Group Development

Group Dynamics

Donelson Forsyth

Forsyth, D. (2018). Group Dynamics (7th ed.). Cengage Learning US.  https://mbsdirect.vitalsource.com/books/9798214344799

Chapter 5. Cohesion and Development

A group is not just a set of individuals, but a cohesive whole that joins the members together in interlocking interdependencies. This solidarity or unity is called group cohesion and is a necessary, if not sufficient, condition for a group to exist. A group may begin as a collection of strangers, but, as uncertainty gives way to increasing unity, the members become bound to their group and its goals. As cohesion and commitment ebb and flow with time, the group’s influence over its members rises and falls.

What is group cohesion, and what are its sources?

How does cohesion develop over time?

What are the positive and negative consequences of cohesion?

Do initiations increase cohesion?

The U.S. Olympic Hockey Team: Miracle Makers

They were underdogs, and they knew it. Their mission: To represent the United States in the 1980 Winter Olympics hockey competition held in Lake Placid, New York. Their task: To defeat teams from such hockey-rich countries as Sweden and Germany. Their goal: To win a medal, preferably a gold one. Their major obstacle: The powerful U.S.S.R. National Championship Team. At a time when Olympic athletes were amateurs, the Russian players were practically professionals. They were members of the Russian army, and they were paid to practice and play their sport. The Russian team had dominated hockey for many years and was poised to take its fifth consecutive gold medal in the sport. In an exhibition game played on February 9, just a few days before the start of the Olympic Games, Russia defeated the U.S. Olympic hockey team 10 to 3.

But strange things can happen when groups compete against groups. The U.S. team made its way through the preliminary rounds and faced the Russian team in the medal round. The U.S. team fell behind by two goals, and it looked as though the Russians would take the victory with ease. But the Americans struggled on, finally tying, and then taking the lead. During the game’s last minutes, the Russians launched shot after shot, but all the while the U.S. coach, Herb Brooks, calmed his players by telling them “Play your game!” As the game’s end neared, the announcer counted down the seconds into his microphone before asking his listeners, “Do you believe in miracles?” What else could explain the game’s outcome? The U.S. Olympic hockey team had just beaten an unbeatable team.

The U.S. team was inferior to the Russian team in nearly all respects. The Americans were mostly college students or recent graduates. They were smaller, slower, and far less experienced. The team had practiced diligently for months before the tournament, but the Soviets had been playing as a team for years. Yet for all their relative weaknesses, they were stronger than the Russian team in one key way: They were more cohesive. They were unified by friendship, a sense of purpose, and esprit de corps. No one player took credit for the victory, but instead spoke only of “we,” repeating “we beat those guys” over and over as the bewildered Russian team looked on. As a writer for the magazine Sports Illustrated explained (Swift, 1980, p. 32):

Individually, they were fine, dedicated sportsmen…. But collectively, they were a transcendent lot. For seven months they pushed each other on and pulled each other along, from rung to rung, until for two weeks in February they—a bunch of unheralded amateurs—became the best hockey team in the world. The best team. The whole was greater than the sum of its parts by a mile.

The U.S. Olympic team that faced the Russians on that February day in Lake Placid had many noteworthy qualities. The players were handpicked to represent their country, and they had trained diligently for months leading up to the game. They were each one highly skilled, for the majority went on to professional careers in hockey after the Olympics. The team’s coach was known for his hard-driving style of leadership, and each player could tell more than one story about the indignities visited upon them by the coach in his dogged pursuit of excellence. But most of all, they were cohesive, and many believed that the team’s cohesiveness was the deciding factor in their victory.

Group cohesion can lay claim to being group dynamics’ most theoretically important concept. Uniquely group-level, cohesion comes about if, and only if, a group exists. Cohesiveness signals, if only indirectly, the health of the group. A cohesive group will be more likely to prosper over time, since it retains its members and allows them to reach goals that would elude a more incoherent aggregate. The group that lacks cohesion is at risk, for it may break into subgroups at the first sign of conflict, lose members faster than it can replace them, and fail to reach its agreed upon goals. The concept of group cohesion provides insight into a host of core processes that occur in groups, including productivity, members’ satisfaction and turnover, morale, formation, stability, influence, and conflict.

5-1. Sources of Cohesion

The Latin word haesus means to cling to; it is the basis of such words as adhesive, inherit, and, of course, cohesive. In physics, things that are cohesive are made up of particles that are bonded together so tightly that they remain in place rather than drifting off or adhering to some other object. Similarly, when human groups are cohesive, the members stick together rather than leave to join other groups. They are joined together by strong interpersonal bonds and by a shared commitment to the group and its goals. Cohesive groups remain united, over time and across situations, whereas less cohesive groups experience frequent changes in their membership, their processes, and their procedures (Cartwright, 1968).

Cohesive groups are unified groups, but their unity is often the result of different causes and processes. Consider, for example, an executive board of a company that is productive and enduring, yet the members never associate with one another after work. In fact, most dislike one another. In contrast, another group may be completely unproductive, but the members are so closely interconnected emotionally that they can move from one problematic experience to another without a loss of synchrony. (Most of the groups who star in popular television series—Friends, Big Bang Theory, Modern Family—fit this category.) These groups may be equally cohesive, but their unity is the result of very different group processes.

The idea that no one condition or process is a necessary or sufficient condition for a group to become cohesive is consistent with systems theory’s principle of equifinality: “final states or objectives may be reached in different ways and from disparate starting points” (Skyttner, 2005, p. 71). But increases in cohesiveness are not entirely unpredictable. Recognizing that our review cannot be comprehensive, the following sections examine five overlapping, but influential, sources of a group’s unity:

Social cohesion: The attraction of members to one another and to the group as a whole.

Task cohesion: A shared commitment among members to achieve a goal and the resulting capacity to perform successfully as a coordinated unit.

Collective cohesion: Unity based on shared identity and belonging.

Emotional cohesion: Group-based emotions, including pride, esprit de corps, and overall affective intensity.

Structural cohesion: The group’s structural integrity, including normative coherence, clarity of roles, and strength and density of relationships linking members.

5-1a. Social Cohesion

The members of the U.S. Olympic hockey team did not gather, as a group, just to play hockey. They lived, traveled, and partied together because their training regimen demanded it, but also because they liked each other. A group’s social cohesion is determined by how much the members like each other and the group itself (Lott & Lott, 1965).

Interpersonal Attraction

Social psychologists Kurt Lewin, Leon Festinger, and their colleagues conducted some of the earliest studies of group cohesion. Lewin (1948) used the term cohesion to describe the forces that keep groups intact by pushing members together as well as the countering forces that push them apart. Festinger and his colleagues also stressed binding social forces when they defined group cohesion as “the total field of forces which act on members to remain in the group” (Festinger, Schachter, & Back, 1950, p. 164). But in their studies of naturally forming groups, they focused on social cohesion: attraction to the group and its members. To measure cohesion, they asked the members to identify all their good friends and calculated the ratio of ingroup choices to outgroup choices. The greater the ratio, the greater was the cohesiveness of the group (see, too, McPherson & Smith-Lovin, 2002).

Attraction between individuals is a basic ingredient for most groups, but when these relations intensify and proliferate throughout a group, they can transform a run-of-the-mill group into a cohesive one. Social psychologists Muzafer and Carolyn Sherif documented this process in a series of unique field studies conducted in 1949, 1953, and 1954. During the summers of those years, the Sherifs ran camps for 11- and 12-year-old boys that seemed like typical summer camps—with canoeing, campfires, crafts, hikes, and sports. But, unbeknown to the campers, the Sherifs also recorded the behavior of the boys as they reacted to one another and to situations introduced by the investigators. In the 1949 study, conducted in northern Connecticut, the 24 campers all bunked in one cabin for three days. During this period, friendships developed quickly based on proximity of bunks, similarities in interests, and maturity. The Sherifs then intervened and broke the large group into two smaller ones by deliberately splitting up any friendship pairs that had formed. They assigned one best friend to one group (named the Bulldogs) and the other to the other group (the Red Devils).

