In everyday life, many decisions are made by groups. Some of these group decisions are relatively inconsequential; however, others serve highly critical functions, such as those made by juries, medical teams, political committees, and safety advisory boards. Therefore, much research has been carried out on the determinants and dynamics of group decision making.
Group Decision Making History
The scholarly analysis of group decision making can be traced as far back as the philosophies of Socrates and Aristotle. The Socratic dialogue, for example, is predicated on the assumption that collective discourse can lead to greater truths than can solitary reflections. While there are some laudable exceptions, it was not until the emergence of social psychology in the 1930s that the study of group decision making took on its contemporary shape. This approach, initiated by such luminaries as Kurt Lewin, Muzafer Sherif, and Floyd Allport, emphasizes the collection of scientific evidence. Notably, while social psychology is the discipline most closely associated with the establishment and current study of group decisions, many important contributions have come from other fields, including sociology, business, education, and political science.
An early debate centered on the basic question of whether or not groups could be considered “real” in the sense of having scientifically measurable properties that transcend their individual members. Allport argued that they did not. He contended that groups could be wholly understood by studying their member individuals and that this was the only scientifically valid position. Others, including Lewin and Sherif, argued to the contrary. Through the collection of empirical data and persuasive theorizing, the debate was largely settled by the 1950s in favor of the groups approach, although vestiges of the rift remain to this day.
Some early studies on group decision making identified pseudo-group effects, outcomes that seem to emerge from groups but are attributable to statistical principles. For example, when several people combine their inputs to generate a single decision, they will reliably outperform individuals working alone. However, this effect can be attributed to the statistical principle of aggregation. Increasing the size of a group can also lead to a better outcome simply because it increases the probability that one of the individuals will have the requisite skills or knowledge. While such effects are real, they can be explained with statistical principles and are generally rejected as true group effects. However, this does not mean that such effects are unimportant or uninteresting. Indeed, mathematical models of group decision making that incorporate such considerations continue to be developed. Nonetheless, for most contemporary researchers, a true group effect necessarily involves individuals who are interdependent and engaged in social interaction.
Studying Group Decision Making
The methods used to study group decision making include experimental designs, which allow for the systematic manipulation and control of variables, and correlational designs, in which naturally occurring variables are carefully measured (but not manipulated) to see if they are reliably associated. Field studies of actual groups (e.g., committees, juries, clubs, fraternities, teams) typically employ correlational designs but may include experimental variables as well. A case study is an in-depth descriptive analysis of a single group, often one that has made a notorious decision. Each of these basic research designs offers unique advantages and disadvantages. For example, experimental designs allow for superior confidence in determining causality (e.g., Does time pressure cause group tension?); correlational designs allow researchers to study variables of interest that cannot be manipulated (e.g., gender, personality); and case studies, while not suitable as scientific evidence, can provide fascinating illustrations of established principles or may serve to stimulate scientific investigations.
Researchers in this field also rely on various measurement strategies. Self-report measures involve directly asking group members questions designed to tap variables of interest (e.g., How much did you enjoy the group interaction?). Objective outcome measures include, for example, the quality of a final decision, the length of group discussion, or the tally of votes for a proposed decision. Researchers may also assess ongoing processes by employing a structured observational measure. Such measures require trained observers to code and classify specified bits of behavior that emerge during a group discussion. For example, Robert Bales’s influential coding system, called interaction process analysis (IPA), consists of social-emotional categories (e.g., displays of solidarity) and task categories (e.g., asking for suggestions). The IPA remains popular and has served as the foundation for numerous theoretical and methodological advancements, such as the System of Multiple Level Observation of Groups (SYMLOG).
Group Decision-Making Process
According to much research, most decision-making groups proceed through three stages: orientation (defining the situation and procedures), evaluation (discussion of ideas, opinions), and decision (deciding what to do). A host of alternative stage models have been proposed, the most notable of which suggests four phases: forming (task identification), storming (dealing with conflict, emotional reactions), norming (developing cohesion, expressing opinions), and performing (solving the problem). Despite variations in terminology and specified number of stages, the various models share the idea of an initial phase that defines the problem, a middle phase (or phases) that involves working on the problem, and a final stage in which a decision is made.
