Small-group communication

Small-group communication refers to the nature of communication that occurs in groups that are between 3 and 7 individuals. Small group communication generally takes place in a context that mixes interpersonal communication interactions with social clustering.

Nature of small groups

Small groups have been described and studied in a number of contexts such as work groups, parties and other event-based or social situations. Most small groups fall under one or more categories: The task group, the relationship group and the influence group. The task group is more of a committee meeting, set up to accomplish a task, such as a Local School Council. The relationship group is more of a meeting of people for the enjoyment of each other's company, such as a gourmet club. The influence group focuses on bettering yourself and those around you, group members have influence over each other, such as a self help group.

Defining small groups

In the context of interpersonal communication, it is difficult to define a "small group." Scores of competing definitions have appeared in different research traditions, most concentrating on one aspect of small groups at the expense of others. Among the characteristics of groups that have underlaid definitions are the following:
*interdependence that develops over time among group members. [Lewin, K. (1948). "Resolving social conflicts". New York: Harper.]
*extent to which communication patterns among members become predictable. [Fisher, B. A. (1974). "Small group decision making". New York: McGraw-Hill.]
*emergence of a structure of roles and norms. [Newcomb, T. M. (1951). Social psychological theory. In J. H. Rohrer and M. Sherif (Eds.), "Social psychology at the crossroads" (pp. 31-49). New York: Harper.]
*the functions (in the sense of structural functionalism)performed by the group. [McDavid, J. W. & Harari, H. (1968). "Social psychology". New York: Harper.]
*extent to which the group satisfies group member needs or rewards them. [Cattell, R. B. (1951). New concepts for measuring leadership, in terms of group syntality. "Human Relations, 4," 161-184.]
*extent to which members perceive themselves as a group. [Smith, M. (1945). Social situation, social behavior, social group. "Psychological Review, 52," 224-229.]

Similarly, attempts to distinguish small groups from large groups through number of members is arbitrary; social psychologist Robert Bales claimed that the former differ from the latter in that each participant in a small group can remember each other participant's presence.Bales, R. F. (1950). "Interaction process analysis". Page 33. Cambridge, MA: Addison-Wesley.]

Input-Process-Output model

The things which affect small-group communication are best conceived of in terms of the "input-process-output" (hereafter I-P-O) model. Input factors are those that exist before a group meets; process factors are those that occur during a group meeting; output factors are the results of the meeting.

There are any number of important input factors; perhaps the most important is the group's task. Group tasks run the gamut from making decisions to performing projects to helping group members learn skills (training, or T-groups) to encouraging self help for members (sensitivity training groups; 12-step groups) to education; even informal groups of friends can be said to have a task if they perform activities together. Other input factors include the personalities of the members, the length of time the group has existed and the extent to which the group follows a formal procedure for performing its task.

The most important process factor is the communication that occurs during group meetings, although there are others (e.g., a group working together building a shed must coordinate their shed-building actions and not get in each other's way).

Output factors can be divided into two major categories. One category includes those relevant to the group's task, such as how much is accomplished, the speed at which the task is completed and the quality of the product. The other category is relevant to the relationships among the members of the group and the individual perceptions of the group members. These include most notably; cohesiveness, which can be defined as the extent to which group members are attracted to and identify with the group, along with individual member morale and degree of satisfaction with the experience. In addition, groups develop role and leadership structures over time; these structures along with group cohesiveness affect later group meetings and as a consequence also serve as input factors.

Group communication

The first important research study of small group communication was performed by social psychologist Robert Bales and published in a series of books and articles in the early and mid 1950s .Bales, R. F. (1950). "Interaction process analysis". Cambridge, MA: Addison-Wesley.] [Bales, R. F., and Strodtbeck, F. L. (1951). Phases in group problem-solving. "Journal of Abnormal and Social Psychology, 46," 485-495.] This research entailed the content analysis of discussions within groups making decisions about "human relations" problems (i.e., vignettes about relationship difficulties within families or organizations). Bales made a series of important discoveries. First, group discussion tends to shift back and forth relatively quickly between the discussion of the group task and discussion relevant to the relationship among the members. He believed that this shifting was the product of an implicit attempt to balance the demands of task completion and group cohesion, under the presumption that conflict generated during task discussion causes stress among members, which must released through positive relational talk. Second, task group discussion shifts from an emphasis on opinion exchange, through an attentiveness to values underlying the decision, to making the decision. This implication that group discussion goes through the same series of stages in the same order for any decision-making group is known as the "linear phase model". Third, the most talkative member of a group tends to make between 40 and 50 percent of the comments and the second most talkative member between 25 and 30, no matter the size of the group. As a consequence, large groups tend to be dominated by one or two members to the detriment of the others.