Even under these conditions—with the factors that produced attraction between the boys minimized—new attractions formed quickly and resulted in two highly cohesive groups. The Sherifs made certain that the boys’ first few days in their new groups were spent in a variety of positive experiences (hiking, cookouts, and games), and before long, the boys, when asked to name their friends, chose members of their new groups rather than the boys they had liked when camp first began. When first split up, 65% of the boys picked as friends those in the other group. But when the groups became cohesive, fewer than 10% named boys as friends who were in the other group. The Sherifs’ well-known Robbers Cave Experiment, which is discussed in more detail in Chapter 14’s analysis of conflict between groups, yielded similar findings.

Is Your Group Cohesive?

For every group that never quite jells is a group of the unlikeliest of allies who become so interlocked that the members fit together like pieces of a jigsaw puzzle. What is this unseen “social chemistry” that transforms humdrum groups into tight-knit teams, cliques, squads, bands, and so on?

Instructions. To explore the concept of cohesion, think of a group you belong to—one that is significant for you personally. Then, put a check in each box if you think the statement accurately describes your group.

Social Cohesion: Attraction

Most of us really like this group.

People get along well in this group.

I like the people who are in this group.

I am friends with many of the members of this group.

Task Cohesion: Goals

We work well together to achieve group goals.

We work together diligently pursuing our goals.

I work enthusiastically in this group.

I am willing to work hard for this group’s goals.

Collective Cohesion: Unity

We stick together.

The group is a unified one.

I feel like a part of this group.

I identify with this group.

Emotional Cohesion: Affect

The group has a great amount of energy.

The group has team spirit.

I share in this group’s excitement (and disappointments).

I am proud of this group.

Structural Cohesion: Integrity

The group is well organized.

This group has a high level of structural integrity.

I understand my place in this group.

My role in this group is well defined.

Interpretation. Is your group a cohesive one? If you checked at least two of the four indicators for each type of cohesion, then the group can be considered cohesive—at least, cohesive in that particular form of cohesiveness. But if your group did not earn many checks then it cannot be considered cohesive. (Note: The first two indicators in each set pertained to group-level cohesion, whereas the second two were individual-level indicators of cohesion.)

Group-Level Attraction

Social cohesion increases when group members like each other. Cohesion, however, is a multilevel process, for group members may be bonded to each other, to their group, and to the organization in which their group is embedded. The players on the hockey team, for example, liked each other, but they also liked the team as a whole (Carless & De Paola, 2000). These two forms of social cohesion usually go hand in hand: if you like many individuals in the group, you likely also like the group itself. But, when cohesion is based only on individual-level attraction and those who are liked leave the group, the remaining members are more likely to quit. When cohesion is based on group-level attraction, people remain members even when specific members leave the group (Ehrhart & Naumann, 2004).

5-1b. Task Cohesion

Studies of task-oriented groups, such as teams, military squads, and expeditions, find that members, when asked to describe their team’s cohesiveness, stress the quality of their group as a performing unit. The U.S. Olympic hockey team players, for example, were not just individuals seeking personal goals, but teammates who combined their strengths and talents to create a single, performance-focused hockey team. They achieved task cohesion: a shared commitment among members to achieve a goal that requires the collective effort of the group (Severt & Estrada, 2015).

Group Motivation

Task cohesion is based on group-level goal motivation. Many of the players on the hockey team were the stars of their college teams, and when they played they wanted to do their personal best by scoring the most goals or defending their own net. But success in hockey is not based on personal performance. A good player may do much to help the team win, but success in hockey requires collaboration, so all members must contribute to the group and its objectives. Group members typically have the choice of working for the group, for themselves, for both the group and themselves, or for neither, and thus do not always choose to strive for group success. If, however, group cohesiveness is so strong that all members feel united in a common effort, then group-oriented motives should replace individualistic motives, and the desire among members for group success should be strong (Zander, 1971).

The coach of the hockey team, Herb Brooks, was careful to emphasize the importance of team goals rather than individual performance as he prepared his team for the Olympics. Rather than appealing to player-centered motivations by emphasizing personal performance and rewarding individual expertise, Brooks instilled a strong desire for group rather than individual success. He deliberately avoided developing personal relationships with the players and reminded each one frequently that, as a hockey player, he was expendable. As one of the players remarked in describing his coach: “He treated us all the same: rotten” (Swift, 1980, p. 32).

Collective Efficacy and Potency

Groups that are cohesive, in terms of task commitment, tend to exhibit high levels of collective efficacy and group potency. Collective efficacy is determined by members’ shared beliefs that they can accomplish all the components of their group’s tasks competently and efficiently. Unlike self-efficacy, which is confidence in one’s own abilities, collective efficacy is a group-level process: Most or all of the members must believe the members will competently coordinate their individual actions in a skilled, collective performance (Pescosolido, 2003). Hence, collective efficacy is “a group’s shared belief in its conjoint capabilities to organize and execute the courses of action required to produce given levels of attainment” (Bandura, 1997, p. 476). A similar construct, group potency, is a generalized positive expectation about the group’s chances for success (Guzzo et al., 1993). High-potency groups tend to select more difficult goals to pursue and they tend to outperform their less potent counterparts (Stajkovic, Lee, & Nyberg, 2009).

5-1c. Collective Cohesion

Brooks, the coach, did not just stress a sense of camaraderie or shared goals, but team unity. His goal, he explained, was to “build a ‘we’ and ‘us’ in ourselves as opposed to an ‘I,’ ‘me,’ and ‘myself’” (Warner HBO, 2001). The team’s collective cohesion reached its peak in the medal ceremonies after the U.S. team had won its gold medal. Team captain Mike Eruzione waved to the team to join him on the small stage, and somehow the entire team crowded onto the platform. Instead of the team captain representing the group, the entire group, as a whole, received the medal.

When a group is cohesive, collectively, members are united; fused to form a single whole. When members talk about themselves and their group, they use more plural pronouns than personal pronouns: “We won that game” or “We got the job done” rather than “I got the job done” (Cialdini et al., 1976). They use words like family, community, or just we to describe their group. They may also refuse to differentiate among the members of the group, as when one member refuses to take responsibility for the victory or win and insists that the team as a whole deserves the credit. Individuals who are members of cohesive groups—with cohesion defined as a strong sense of being a part of a larger whole—are more committed to their groups, where commitment is indicated by the degree of attachment to the group, a long-term orientation to the group, and intentions to remain within the group (Arriaga & Agnew, 2001). Physically, member stand closer together and they position themselves to prevent nonmembers from intruding on the group’s space (Knowles & Brickner, 1981).

Cohesion and Entitativity

A group that is higher in collective cohesion will, in most cases, be a group that is higher in entitativity: It will be perceived to be a single, unified entity that resists disintegration. As noted in Chapter 1, small, highly unified groups such as families, cliques of friends, gangs, and sports teams are all thought to be high in entitativity, and these are also the types of groups that are high in cohesiveness. Confirming this close association between cohesiveness and entitativity, one team of investigators found that level of entitativity, as measured by such items as “do you think about this collection of people as a number of individuals or as a whole” was correlated with both social cohesion (r = .55) and task cohesion (r = .67). They also found that qualities that were associated with increases in entitativity, such as the importance of the group for the members, the amount of interaction among members, and the duration of the group, also predicted increases in cohesion (Thurston, 2012).

Belonging and Identity

The members of the U.S. Olympic Hockey team did not just meld together to form a single, unified group—each individual player also came to strongly identify himself as a member of that group. Members of groups that are high in collective cohesiveness, when asked to comment directly on their sense of connection to the group, are more likely to say “I feel a sense of belonging to my group” (Bollen & Hoyle, 1990), “I think of this group as a part of who I am” (Henry, Arrow, & Carini, 1999), and “I see myself as a member of the group” (Smith, Seger, & Mackie, 2007). They not only consider their group to be a single, unified entity, but they also consider themselves to be a component part of that inseparable whole.

This increased identification of individuals with cohesive groups is predicted by social identity theory (Hogg, 1992, 2001). When a group is highly cohesive, members’ identities will be based more on their membership in that group rather than their own personal, unique qualities. In consequence, their sense of self will become depersonalized: They will view themselves and their fellow members as relatively interchangeable parts of the whole, and their sense of membership in the group will become more important to them than their personal relationships with specific group members. Any factor that increases members’ tendency to categorize themselves as group members (e.g., conflict with other groups, the presence of outgroups, and activities that focus members’ attention on their group identity) will reduce personal attraction but increase depersonalized attraction to the group as a whole.