There is clear evidence for value of the initial stage. Groups that devote relatively more time and effort to orientation issues generally produce higher-quality solutions and are generally more satisfied with the interactions and end result. Unfortunately, research also indicates that most groups seldom discuss orientation issues or strategies.
For middle stage processes, it has been demonstrated that varied contributions, critical appraisals, expressions of commitment, and ongoing assessments of performance all facilitate effective group decisions. However, studies have shown that groups also frequently engage in counterproductive processes at this stage, such as procrastinating, ignoring plausible solutions, withholding critical comments, trivializing the discussion, or avoiding responsibility for the decision.
The most notable aspect of the final stage is the implementation of a decision rule (or social decision scheme) that dictates how the preferences of individual group members will be combined to generate a single collective decision. The most prominent explicit decision rule across all types of human groups is the majority rule (or the closely related plurality rule). This is one of several voting rules, in which each group member receives one vote and the alternative with the most votes is adopted by the group. There is good evidence that the majority rule yields the most efficient and accurate outcomes across a range of conditions. Other explicit rules include consensus (discussion, usually with compromise, until unanimous agreement is reached), averaging individual inputs (some midpoint of the expressed preferences is calculated), and delegation (an individual group member, such as a leader or expert, or subgroup, is given authority to decide for the group). The “truth wins” rule is an implicit rule in which the correct decision emerges and is adopted, as its correctness is recognized by the group as a whole.
Little research has been conducted on how decisions, once made, are actually implemented or carried out. There is evidence that implementation is more successful if group members are closely involved in the decision-making process. Reluctance is more likely if group members are simply ordered to implement a decision that they had no role in determining. Recognizing this potential, many organizations now utilize such strategies as quality circles, autonomous work groups, self-directed teams, and total participation groups.
Group polarization refers to the well-established principle that, after a group discussion, people tend to take more extreme positions compared to their prediscussion inclinations. Thus, for example, a group of moderately prejudiced individuals will become more strongly prejudiced after group discussion, whereas a group of individuals somewhat low in prejudice will become even less prejudiced after a discussion. The implications for group decision making are clear: A group of individuals leaning somewhat toward a particular decision (e.g., to initiate a conflict, make an investment, declare a defendant guilty, not hire a candidate) is likely to become more solidified and extreme in that position as a function of group discussion.
After decades of research, two primary determinants of group polarization have been identified. According to the persuasive arguments or informational influence explanation, the group discussion generates a large pool of arguments or information that supports the initial proclivities of the group members. This exposure strengthens each member’s confidence in the correctness of his or her view, which leads to a more extreme stance. Thus, this explanation stresses the desire to be correct as a motivating force. The social comparison or normative influence explanation suggests that group members determine what an appropriate stance is by comparing their own views with the views of others. They then tend to shift their views to be more in line with that of the group as a whole. This explanation emphasizes the desire to be liked or to be held in high esteem, which acts as a motivating force.
Which of two major determinants will have a stronger effect depends, in part, on the type of decision under discussion. Research suggests that persuasive arguments have a greater influence on groups deliberating intellective tasks (problems with relatively objective, factual solutions) or when the group is more task- than friendship-based. Social comparison processes have a greater impact on groups dealing with ambiguous or judgmental tasks (problems with a relative emphasis on values, tastes, preferences), or when the individual’s group identity is more salient than his or her personal identity.
The term groupthink was brought into prominence by researcher Irving Janis who examined several infamously bad group decisions, including those associated with the Pearl Harbor bombing, the Vietnam War, and the Bay of Pigs invasion. According to Janis, groupthink occurs when the members of a group are so intent on reaching unanimity regarding a decision that they fail to critically appraise the potential flaws of their decision or to seriously consider alternative courses of action. Since Janis’s work, the concept has been used to explain several other disastrous real-world group decisions, including the launching of the Space Shuttle Challenger and the intelligence failures leading up to 9/11 and the Iraq war.