Linear phase model

The most influential of these discoveries has been the latter; the linear phase model. The idea that all groups performing a given type of task go through the same series of stages in the same order was replicated through the 1950s, 1960s and 1970s; with most finding four phases of discussion. For example, communication researcher B. Aubrey Fisher showed groups going sequentially through an orientation stage, a conflict stage, a stage in which a decision emerges and a stage in which that decision is reinforced. [Fisher, B. A. (1970). Decision emergence: Phases in group decision making. "Speech Monographs, 37," 53-66.] Much of this research (although not necessarily Fisher's) had two fundamental flaws. First, all group data was combined before analysis, making it impossible to determine whether there were differences among groups in their sequence of discussion. Second, group discussion content was compared across the same number of stages as the researcher hypothesized, such that if the researcher believed there were four stages to discussion, there was no way to find out if there actually were five or more. In the 1980s, communication researcher Marshall Scott Poole examined a sample of groups without making these errors and noted substantial differences among them in the number and order of stages. [Poole, M. S., & Roth, J. (1989). Decision development in small groups IV: A typology of group decision paths. "Human Communication Research, 15," 323-356.] He hypothesized that groups finding themselves in some difficulty due to task complexity, an unclear leadership structure or poor cohesion act as if they feel the need to conduct a "complete" discussion and thus are more likely to pass through all stages as the linear phase model implies, whereas groups feeling confident due to task simplicity, a clear leadership structure and cohesion are more likely to skip stages apparently deemed unnecessary.

Idea development

Another milestone in the study of group discussion content was early 1960s work by communication researchers Thomas Scheidel and Laura Crowell regarding the process by which groups examine individual proposed solutions to their problem. [Scheidel, T. M., & Crowell, L. (1964). Idea development in small discussion groups. "Quarterly Journal of Speech, 50," 140-145.] They concluded that after a proposal is made, groups discuss it in an implied attempt to determine their "comfort level" with it and then drop it in lieu of a different proposal. In a procedure akin to the survival of the fittest, proposals viewed favorably would emerge later in discussion, whereas those viewed unfavorably would not; the authors referred to this process as "spiralling." Although there are serious methodological problems with this work, other studies have led to similar conclusions. For example, in the 1970s, social psychologist L. Richard Hoffman noted that odds of a proposal's acceptance is strongly associated with the arithmetical difference between the number of utterances supporting versus rejecting that proposal. More recent work has shown that groups differ substantially in the extent to which they spiral. [Hoffman, L. R. (1979). "The group problem-solving process". New York: Praeger.]

None of this work has attempted to link discussion content with task output. The most successful attempt at that can be found in a 1980s research program of communication researcher Randy Y. Hirokawa. [Hirokawa, R. Y. (1985). Discussion procedures and decision-making performance. "Human Communication Research, 12," 203-224.] The implication of this program is that to an extent, depending upon task, the quality of a group's decision appears to be associated with the extent to which the group examines the problem it faces, identifies the requirements of an ideal solution and evaluates the positive and negative features of proposed solutions. Although this reads like Bales's linear phase model, Hirokawa (like Poole at about the same time) demonstrated that these decision functions need not occur in any particular order. Communication researchers Renee Meyers and Dale Brashers have also had some success in correlating group decisions with the pattern of arguments (in the sense of argumentation theory) that occur during discussion. [Meyers, R. & Brashers, D. (1998). Argument in group decision-making: Explicating a process model and investigating the argument-outcome link. "Communication Monographs, 65," 261-281.]