Can Members Get Lost in Their Groups?

When individuals identify with their group, their sense of self combines elements drawn from both their individualized, personal self and their collective, group-level self. But what happens if this distinction between the individual self and collective self dissolves—if the individual’s personal self becomes fused with the collective? Identity fusion theory suggests that in such cases—which are admittedly rare—both the personal self and the collective selves become amplified, with the result that individuals are willing to engage in extreme forms of behavior on behalf of their group. Individuals who sacrifice themselves for their group—heroes in combat but also suicide bombers—perform actions that seem objectively inexplicable; they sacrifice their own lives to either save group members or to harm others who they believe are their enemies. Identity fusion theory suggests that individuals engage in these actions because their identification with their group is so great that they no longer distinguish between themselves and their group (Swann et al., 2012; Talaifar & Swann, 2016).

Researchers studied this process by surveying a large number of students, asking them if they agreed with such items as “I am one with my group” and “I feel immersed in my group” (paraphrased from Gómez et al., 2011, p. 992). Individuals who agreed with these items, when recontacted six months later, were asked if they were willing to fight and die for their group. Many said they would. Fusion also predicted how they responded to a modified version of the trolley-car dilemma used in studies of ethics. Participants were asked to imagine that they were on a footbridge and could see that a runaway trolley was about to run over five members of their group. The only thing they could do to stop the trolley would be to jump from the bridge into its path. More of the individuals with fused identifies said they would do so (Gómez et al., 2011).

5-1d. Emotional Cohesion

Napoleon is said to have proclaimed that the great strength of an army lies not in the skill of its leaders, but in the élan—the emotional intensity—of its members. The sociologist Émile Durkheim, in discussing the nature of ritualized interactions in cohesive groups, stressed how they develop intense emotional experiences, for when all “come together, a sort of electricity is formed by their collecting which quickly transports them to an extraordinary degree of exaltation” (1912/1965, p. 262). Durkheim was describing the large gatherings of local communities in New Guinea, but he believed that collective effervescence resulted from the sharing of emotional reactions within a group. As the positive and elevated mood of one person is picked up by the next, the group members eventually display a shared emotional intensity.

A variety of terms is used to describe group-level emotional states, including élan, morale, pride, esprit de corps, and positive affective tone, but no matter what its label, this shared emotional cohesion is one of the most obvious features of many unified groups. The Russian and U.S. teams were equal in confidence and collective efficacy, for both groups had the talent needed to win at hockey. But they differed dramatically in their level of emotionality. The Russian team was confident but unenthusiastic. The U.S. team was not so confident, but the team was brimming with energy, enthusiasm, and team spirit. A group with high levels of collective efficacy may expect to succeed, but a group with emotional cohesion has vitality, passion, vim, and vigor. It was this emotionality that Coach Brooks whipped up to its peak intensity before the U.S. team’s game with the Russians. He told them that the Russians were taking their victory for granted, but “we can beat them.” He told his team: “you were born to be a player,” you were “destined to be here today,” and this is “our time.” When he told them to “spit in the eye of the tiger,” they did.

Group-Level Emotions

Emotional cohesion, like the other components of cohesion, is a multilevel process. Emotions, although traditionally thought to be personal and private are more often interpersonal and socially shared (Menges & Kilduff, 2015). The management and organizational behavior researcher Jennifer George (1996), for example, suggests cohesive groups are more likely to display collective mood states; members’ emotions and moods become synchronized, as if they had reached consensus on the feelings they should be experiencing. This group affective tone is not tied to any specific aspect of the group’s activities or to any one individual, but rather pervades all the group’s day-to-day activities. The group’s mood may be so taken-for-granted that members do not realize its influence, but George believes that a positive group affect will lead to increases in a number of pro-group actions, including helping out other members, protecting the group, making constructive suggestions, and even enhancing survival (Barsade & Gibson, 2012). In some cases, the members of groups may experience emotions in response to events and outcomes that they personally did not experience, as when all members of a group are happy when one of the members receives an award or all become angry when they learn one of their own has been mistreated (Smith, Seger, & Mackie, 2007). Group-level emotions can even emerge when people are isolated from their groups, but nonetheless identify with them. For example:

Imagine all of the new college students who wander across campuses and settle into dorm rooms at the beginning of their freshman semester. Despite geographic divides, these students may nevertheless feel very similar emotions when thinking about themselves as college freshmen. Or consider the business traveler, holed up alone in a foreign hotel room. Despite the absence of even a single compatriot, she may nevertheless feel the same surge of pride as she reads about her country’s exploits at the summer Olympics that her fellow countrymen thousands of miles away do (Moons et al., 2009, p. 760).

Affect and Relational Cohesion

A number of theorists believe that the positive emotions that generate cohesion arise spontaneously during the course of routine interactions in groups—so long as these interactions are relatively pleasant. For example, sociologists Edward Lawler, Shane Thye, and Jeongkoo Yoon (2014), in their relational cohesion theory, argue that group members, because they are linked to one another in recurring (and mostly positive) exchange relationships, eventually experience positive emotions when interacting with one another. Particularly when “jointness” is high—members must align their behaviors with each other in order to reach their goals—then members will attribute their positive feelings to the group and become more strongly committed to it.

Lawler, Thye, and Yoon (2000) tested the theory by arranging for groups of three students seated in separate cubicles to work together on an economic decision-making task. Participants were told on each round they would have the opportunity to draw out points from a common pool and that at the end of 20 rounds the points they earned would be converted into monetary payment. The group members needed to negotiate the amounts each withdrew among them; if too much was taken, no one would receive any points at all. After numerous rounds of these negotiations the members were asked to describe their emotions using such adjectives as pleased, happy, satisfied, and contented. They also indicated if they felt their group was cohesive (e.g., close, solid, and coming together) or divisive (e.g., distant, fragile, or coming apart). As relational cohesion theory predicts, the more frequently the groups successfully completed their negotiations, the more positive their emotions, and the more positive their emotions, the higher their group’s level of cohesion.

Does Collective Movement Build Cohesion?

Some rituals and practices, such as collective singing, chanting, praying, and marching, result in the development of a shared emotional elevation among group members. Historian William McNeill (1995), in his book Keeping Together in Time, describes this feeling by drawing on his personal experience as a new recruit during basic training in the U.S. Army.

Marching aimlessly about on the drill field, swaggering in conformity with prescribed military postures, conscious only of keeping in step so as to make the next move correctly and in time somehow felt good. Words are inadequate to describe the emotion aroused by the prolonged movement in unison that drilling involved. A sense of pervasive well-being is what I recalled; more specifically, a strange sense of personal enlargement; a sort of swelling out, becoming bigger than life, thanks to participation in collective ritual (p. 2).

McNeill suggests that much of the history of modern forms of warfare can be traced to the cohesion-building effects of close-group training. His collective-movement hypothesis offers, for example, a solution to one of military history’s great mysteries: How did the Greek forces of Athens and Sparta, in the period from 600 B.C. to 300 B.C., manage to overwhelm vastly superior forces? McNeill’s proposal: The Greeks relied on highly cohesive groups of ground forces that moved forward as a synchronized unit. This formation is known as a phalanx, from the Greek word for fingers. These units varied in size, but were typically at least eight rows deep and stretched wide enough across a field of battle to prevent flanking. In some cases, each man’s shield was designed so that it covered the soldier beside him as well, thereby further increasing the unity of the group. The men of these phalanxes trained together over long periods of time, and they became synchronized to the point that they acted as a single unit that could inflict great damage against even the best-trained individual soldiers. These phalanxes eventually gave way to other means of organizing men in battle, given their vulnerability to cavalry and more maneuverable adversaries.

5-1e. Structural Cohesion

Structural cohesion is unity of a group that results from the integrity of its structural features, including norms, roles, and intermember relations. The U.S. Olympic team, for example, was a well-structured group, in terms of roles, norms, and relationships. Each player had a position on the ice that he played, and his responsibilities in that role were well defined. The players, through practice, knew what they were supposed to do when on offense and defense, and their success in enacting these duties was reviewed regularly by the coaching staff that had authority over all the players. The group also had clear norms about how it operated, what kinds of behaviors were acceptable, and the goals that it sought. The group was also closed, rather than open: only certain individuals could join, and membership was regarded by most as a noteworthy accomplishment (Ziller, 1965). These social structures regulated members’ behavior, minimized conflict within the group, routinized communication and interdependencies, and, in doing so, increased the group’s cohesiveness (Eys & Carron, 2001; Moody & White, 2003).