In theory, the following factors increase the likelihood of groupthink: a high degree of social cohesion or comradery among the group members, isolation of the group from outside scrutiny, and a biased group leader who strongly favors and promotes a particular decision outcome. The symptoms of groupthink include an illusion of invulnerability and morality (“We can’t possibly fail” and ‘We are morally justified in our decision”), constrained flow of information (don’t rock the boat), self-censorship (opposing opinions are avoided, minimized, or ridiculed), mindguards (group members who squelch dissent), and faulty analyses of goals, processes, and information.
Although the theory of groupthink has been quite influential and has enjoyed popular appeal, direct empirical evidence for it is mixed. This is partly due to the fact that the highly stressful and consequential situations upon which the theory was developed are difficult to replicate in laboratories that could allow for precise scientific study. Many contemporary researchers have opted to incorporate aspects of the theory into more general models of group decision making.
Minority influence refers to those instances when a group’s decision is substantially influenced by the views of an individual or a small subset of individuals that are not in line with the views of most group members. A good fictional illustration of this phenomenon can be seen in the 1957 movie Twelve Angry Men. In this film, 11 jurors quickly agree on the guilt of a defendant but are slowly influenced by the contrary views of a lone juror. About a decade after the film was released, Serge Moscovici, a prominent European social psychologist, formulated a theory of minority influence and conducted a series of classic studies that empirically demonstrated its power. Subsequent studies have continued to demonstrate that a minority position can indeed change the viewpoint of the majority.
Research has shown that minority influence is most likely to occur when the person or persons holding the converse opinion are steadfast in their views, but do not appear overly dogmatic or rigid, and are willing to compromise. Minority views also are more likely to have an influence on the majority if they offer a compelling argument against the majority’s position, the minority position is held by more than one group member, and there is not an obvious selfish explanation for the minority position (e.g., the minority would benefit financially).
Even under the best of circumstances, a minority viewpoint may not be accepted. And even if accepted privately, publicly expressed acceptance may be hindered by the fear of disapproval by the larger group or powerful leaders. Nevertheless, there is good evidence that even if a minority position is not fully or even partially adopted immediately, the process may stimulate more in-depth and creative thinking about the issues under consideration and can lead to more long-term shifts in opinions.
Information Processing in Groups
To help explain certain decision-making processes, some researchers conceptualize decision-making groups as collective information processors. One prominent model in this vein considers situations in which different members of a group are responsible for different domains of knowledge. Their combined cognitive effort of collecting, analyzing, and communicating information is termed a transactive (or collective) memory system (TMS). In short, a TMS is a cooperative division of mental labor. Research suggests that such systems have limited benefits with newly established or short-term groups but do benefit long-term groups. It seems that as a group stays together over time, the members become more proficient at coordinating their cognitive efforts, more trusting in their mutual reliance, and typically improve in their decision-making performance.
Relatedly, the information sampling model was developed, in part, to examine the commonly held assumption that group members tend to pool their unique bits of knowledge and this leads to higher-quality decisions. Indeed, studies confirm that the extent to which unshared information (information held by only one or a few members of the group) is discussed is a good predictor of ultimate decision quality. However, consistent with the model’s predictions, studies have also found that groups tend to spend an inordinate amount of time discussing shared information (information that each member of the group possessed) and very little time discussing unshared information. This information sampling bias leads to faulty decision-making outcomes (sometimes referred to as the common knowledge effect). There is some evidence that the sampling bias can be mitigated if the decision task is intellective rather than judgmental and if the group is motivated to generate the correct solution.
- Baron, R. S. (2005). So right it’s wrong: Groupthink and the ubiquitous nature of polarized group decision making. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 37, pp. 219-253). San Diego, CA: Academic Press.
- Forsyth, D. R. (2006). Group dynamics (4th ed.). Belmont, CA: Wadsworth.
- Hastie, R., & Kameda, T. (2005). The robust beauty of majority rules in group decisions. Psychological Review, 112, 494-508.
- Kerr, N. L., & Tindale, S. R. (2004). Group performance and decision-making. Annual Review of Psychology, 55, 623-655.
- Nemeth, C. (1986). Differential contributions of majority and minority influence. Psychological Review, 93, 23-32.