Social influence in groups

Work relevant to social influence in groups has a long history. Two early examples of social psychological research have been particularly influential. The first of these was by Muzafer Sherif in 1935 using the autokinetic effect. Sherif asked participants to voice their judgments of light movement in the presence of others and noted that these judgments tended to converge. [Sherif, M. (1935). A study of some social factors in perception. "Archives of Psychology, 27"(187).] The second of these was a series of studies by Solomon Asch, in which naive participants were asked to voice their judgments of the similarity of the length of lines after hearing the "judgments" of several confederates (research assistants posing as participants) who purposely voiced the same obviously wrong judgment. On about 1/3 of the cases, participants voiced the obviously wrong judgment. When asked why, many of these participants reported that they had originally made the correct judgment but after hearing the confederates, decided the judgments of several others (the confederates) should be trusted over theirs. [Asch, S. E. (1956). Studies of independence and conformity: 1. A minority of one against a unanimous majority. "Psychological Monographs, 70"(9), Whole #416.] As a consequence of these and other studies, social psychologists have come to distinguish between two types of social influence; informational and normative (see conformity). Informational influence occurs when group members are persuaded by the content of what they read or hear to accept an opinion; Sherif's study appears to be an example. Normative influence occurs when group members are persuaded by the knowledge that a majority of group members have a view. Normative influence should not be confused with compliance, which occurs when group members are not persuaded but voice the opinions of the group majority. Although some of the participants in the Asch studies who conformed admitted that they had complied, the ones mentioned above who believed the majority to be correct are best considered to have been persuaded through normative influence.

Group decisions

By the end of the 1950s, studies such as Sherif's led to the reasonable conclusion that social influence in groups leads group members to converge on the average judgment of the individual members. As a consequence, it was a surprise to many social psychologists when in the early 1960s, evidence appeared that group decisions often became more extreme than the average of the individual prediscussion judgments. [Wallach, M. A., Kogan, N., & Bem, D. J. (1962). Group influence on individual risk taking. "Journal of Abnormal and Social Psychology, 65," 75-86.] This was originally thought to be a tendency for groups to be riskier than their members would be alone (the risky shift), but later found to be a tendency for extremity in any direction based on which way the members individually tended to lean before discussion (group polarization). Research has clearly demonstrated that group polarization is primarily a product of persuasion not compliance. Two theoretical explanations for group polarization have come to predominate. One is based on social comparison theory, claiming that members look to one another for the "socially correct" side of the issue and if they find themselves deviant in this regard, shift their opinion toward the extreme of the socially correct position. [Baron, R. S., Dion, K. L., Baron, P. H., & Miller, N. (1971). Group consensus and cultural values as determinants of risk taking. "Journal of Personality and Social Psychology, 20," 446-455.] This would be an example of normative influence. The other 'persuasive arguments theory' (PAT), begins with the notion that each group member enters discussion aware of a set of items of information favoring both sides of the issue but lean toward that side that boasts the greater amount of information. Some of these items are shared among the members (all are aware of them), others are unshared (only one member is aware of each). Assuming most or all group members lean in the same direction, during discussion, items of unshared information supporting that direction are voiced, giving members previously unaware of them more reason to lean in that direction. [Vinokur, A., & Burnstein, E. (1974). Effects of partially shared persuasive arguments on group induced shifts: A group problem-solving approach. "Journal of Personality and Social Psychology, 29," 305-315.] PAT is an example of informational influence. Although PAT has strong empirical support, it would imply that unshared items of information on the opposite side of the favored position would also come up in discussion, cancelling the tendency to polarize. Research has shown that when group members all lean in one direction, discussion content is biased toward the side favored by the group, inconsistent with PAT. This finding is consistent with social comparison notions; upon discovering where the group stands, members only voice items of information on the socially correct side. It follows that an explanation for group polarization must include information influence and normative influence.

The possibility exists that the majority of information known to all group members combined, supports one side of an issue but that the majority of information known to each member individually, supports the other side of the issue. For example, imagine that each member of a 4-person group was aware of 3 items of information supporting job candidate A that were only known to that member and 6 items of information supporting job candidate B that were known to all members. There would be 12 items of information supporting candidate A and 6 supporting candidate B but each member would be aware of more information supporting B. Persuasive arguments theory implies that the items of information favoring A should also come up, leading to each member changing their mind but research has indicated that this does not occur. Rather, as predicted by the merging of PAT and social comparison theory, each member would come into discussion favoring B, that discussion would be heavily biased toward B and that the group would choose B for the job. This circumstance, first studied by Stasser and Titus, is known as a "hidden profile" and is more likely to occur as group size increases and as the proportion of shared versus unshared items of information increases. [Stasser, G., & Titus, W. (1985). Pooling of unshared information in group decision making: Biased information sampling during discussion. "Journal of Personality and Social Psychology, 53," 81-93.]

References

* Homans, George Kaspar (1974), "Social Behavior: Its Elementary Forms" (Rev. ed.), Harcourt Brace Jovanovich.

See also

*Group Dynamics