Just as a well-designed building can withstand the vicissitudes of time and weather, so can a structurally cohesive group withstand stresses and strains that would cause a less coherent group to crumble. For example, the two groups studied by the Sherifs (1953, 1956) discussed earlier in the chapter—the Bulldogs and Red Devils—developed very different organizational structures. As Figure 5.1 indicates, the Bulldogs had a dense network of relationships linking members, whereas the Red Devils team was more stratified. When the boys were asked to name as many as five friends at the camp, 9 of the 12 Bulldogs named each other in a tightly knit pattern of reciprocal and overlapping choices. The remaining three individuals received no friendship nominations, but they picked others who were part of the main cluster as friends and were not rejected. In the Red Devils, liking was more concentrated: the two most-liked individuals in the group garnered 50% of all friendship choices. The Red Devils’ group structure also included a large subgroup—50% of the members were nested in a clique linked to one of the mid-level leaders. Such subgroups create fault lines in groups, and, when the group experiences turmoil, it can break apart along these lines (Lau & Murnighan, 1998). These differences in structure corresponded to differences between the two groups both in cohesiveness and in performance. The Bulldogs were a more tight-knit, cohesive group, and they were the victors when the two teams played each other in a series of competitions. “The results of the intergroup competition for the Bulldogs were elation and heightened ingroup pride and identification” (Sherif & Sherif, 1956, p. 294).

5-1f. Assumptions and Assessments

The concept of cohesion, given its theoretical importance, has been defined and redefined by dozens of different theorists. Some consider cohesion to be strong feelings of attraction that link members together. Others, in contrast, focus on morale and trust, and still others stress the cohesive group’s capacity to combine members in a highly productive unit. Cohesiveness’s many definitions have caused some to complain that the concept, ironically, lacks cohesion (e.g., Casey-Campbell & Martens, 2009).

The Multicomponent Assumption

A multicomponent approach embraces this definitional diversity by suggesting that many different factors contribute to the unity of a group. The members of the U.S. Olympic hockey team, for example, became good friends as well as teammates, they worked well together, they identified with their team, they played with great emotional intensity, and the group was highly structured. But, as is consistent with the concept of equifinality, other cohesive groups may not exhibit all of these qualities.

Some kinds of cohesion are more common than others. For example, studies of performance-focused groups, such as military squads and sports teams, identify two components—social cohesion and task cohesion—as the two primary forms of cohesion and distinguish between these two forms and secondary forms of cohesion (Siebold, 2007, 2015). Sociological and network studies, in contrast, tend to emphasize the importance of structural features as critical determinants of cohesion (e.g., Burke, Davies, & Carron, 2014; McPherson & Smith-Lovin, 2002).

The Multilevel Assumption

Cohesion is also a multilevel process as well as a multicomponent one. Social cohesion includes both liking for specific members but also liking for the group itself. Task cohesion is commitment to one’s personal goals, but also the goals that the group is pursuing. But each of the other sources of cohesion considered in this section—identity, emotions, and structure—also operate at multiple levels. Members of cohesive groups not only identify with other members, mimic their emotions, and meet their role obligations: They embrace the group’s identity, share its emotions, and fit into its structure. A multilevel analysis must also take into account vertical and organizational bonding (Siebold, 2007). Cohesion is substantially influenced by the strength of the relations between members and their leaders (vertical bonding) and the relationship between the group and the organization or institution in which it is embedded (organizational bonding).

The Multimethod Assumption

Since cohesion is multifaceted, researchers use a wide variety of methods to measure it. Some make use of social network methods, indexing the unity of a group by considering who likes who and the group’s structure. Others rely on observational strategies, monitoring interpersonal relations among members, noting instances of conflict or tension, and members’ physical locations over time. In many cases, too, investigators hope that group members are accurate observers of their group’s cohesiveness and, if asked, will share these perceptions. They might ask members of a group to only describe their own attraction to and commitment to the group through such questions as, “Are you attracted to the group?” or “Do you feel a strong sense of belonging to the group?” Other researchers, in contrast, may ask group members to estimate the group’s cohesion directly through such questions as, “Are members attracted to this group?” and “Is this group a cohesive one?” (Salas et al., 2015).

This plethora of operational definitions can create challenges for researchers. When they measure cohesiveness in different ways, they often report different conclusions. A study using a self-report measure of cohesion might find that cohesive groups produce more than cohesive ones, but other investigators may not replicate this finding when they use observational measures. Moreover, some operational definitions of cohesion may correspond more closely to the theoretical definition than others. A measure that focuses only on group members’ perceptions of their group’s cohesiveness, for example, may be assessing something very different than measures of the actual strength of the relationships linking individuals to their group.

5-2. Developing Cohesion

The U.S. Olympic hockey team that faced the Russian team in February of 1980 was extraordinarily unified, but the group did not become cohesive all at once. When Coach Brooks invited the best amateur hockey players to a training camp in Colorado Springs in July 1979, the players showed few signs of camaraderie, fellowship, or cohesion. Many had played against one another in college and still held grudges, and some were so temperamental that no one would befriend them. But the hockey team changed over time, transforming from a collection of talented individuals into a cohesive team.

5-2a. Theories of Group Development

New groups are different from established groups. The committee meeting for the first time will not act the way it will during its tenth meeting. The team playing its first game of the season will not perform in the same way it will on its last game. The partygoers at 2 AM don’t do the sorts of things they did at the party’s start at 9 PM. Some of the changes that a group and its members undergo are specific to that particular group, for they are the result of the unique characteristics of the members, the distinctive way these individuals interact with each other, and the group’s reaction to external pressures that it may encounter. But along with these idiographic changes are more predictable patterns of change that are common to most groups the longer their duration.

Theories of group development seek to describe these recurring patterns of change in a group’s structure and interactions that occur over the course of the group’s existence. Some theories—successive-stage models—suggest that groups move through a series of separable stages as they develop. The U.S. Olympic team, for example, became unified, but only after progressing through earlier stages marked by confusion, conflict, and growing group structure. Cyclical models, in contrast, argue that groups repeatedly cycle through periods or phases during their lifetimes, rather than just moving through each stage once. The U.S. Olympic team, for example, experienced substantial shifts in its levels of conflict between the coach and the players throughout its existence, but these shifts triggered processes that worked to control tension and increase harmony. Still other theories mix elements of both stage and cycle models and extend these two basic perspectives in various ways. We consider examples of these approaches to the analysis of group development in this section (Arrow et al., 2005; Wheelan, 2005).

5-2b. Five Stages of Development

Just as humans mature from infancy to childhood, adolescence, adulthood, and old age, stage models of group development theorize that groups move from one stage to the next in a predictable, sequential fashion. It is not the passage of time, per se, which is critical, but rather the processes that take place as the group transforms from an amalgam of individuals into a cohesive unit. As the group deals with the challenges that it must confront at each stage, the group’s interactions stabilize, the relations joining the members strengthen, and the group becomes more proficient. A group that has completed its movement through each stage in the sequence should, in theory, function more effectively than one that has not.

The number and names of the stages vary among theorists, but many models highlight certain interpersonal outcomes that must be achieved in any group that exists for a prolonged period (see Table 5.1). At the outset, the group members must become oriented toward one another. Second, they often find themselves in conflict, and some solution is sought to improve their working relationship. Next, norms and roles develop that regulate behavior, and the group achieves greater unity. In the fourth phase, the group can perform as a unit to achieve desired goals. The final stage ends the sequence of development with the group’s adjournment. The educational psychologist Bruce Tuckman (Tuckman, 1965; Tuckman & Jensen, 1977) gave these stages in the five-stage model of group development poetically pleasant names: forming (orientation), storming (conflict), norming (structure development), performing (work), and adjourning (dissolution).

Forming: The Orientation Stage

The first few minutes, hours, days, or even weeks of a newly formed group’s life are often marked by tension, guarded interchanges, and relatively low levels of interaction. During this initial forming stage, members monitor their behavior to avoid any embarrassing lapses of social poise and are tentative when expressing their personal opinions. Because the group’s structure has not had time to develop, the members are often uncertain about their role in the group, what they should be doing to help the group reach its goals, or even who is leading the group.

