks have been gained by comparing the decision-making behavior of individuals acting alone to that of persons in groups. It is apparent that the way people behave in groups is different from the way they behave alone. As a consequence, decisions made in groups can differ radically from those made by the same persons acting alone. It is not uncommon for group decisions to be more extreme than an average of members’ individual decisions.
Such a “choice shift” (the difference between the decision of a group and the average decision of group members as individuals) can lead to group decisions that are better than—or not as good as—the average member’s decision. The average group member’s judgment or choice provides one standard against which group decision quality can be compared. Another is the quality of the best decision from an individual member. Steiner called any decrease in quality from the decision of the best member, acting alone, to the group decision “process loss.” Reviews of group-decision research typically conclude that an average group performs above the level of its average member but below that of its best member. It is possible, however, for groups to reach better decisions than can any of their members alone.
Social psychologists recognize two types of influence that can cause group decisions to differ from those of individual group members: normative and informational. Normative influence comes merely from knowledge of the positions of others. One may come to doubt the quality of the alternative that one has selected simply by learning that everyone else believes that another alternative is superior. Confidence in a belief is difficult to maintain in the face of others who are in consensus about a contrary belief. The second type of influence, informational influence, results from logic or argument concerning the relative merits of various choice alternatives. Both types of influence usually operate in group decision making. Through normative or informational influence, individual group members can shift their positions.
Since groups usually make decisions under conditions of uncertainty, it can be difficult to evaluate the actual quality of a group decision. For this reason, the study of “group confidence” has become increasingly important. In general, groups exhibit greater confidence about the quality of their decisions than do individuals. This can be a desirable outcome if commitment to the group decision is necessary. As Irving Janis has shown in his analyses of political and managerial decision making, however, groups can be highly confident even while making disastrous decisions. Proper evaluation of the quality and confidence of the group decision requires an ability to evaluate objectively the decision outcome. This has been done in a number of laboratory studies.
Experiments on group judgment by Janet A. Sniezek and her colleagues show how different group and individual decision making can be. These studies examine two aspects of group performance in judgment tasks. The first, accuracy, refers to the proximity of the consensus group judgment of some unknown value to the actual value. The other performance measure is one of confidence. Results show that consensus group judgments are typically far more accurate, and somewhat more confident, than the independent judgments of two or more comparable individuals.
Two factors appear to be related to an increase in judgment accuracy in groups. One is the tendency of some groups to develop group judgments that are quite different from the members’ individual judgments. A representative study asked students to judge various risks by estimating the annual frequency of deaths in the United States from each of several causes. A minority of group judgments were either higher or lower than all the members’ individual estimates. For example, individual members’ estimates for a given cause of death were 300, 500, and 650, but the consensus of these persons as a group was 200. This phenomenon is often associated with great gains in group judgment accuracy, but unfortunately, such radical shifts in judgment as a result of grouping can also lead to extreme process loss. Some groups become far more inaccurate than their average members by going out of the range of individual estimates.
The second factor that is related to improvements in the accuracy of group judgment is heterogeneity (variety in a group) within the group. On the whole, groups that begin with a wide variety of judgments improve more in comparison to their average members than groups that begin with more homogeneity. This supports the creation of groups with members from different ethnic, racial, religious, or educational backgrounds. Such differences are likely to promote heterogeneity because the members of the group will have different sources of information and varying perspectives. Groups that lack sufficient heterogeneity face the danger of merely averaging their individual contributions. The result of averaging is to fail to improve appreciably in comparison to the level of quality of the average member.
There have been many efforts to identify procedures that are better than discretionary procedures in improving the quality of group decisions for optimizing tasks. Many group techniques have been developed in an attempt to eliminate factors that are thought to contribute to process loss. The most popular techniques are designed to alter group discussion. Some inhibit normative influence by restricting the extent to which group members can reveal their preferences. Instead, group discussion is limited—at least initially—to a thorough evaluation of all options.
Other techniques for enhancing group decision making are designed to suppress the extent to which members are influenced by irrelevant factors. For example, status effects can operate in groups, causing the person with the highest status to exert a greater influence on the group decision than other members. This is undesirable if the high-status person’s judgments are no more accurate—or even less accurate—than those of other group members. Other factors that have been shown to be irrelevant include the amount of participation in group discussion and self-confidence. Perhaps the most well-known technique developed to maximize informational influence and minimize irrelevant influences is the Delphi technique. This procedure prevents any potential problems of noninformational influence by not allowing face-to-face interaction. Instead, group members are given periodic anonymous feedback about the current positions of other group members. Often, group members provide information and logic to support their positions. More advanced technology, such as that available with computerized group-decision support systems, has greatly expanded the ability to control group decision-making processes.