With time, tension is dispelled as the ice is broken and group members become better acquainted. After the initial inhibitions subside, group members typically begin exchanging information about themselves and their goals. To better understand and relate to the group, individual members gather information about their leaders’ and members’ personality characteristics, interests, and attitudes. In most cases, too, members recognize that the others in the group are forming an impression of each other, and so they facilitate this process by revealing some private, personal information during conversations and Internet-based exchanges. This gradual, and in some cases tactical, communication of personal information is termed self-disclosure, and it serves the important function of helping members get to know one another (Jourard, 1971). Eventually, the group members feel familiar enough with one another that their interactions become more open and spontaneous.

Storming: The Conflict Stage

As the relatively mild tension caused by the newness of a group wanes, tension over goals, procedures, and authority often waxes. On the U.S. Olympic team, for example, the players from the schools in the eastern part of the United States often excluded the players from the Midwest. Several players were considered hotshots more interested in their personal performance than in team success. And nearly all the players rebelled against the hard-driving coaching style of Herb Brooks. He would yell, insult, swear, and curse the players whenever they failed to perform up to his standards, and he often threatened to cut players from the team.

The storming stage is marked by a “lack of unity” (Tuckman, 1965, p. 386), including personal conflicts between individual members who discover that they just do not get along, procedural conflict over the group’s goals and procedures, and competition between individual members for authority, leadership, and more prestigious roles. In groups that have an official leader, like the U.S. Olympic team, the conflict often centers on relationships between the leader and the rest of the group. In the orientation stage, members accept the leader’s guidance with few questions, but as the group matures, leader–member conflicts disrupt the group’s functioning. Members may oscillate between fight (counter dependency) and flight (withdrawal) as some openly challenge the leader’s authority and others exude submissiveness. In groups that have no formally appointed leader, conflicts erupt as members vie for status and roles within the group. Once stable patterns of authority, attraction, and communication have developed, conflicts subside, but until then, group members jockey for authority and power (Forsyth & Diederich, 2014).

Many group members are discouraged by this outbreak of conflict in their young groups, but conflict is the yang to the yin of group harmony. As Chapter 13’s analysis of the roots of conflict suggests, the dynamic nature of the group ensures continual change, but along with this change come stresses and strains that surface in the form of conflict. In rare instances, group members may avoid all conflict because their actions are perfectly coordinated; but in most groups, the push and pull of interpersonal forces inevitably exerts its influence. Low levels of conflict in a group can be an indication of remarkably positive interpersonal relations, but it is more likely that the group members are simply uninvolved, unmotivated, and bored (Chin et al., 2017).

Conflict is not just unavoidable, however; it may be a key ingredient for creating group cohesion. If conflict escalates out of control, it can destroy a group. But in some cases, conflict settles matters of structure, direction, and performance expectations. Group members must learn to live and work together, and to do so they may need to openly discuss—and even argue over—a specific problem or task they are dealing with (task conflict), the way they work together (process conflict), the stresses and strains of their interpersonal relationships (personal conflict), and who is in charge (status conflict). To move along in their developmental process, the members of cohesive groups must learn to manage their conflict: when hostility surfaces it must be confronted and resolved (Bradley et al., 2015). Successful conflict resolution may even “serve to ‘sew the social system together’” by identifying and resolving disagreements that if left untended could destabilize the group (Coser, 1956, p. 801). Groups that spend too much time in the storming stage, however, may find that the harmful effects of protracted conflict-laden interactions cannot be undone (De Dreu, 2010).

Can Cohesion Make a Bad Job Good?

Cohesiveness increases with interaction, for the more people do things together as a group—talking, working, eating, relaxing, socializing, traveling, and so on—then the more cohesive the group will become. This generalization, however, comes with qualifications, for any number of situational factors can shift interaction out of the plus column into the minus. If the interactions take place in a hostile environment; if a substantial number of group members are interpersonally irritating in some way; if the group interactions are uncoordinated and boring; if many of the members feel that they are unfairly excluded from the group’s activities, the group’s development of cohesiveness will be delayed.

Cohesion, however, can manifest in even the most challenging of group circumstances, as sociologist Donald Roy’s (1959) “banana time” case study reveals. Roy worked, for two months, in 12-hour shifts lasting from 8 AM to 8:30 PM, with three other men in an isolated room in a garment factory operating a press machine. The work was not just tedious, but menial, repetitive, and tiring. Roy felt he could not last more than a week, but that was before he was drawn into the interaction of the small group. The group filled its days with jokes, teasing, kidding around, and horseplay that gave structure and meaning to their work. To break up the day into smaller segments, the men stopped from time to time for various refreshments and breaks. There was, of course, lunch time, but the men added many others, such as coffee time, peach time, fish time, and banana time. These rituals and social activities, collectively called “banana time” by Roy, turned a bad job into a good one.

All cohesive groups have their banana times—interaction rituals that elevate the degree of social connection among the members. Like traditional rituals, such as grace said before meals or singing the national anthem, such interaction rituals provide structure and meaning for the group and its members. Reading the minutes of the last meeting, introducing new members, joking about the member who is always late, or commenting on someone’s appearance, are all simple rituals that ensure that the group’s activities will unfold in a predictable and orderly way. Rituals and ceremonies have been linked to increases in a shared focus among members, emotional energy, and overall cohesiveness (Collins, 2004).

Norming: The Structure Stage

With each crisis overcome, the U.S. Olympic hockey team became more stable, more organized, and more unified. The players revised their initial impressions of each other and reached more benevolent conclusions about their teammates. The players still complained about the team rules, the practice schedules, and the coach’s constant criticisms, but they became fiercely loyal to the team, their teammates, and their coach. Whereas groups in the orientation and conflict stages are characterized by low levels of intimacy, friendship, and continuity, in the norming stage members become more trusting, supportive, and cooperative. The group becomes cohesive.

As the group becomes more organized, it resolves the problems that caused earlier conflicts—uncertainty about goals, roles, and authority—and prepares to get down to the work at hand. Norms emerge more clearly and guide the group members as they interact with one another. The group begins to display interaction rituals that provide structure and meaning for the group and its members, such as regularities in small talk or repetitive mundane practices (Collins, 2004). Differences of opinion still arise, but now they are dealt with through constructive discussion and negotiation. Members communicate openly with one another about personal and group concerns, in part because members know one another better. On the U.S. Olympic team, the players did not always agree with the coach, but they changed the way they dealt with disagreement. Instead of grumbling about their treatment, several players started compiling a book of Brooksisms—the odd expressions Coach Brooks used during practice to motivate his players. Nearly every player, interviewed 20 years after they played for Brooks, remembered such Brooksisms as “You are playing worse every day and now you are playing like the middle of next week” and “Gentlemen, you don’t have enough talent to win on talent alone.”

Performing: The Work Stage

The U.S. Olympic hockey team played 41 games against other teams in preparation for the Olympics and won 30 of those matches. They reached their peak of performance when they beat the Russian team to qualify for the final, gold-medal game against Finland. Before that game, Coach Brooks did not give them a pep talk, as he had before the Russian game. Instead, he only said, “You lose this game and you will take it to your [expletive deleted] graves” (Warner HBO, 2001). They won.

Few groups are productive immediately; instead, productivity must usually wait until the group matures. Various types of groups, such as conferences, factory workers assembling relay units, workshop participants, and the members of expeditions, become more efficient and productive later in their group’s life cycle (Hare, 1967, 1982; Hare & Naveh, 1984). The more “mature” a group, the more likely the group will spend the bulk of its time working toward its chosen goals rather than socializing, seeking direction, or arguing (Dierdorff, Bell, & Belohlav, 2011). When researchers coded the content of group members’ verbal interactions, they discovered task-focused remarks occur later rather than sooner in the group’s life (e.g., Bales & Strodtbeck, 1951; Borgatta & Bales, 1953; Heinicke & Bales, 1953). Conflict and uncertainty also decrease over time as work-focused comments increase. Groups that have been together longer talk more about work-related matters, whereas younger groups are more likely to express conflict or uncertainty and make requests for guidance (Wheelan, Davidson, & Tilin, 2003). Once the group reaches the performing stage “members shift their attention from what the group is to what the group needs to do” (Bushe & Coetzer, 2007, p. 193).

Not all groups, however, reach this productive work stage. If you have never been a member of a group that failed to produce, you are a rare individual indeed. In a study of neighborhood action committees, only 1 of 12 groups reached the productivity stage; all the others were bogged down at the forming or storming stages (Zurcher, 1969). An early investigation of combat units found that out of 63 squads, only 13 could be clearly classified as effective performance units (Goodacre, 1953). An analysis of 18 personal growth groups concluded that only 5 managed to reach the task performance stage (Kuypers, Davies, & Glaser, 1986). These studies and others suggest that time is needed to develop a working relationship, but time alone is no guarantee that the group will be productive (Gabarro, 1987). Chapters 10, 11, and 12 examine issues pertaining to team performance in detail.

Adjourning: The Dissolution Stage

After its astounding victory, the U.S. Olympic hockey team was celebrated by a nation of sudden hockey fans, including the president of the United States. But the ceremony at the White House marked the end of the group’s existence, for the team never reconvened or played again. After meeting the president, the teammates clapped one another on the back one last time, and then the group disbanded.

A group’s entry into the dissolution stage can be either planned or spontaneous. Planned dissolution takes place when the group accomplishes its goals or exhausts its time and resources. The U.S. Olympic hockey team meeting the president, a wilderness expedition at the end of its journey, a jury delivering its verdict, and an ad hoc committee filing its final report are all ending as scheduled. Spontaneous dissolution, in contrast, occurs when the group’s end is not scheduled. In some cases, an unanticipated problem may arise that makes continued group interaction impossible. When groups fail repeatedly to achieve their goals, their members or some outside power may decide that maintaining the group is a waste of time and resources. In other cases, the group members may no longer find the group and its goals sufficiently satisfying to warrant their continued membership. As social exchange theory maintains, when the number of rewards provided by group membership decreases and the costly aspects of membership escalate, group members become dissatisfied. If the members feel that they have no alternatives or that they have put too much into the group to abandon it, they may remain in the group even though they are dissatisfied. If, however, group members feel that other groups are available or that nonparticipation is preferable to participation in such a costly group, they will be more likely to let their current group die (Vandenberghe & Bentein, 2009).

The dissolution stage can be stressful for members (Birnbaum & Cicchetti, 2005). When dissolution is unplanned, the final group sessions may be filled with conflict-laden exchanges among members, growing apathy and animosity, repeated failures at the group’s task, and loss of trust within the group. Even when dissolution is planned, the members may feel distressed. Their work in the group may be over, but they still mourn for the group and suffer from a lack of personal support. In many cases, too, group members deal with the demise of the group for blaming each other for the group’s errors and misadventures (Gillespie & Dietz, 2009).

5-2c. Cycles of Development

Tuckman’s model, which is operationalized in Figure 5.2, is a successive-stage theory: It specifies the usual order of the phases of group development. Sometimes, however, group development takes a different course. Although interpersonal exploration is often a prerequisite for group solidarity, and cohesion and conflict often precede effective performance, this pattern is not universal. Some groups manage to avoid particular stages; others move through the stages in a unique order; still others seem to develop in ways that cannot be described by Tuckman’s five stages (Bonebright, 2010). Also, the demarcation between stages is not clear-cut. When group conflict is waning, for example, feelings of cohesion may be increasing, but these time-dependent changes do not occur in a discontinuous, stepwise sequence (Arrow, 1997).

Charting the development of groups. The five-stage model of group development assumes that groups naturally progress through five stages as their processes, procedures, and personal relationship change over time. To apply the model, think of a group to which you currently belong. Then, beginning at the topmost box, ask if your group displays the qualities of a “young” group, or if it has moved on to a later stage of development. If the group you are analyzing is one that is responsible for completing a task, the sooner it reaches the performing stage, then the more likely it will be successful in reaching its goals.

Many theorists believe that groups repeatedly cycle through stages during their lifetime, rather than just moving through each stage once. These cyclical models agree that certain issues tend to dominate group interaction during the various phases of a group’s development, but they add that these issues can recur later in the life of the group. Very long-term groups, such as teams of software engineers who work on products for many years, show signs of shifting from task-focused stages back to conflict (re-storming) and norming (re-norming) stages (McGrew, Bilotta, & Deeney, 1999). Production crews in progress meetings spend much of their time discussing the work itself—the tasks they have completed and those still undone—but their conversations also include relational, group-focused topics that increase cohesiveness. Notably, groups that balance the task side with the relational side are more likely to complete their projects on time than are groups whose meetings are all business and no relationships (Gorse & Emmitt, 2009).

The sociologist Robert Freed Bales’ equilibrium model of group development assumes that group members strive to maintain a balance between accomplishing the task and enhancing the quality of the interpersonal relationships within the group. In consequence, groups cycle back and forth between what Tuckman called the norming and performing stages: A period of prolonged group effort must be followed by a period of cohesion-creating, interpersonal activity. The discussion groups that Bales studied followed this general pattern of oscillation between the two types of group activity (Bales & Cohen, 1979).

Punctuated equilibrium models agree with Bales’ view, but they add that groups often go through periods of relatively rapid change. These changes in the group’s developmental trajectory can be triggered by a barometric event: an incident or outcome that causes a significant shift in the interpersonal dynamics of the group (Bennis & Shepard, 1956). The halfway point in the group’s life, too, can trigger dramatic changes in the group, as members realize that the time they have available to them is dwindling (Gersick, 1989). Groups must deal with deadlines and time pressures, and as time runs out conflict and tension can rise, whereas the group’s cohesiveness can drop (Mohammed, Hamilton, & Lim, 2009).

The U.S. Olympic team’s development, although stage-like in many respects, changed more rapidly following critical events. Perhaps the most dramatic turning point in the group’s life occurred when the team lost an exhibition game to a weak team. Coach Brooks believed the team played without any heart or energy, and after the game he kept them on the ice rather than letting them shower and change. He made the players skate back and forth between the goals (the players called these drills “Herbies”) for what seemed like hours. Even when the arena manager turned off the lights and went home, Brooks kept the team skating back and forth in the dark. The experience created a feeling of unity in the group, and this cohesiveness carried them through the remainder of their games and on to victory. Such turning points may, however, be relatively rare in groups. More typically, the shift from an initial orientation focus to a task focus occurs gradually as groups and their members pace their progress toward the completion of their final goal (Maznevski & Chudoba, 2000).

How Do Online Groups Build Cohesion?

Many online groups are as collaborative, cohesive, and continuous as their corresponding offline cousins. For example, online project-focused groups and teams—such as groups created in college courses that complete group-level learning activities and groups in organizational contexts that have explicit work-related purposes—are initially less unified than offline groups, but most eventually catch up to groups that interact in face-to-face settings (e.g., Bateman, Gray, & Butler, 2011).

Some types of online groups, however, are less successful when it comes to reaching high levels of cohesiveness. Groups that form in social networking sites, for example, often exhibit one telling sign of low cohesion: membership instability. In a social network site, such as Facebook, people can leave or join a group or sever a connection (“unfriend” someone) at the push of a button. Similarly, studies of the guilds that form in the multiplayer game World of Warcraft indicate that members join and leave these groups at a relatively high rate (Ducheneaut &Yee, 2014).

Robert Ziller’s (1965) theory of open and closed groups offers an explanation. Ziller suggests that groups with permeable boundaries—open groups—are especially unlikely to become cohesive, since members do not remain in the group long enough to develop attraction bonds, commitment to group goals, or a sense of solidarity. Members of closed groups, in contrast, are more willing to invest in their groups, because they consider their membership to be more enduring than temporary.

Many online groups, recognizing the benefits of a stabilized membership and the cohesiveness that stability brings, use various technological and interpersonal strategies to transform their open groups into closed ones. Those who create e-groups often set up the membership procedures so that the group has clear boundaries between those who belong and those who do not. In some cases, those seeking to join must be approved by the group before they are admitted, and only then will they be able to access the contents of the group’s communications (Backstrom et al., 2008). Many of the most successful, long-term guilds in World of Warcraft, for example, are highly selective, accepting only applicants with certain credentials or those who have been referred to the guild by a current member. Guilds also increase the commitment of members to the group by requiring each member to contribute some of their resources to a common pool—a “guild bank”—and those resources are forfeited if one leaves the guild (Malone, 2009). Although research is needed to evaluate the impact of these methods of increasing the unity of online groups, anecdotal evidence suggests they are effective means of increasing the cohesiveness of e-groups. A Google search of the phrase “I love my guild” yielded about 47,000 results, whereas a search for “I hate my guild” had only 4,500 hits.