In addition to the goal of improving the quality of group decision making, some theorists have stressed the importance of increasing group members’ satisfaction with their decisions. This objective remains somewhat controversial, because it is not always the case that people are more satisfied with better decisions and less satisfied with inferior ones. Ironically, people appear to be most satisfied when given the opportunity for group discussion—though this is precisely what is often eliminated in the hope of improving group decision quality.
Nevertheless, high group confidence can be an important end in itself. Presumably, groups with more confidence in their decisions are more committed to implementing them successfully. The increasing use of groups for decision making in organizations is based in part on this principle. Confidence in decision making can be increased by encouraging the participation of employees from various segments and levels of the organization. With such participative decision making, not only is confidence increased, but also the number of organizational members who support the decision.
Historically, scientific interest in group phenomena in general has been linked to social movements. Group research seems to thrive in the “we” decades, compared with the “me” decades. Interest specifically in group decision making, however, has tended to be stimulated by political and economic events.
Janis carefully analyzed decision making by groups in President John F. Kennedy’s administration. He diagnosed numerous problems regarding the way in which the Bay of Pigs Invasion of 1961 was handled during group meetings. Collectively, the symptoms represent “groupthink,” a narrow-minded approach to decision making that is caused primarily by a strong attachment of the group to its prevailing viewpoints. Janis shows how Kennedy altered the group decision process to accommodate and stimulate diverse perspectives during meetings about the Cuban Missile Crisis one-and-a-half years later.
For many reasons, the study of decision making by groups is likely to become increasingly important to psychologists. As a result of the collectivist nature of most cultures, the discipline of psychology will need to become less individualistic as it grows in non-Western societies and as social psychology expands to encompass more cultures. In addition, the growing interdependence of nations means that more and more global decisions are being made by groups of leaders, not by individual leaders.
Events within the United States can also be expected to create further demand for scientific investigation of group decision making. American organizations are using groups to a larger extent than ever before. For example, the group meeting is the most common approach to forecasting within organizations. The movement toward group decision making has been influenced in part by the apparent success of groups in Japanese firms. The desire to provide greater representation of workers in decisions should result in the increased use of groups for decision making.
While group decision-making research and applications are encouraged by national and global changes, these are insufficient to bring about a genuine revolution in the making of decisions. There must also be an increase in the capacity to use groups. Here, too, it is reasonable to expect developments that support the use of groups in decision making. Major advances in communications allow more people to participate in the decision process in a timely fashion. These advances also have the potential for creating techniques that lead to higher-quality decision making than can be provided by traditional meetings.
Alter, Adam. "The Folly of Crowds." Psychology Today. Sussex, 22 Aug. 2011. Web. 20 May 2014.
"Are Six Heads as Good as Twelve?" American Psychological Association. American Psychological Assn., 28 May 2004. Web. 20 May 2014.
Ariely, Dan. Predictably Irrational: The Hidden Forces That Shape Our Decisions. New York: Harper, 2010. Print.
Cannon-Bowers, Janis, and Eduardo Sales, eds. Making Decisions under Stress: Implications for Individual and Team Training. Washington: Amer. Psychological Soc., 2000. Print.
Davis, James H. “Social Interaction as a Combinatorial Process in Group Decision.” Group Decision Making. Ed. H. Brandstatter, James H. Davis, and Gisela Stocker-Kreichgauer. London: Academic, 1982. Print.
Guzzo, Richard A., ed. Improving Group Decision Making in Organizations. New York: Academic, 1982. Print.
Hastie, Reid. “Experimental Evidence on Group Accuracy.” Decision Research. Vol. 2. Ed. B. Grofman and G. Owen. Greenwich: JAI, 1986. Print.
Janis, Irving Lester. Victims of Groupthink. Boston: Houghton, 1972. Print.
Laughlin, Patrick R. Group Problem Solving. Princeton: Princeton UP, 2011. Print.
McGrath, Joseph Edward. Groups: Interaction and Performance. Englewood Cliffs: Prentice, 1984. Print.
Parks, Craig D., and Lawrence J. Sanna. Group Performance and Interaction. Boulder: Westview, 1999. Print.
Sniezek, Janet A., and Rebecca A. Henry. “Accuracy and Confidence in Group Judgment.” Organizational Behavior and Human Decision Processes 43 (1989): 1–28. Print.
Steiner, Ivan Dale. Group Process and Productivity. New York: Academic, 1972. Print.
Surowiecki, James. The Wisdom of Crowds. New York: Anchor, 2005. Print.
Witte, Erich H., and James H. Davis, eds. Understanding Group Behavior: Consensual Action by Small Groups. Vol. 1. New York: Psychology, 2014. Print.