5-3. Consequences of Cohesion

Cohesion is something of a purr word. Most people, if asked to choose between two groups—one that is cohesive and another that is not—would likely pick the cohesive group. But cohesiveness has its drawbacks. A cohesive group is an intense group, and this intensity affects the members, the group’s dynamics, and the group’s performance in both positive and negative ways. Cohesion leads to a range of consequences—not all of them desirable.

5-3a. Member Satisfaction and Adjustment

The men who were part of the U.S. Olympic hockey team had no regrets. The work load was grueling, the pressure to improve was relentless, the flow of criticism from their coach never ceased, but nonetheless: each one considered their six months together in 1980 to be a special time in their lives. As one of the team’s coaches, Craig Patrick, explained: “in the end it was the camaraderie of the unit that developed; it was something special” (Warner HBO, 2001).

Across a range of groups in industrial, athletic, and educational settings, people who are members of highly compatible, cohesive groups report more satisfaction and enjoyment than members of noncohesive groups. One investigator studied teams of masons and carpenters working on a housing development. For the first five months, the men worked at various assignments in groups formed by the supervisor. This period gave the men a chance to get to know virtually everyone working on the project and natural likes and dislikes soon surfaced. The researcher then established cohesive groups by making certain that the teams only contained people who liked each other. As anticipated, the masons and carpenters were much more satisfied when they worked in cohesive groups. As one of them explained, “Seems as though everything flows a lot smoother.… The work is more interesting when you’ve got a buddy working with you. You certainly like it a lot better anyway” (Van Zelst, 1952, p. 183).

A cohesive group creates a healthier workplace, at least at the psychological level. Because people in cohesive groups respond to one another in a more positive fashion than the members of noncohesive groups, people experience less anxiety and tension in such groups (Seashore, 1954). People who belong to cohesive groups—with cohesion defined as a strong sense of belonging to an integrated community—are more actively involved in their groups, are more enthusiastic about their groups, and even suffer from fewer social and interpersonal problems (Hoyle & Crawford, 1994). They are also happier people. Who responds the most positively when asked such questions as “How happy would you say you are?” and “Would you say you are satisfied with your life right now?” People who live and work in cohesive groups, communities, and societies (Delhey & Dragolov, 2016).

Cohesive groups can, however, be emotionally demanding (Forsyth & Elliott, 1999). The old sergeant syndrome, for example, is more common in cohesive military squads. Although the cohesiveness of the unit initially provides psychological support for the individual, the loss of comrades during battle causes severe distress. When the unit is reinforced with replacements, the original group members are reluctant to establish emotional ties with the newcomers, partly in fear of the pain produced by separation. Hence, they begin restricting their interactions, and these “old sergeants” can eventually become completely isolated within the group (Sobel, 1947). Some highly cohesive groups may also sequester members from other groups in an attempt to seal members off from competing interests. Individuals who leave highly cohesive groups, such as small religious communities, combat units, and gangs, may experience loneliness, chronic guilt, and isolation when their membership in such groups ends. As groups become highly cohesive, they can become so self-contained that members’ links to nonmembers are severed, virtually isolating the group and its members (Junger, 2015).

Cohesive groups can also make those few members who are not closely bonded with the group feel like outsiders. Individuals who dropped out of cohesive groups often recognized that the group was cohesive, but they did not feel that they were part of that close-knit unit (Robinson & Carron, 1982). Being part of a cohesive group is not that enjoyable for those on the fringe.

5-3b. Group Dynamics and Influence

As cohesion increases, the internal dynamics of the group intensify, so pressures to conform are greater in cohesive groups, and individuals’ resistance to these pressures is weaker. Anecdotal accounts of highly cohesive groups—military squads, adolescent peer groups, sports teams, gangs, fraternities and sororities, and cults—often describe the strong pressures that these groups put on their members (Goldhammer, 1996). Cohesive groups—as the final section of this chapter concludes—are more likely to initiate their members. Drug use and illegal activities are often traced back to conformity pressures of adolescents’ peer groups (Giordano, 2003). Cohesive gangs exert strong pressure on members (Peterson & Panfil, 2014). Alternative religious groups (“cults”) may demand extreme sacrifices from members, including suicide. Sports teams, if highly cohesive, may extract both compliance and sacrifice from members (Prapavessis & Carron, 1997). Highly cohesive decision-making groups may err in their judgments when members feel pressured to keep their concerns and disagreements to themselves (Janis, 1972).

Cohesion can also increase negative group processes, including hostility, scapegoating, and hostility toward other groups (Pepitone & Reichling, 1955). In one study, cohesive and noncohesive groups worked on a series of unsolvable problems. Although all the groups seemed frustrated, coalitions tended to form in noncohesive groups, whereas cohesive groups vented their frustrations through interpersonal aggression: overt hostility, joking hostility, scapegoating, and domination of subordinate members. The level of hostility became so intense in one group that observers lost track of how many offensive remarks were made; they estimated that the number surpassed 600 comments during the 45-minute work period (French, 1941).

5-3c. Group Productivity

The cohesive, unified group has, throughout history, been lauded as the most productive, the most likely to win in battle, and the most creative. The Spartans who held the pass at Thermopylae were a model of unity, courage, and strength. The crew of the ship Endurance, which was crushed by ice floes during a voyage to the Antarctic, survived by working together under the able leadership of Ernest Shackleton. The interpersonally enmeshed engineers at the Palo Alto Research Center (PARC) invented the personal computer and other assorted technologies, including the mouse, a graphical interface (clickable icons), e-mail, and laser printers. When the U.S. Olympic hockey team won, most sports commentators explained the victory by pointing to the U.S. team’s cohesiveness, even suggesting that a unified team could work “miracles.” But is this folk wisdom consistent with the scientific evidence? Are cohesive groups really more productive?

The Cohesion → Performance Relationship

Both correlational and experimental studies of all kinds of groups—sports teams, work groups in business settings, expeditions, military squads, and laboratory groups—generally confirm the cohesion → performance relationship: Cohesive groups tend to outperform less unified groups. One analysis of 49 studies of 8,702 members of a variety of groups reported that 92% of these studies supported cohesive groups over noncohesive ones (Mullen & Copper, 1994). Another review, this time aggregating the results of 19 studies of 778 groups, found that increases in cohesion were reliability associated with increases in performance (r = .30; Beal et al., 2003). And when researchers directly manipulate the cohesiveness of the groups they are studying experimentally, they routinely find that groups assigned to the high-cohesion conditions outperform those assigned to the low-cohesion conditions (e.g., Back, 1951; Zaccaro & Lowe, 1986).

What is it about a cohesive group that makes it more successful? Does the high level of attraction among members reduce conflict, making it easier for the group to concentrate on its work? Or perhaps group members are more dedicated to their group if it is cohesive, and this sense of dedication and group pride prompts them to expend more effort on behalf of their group.

The success of cohesive groups lies, in part, in the enhanced coordination of their members. In noncohesive groups, members’ activities are uncoordinated and disjointed, but in cohesive groups, each member’s contributions mesh with those of the other group members. Cohesion thus acts as a “lubricant” that “minimizes the friction due to the human ‘grit’ in the system” (Mullen & Copper, 1994, p. 213). Members of cohesive groups all share the same “mental model” of the group’s task and its demands, and this shared prescription for how the task is to be accomplished facilitates their performance. Hence, cohesive groups are particularly likely to outperform noncohesive groups when the group’s task requires high levels of interaction and interdependence. The degree of interdependency required by the type of tasks the group is working on also determines the size of the cohesion–performance relationship; the more group members must coordinate their activities with one another, the more likely a cohesive group will outperform a less cohesive one (Beal et al., 2003; Gully et al., 1995).

These meta-analytic studies also show support for the value of a multicomponent conceptualization of cohesion, for they suggest that even when cohesion is operationalized in different ways, the cohesion–performance relationship still holds true. In their analysis, social psychologists Brian Mullen and Carolyn Copper (1994) gave the edge to task cohesion, particularly in studies involving bona fide groups rather than artificial ones. Subsequent analyses, however, found evidence that all three components—social, task, and perceptual (“group pride”) cohesion—were related to performance when one looked only at group-level studies (Beal et al., 2003).

The Performance → Cohesion Relationship

When the U.S. Olympic hockey team received their gold medals in the 1980 games, the team likely reached the apex of its cohesion. They not only liked one another, shared positive emotions, and were structurally organized, but they were also victorious: they were a successful group, whereas the Russian team had failed—and this achievement likely further enhanced the group’s overall level of cohesiveness.

The cohesion → performance relationship is actually a reciprocal one: When a group performs well at its identified task, the level of cohesion in the group increases, but should it fail, disharmony, disappointment, and a loss of esprit de corps are typically observed. These effects of performance on cohesion occur even when groups are identical in all respects except one—when some are arbitrarily told they performed well, but others are told they did not do well. Even under these highly controlled circumstances, groups given positive feedback became more cohesive than groups that are told they performed poorly. These studies suggest that cohesion is related to performance, not because cohesion causes groups to perform better, but because groups that perform better become more cohesive (e.g., Forsyth, Zyzniewski, & Giammanco, 2002).

Mullen and Copper (1994) examined the flow of causality in the cohesion–performance relationship by comparing experimental studies that manipulated cohesion with studies that used correlational designs. Because the cohesion–performance relationship emerged in both types of studies, they concluded that cohesion causes improved performance. However, the relationship between cohesion and performance is stronger in correlational studies. This disparity suggests that cohesion aids performance, but that performance also causes changes in cohesiveness. Mullen and Copper closely examined seven correlational studies that measured cohesion and performance twice rather than once. These studies suggested that a group’s cohesiveness at Time 1 predicted its performance at Time 1 and at Time 2. But in these studies, group performance at Time 1 was a particularly powerful predictor of cohesiveness at Time 2. These findings prompted Mullen and Copper to conclude that the cohesion–performance relationship is a reciprocal one: Cohesion makes groups more successful, but groups that succeed also become more cohesive (see Figure 5.3). Other researchers have confirmed the bidirectional nature of the cohesion–performance relation, but further suggest that the impact of cohesion on performance becomes stronger over time, whereas the impact of performance on cohesion remains relatively stable (Mathieu et al., 2015).

The hypothetical relationship between five sources of cohesion (social, task, collective, emotional, and structural), cohesion, and performance. Meta-analyses suggest that the cohesion–performance relationship is a reciprocal relationship: cohesion causes improvements in performance, but performance also feeds back and causes changes in cohesion. The magnitude of the cohesion → performance and performance → cohesion relationships varies over time and settings.

Exceptions and Conditions

Increasing a group’s cohesion does not guarantee that it will perform more effectively. The cohesion → performance relationship is stronger

(1)

in bona fide groups than in ad hoc laboratory groups,

(2)

in correlational studies than in experimental studies,

(3)

in smaller groups than in larger groups,

(4)

in sports teams rather than in other types of groups, and

(5)

in project-focused teams rather than in production or service teams (Carron et al., 2002; Chiocchio & Essiembre, 2009; Gully, Devine, & Whitney, 1995; Mathieu et al., 2015; Mullen & Copper, 1994; Oliver et al., 1999).

The relationship also takes time to reach its full strength. The performance-enhancing effects of cohesion may not yield gains when groups are only beginning their work, but will instead emerge over time (Mathieu et al., 2015).

In some cases, too, cohesion can actually undermine performance. In a field study of this process, researchers surveyed 5,871 factory workers who worked in 228 groups. They discovered that the more cohesive groups were not necessarily more productive, but their productivity level from one member to the next was less variable. The individuals working in cohesive groups produced nearly equivalent amounts, but individuals in noncohesive groups varied considerably from one member to the next in their productivity. Furthermore, fairly low standards of performance had developed in some of the highly cohesive groups; thus, productivity was uniformly low in these groups. In contrast, in cohesive groups with relatively high-performance goals, members were extremely productive (Langfred, 1998; Seashore, 1954). As Figure 5.4 indicates, so long as group norms encourage high productivity, cohesiveness and productivity are positively related: a highly cohesive group produces more than a less cohesive one. If group norms encourage low productivity, however, the relationship is negative.

The relationship between cohesion and productivity when norms stress high and low productivity. If the group’s norms encourage productivity, cohesiveness and productivity will be positively correlated. If the group standards for performance are low, however, cohesiveness will actually undermine productivity.

This tendency for the group’s norms about productivity to moderate the strength of the cohesion–performance relationship was also confirmed experimentally by manipulating both cohesion and production norms (Berkowitz, 1954; Gammage, Carron, & Estabrooks, 2001). In one illustrative study, cohesive and noncohesive groups worked on a simple assembly-line-type task. Then, during the task, messages were ostensibly sent from one worker to another to establish performance norms. In some instances, the messages called for increased production (positive messages), but in other instances, the messages requested a slowdown (negative messages). As expected, the impact of the messages was significantly greater in the cohesive groups than in the noncohesive groups. Furthermore, the decreases in productivity brought about by the negative messages were greater than the increases brought about by the positive messages (Schachter et al., 1951).

The take-home lesson from these studies—that creating social cohesiveness may make members happy but not productive—does not apply to the U.S. team. Every one of the team members was committed to the goal of winning the Olympics, so there was no worry that the performance norm would be set too low. In addition, because of the intervention of a thoughtful coach who skillfully built the group’s unity, their cohesion developed over time until its peak during the Olympics. The team’s triumph was called a miracle by some, but in retrospect, it was due to effective group dynamics.

5-4. Application: Explaining Initiations

Cohesiveness is no cure-all for what ails the ineffective group, but many groups in a variety of contexts—sports teams, military units, educational fraternities, sororities, clubs, work squads, and so on—nonetheless often take steps to deliberately increase their cohesion in the hopes their performance will improve. As Chapter 12’s analysis of teambuilding explains, some of these interventions are effective—when groups identify shared goals, improve coordination, and identify sources of conflict, performance often improves. This section, however, concludes the analysis of cohesion by examining a common, but more controversial, means of unifying a group: initiations.

5-4a. Cohesion and Initiations

Many groups require individuals to demonstrate their commitment to the group before they are allowed to become full-fledged group members. Elite military squads, for example, require a new member to pass extremely demanding tests of physical ability. Religious organizations typically require members to study the group’s beliefs and practices and then pass tests on the material before they gain entrance. Other groups require the members to invest considerable time, energy, and personal resources in the group before they can join.

These investments may strengthen the bond between the individual and the group. Groups with admission standards and policies may be more attractive to members, since their exclusiveness may make them seem more prestigious. Since membership must be earned, people who join do so more intentionally and therefore will more likely be active, contributing members. Groups with less stringent requirements are hampered by the unevenness of the contributions of their members: Some may contribute a great deal to the group, but others may actually draw out more resources than they contribute. Groups with strict membership policies, including initiations, avoid this problem by screening and monitoring members closely and dismissing those individuals who do not demonstrate their worth (Iannaccone, 1994).

Initiations and Commitment

Leon Festinger’s (1957) theory of cognitive dissonance offers an intriguing explanation for the relationship between how much new members invest in the group and their commitment to the group. This theory assumes that people prefer to maintain consistency in their thoughts, attitudes, and beliefs. Their belief, for example, that they invested considerable time in a group would be consonant with the belief that this group was of high caliber. But what if, once they joined a group, they discovered it was worthless? Such a situation would generate dissonance and would cause the member psychological discomfort. Although people can reduce cognitive dissonance in many ways, one frequent method is to emphasize the rewarding features of the group while minimizing its costly characteristics. As the group demands more and more in the way of personal commitment, members become more enamored of the group to justify their investment in it.

How Do Groups Respond When Their Prophecies Fail?

Festinger developed cognitive dissonance theory after studying a small “doomsday” group (Festinger et al., 1956). This group formed around a psychic named Marion Keech, who believed she was receiving messages from aliens—the Guardians—who lived on a planet named Clarion. Through these messages, the Guardians warned Ms. Keech of the impending destruction of the world by flood, but they assured her that the small group of men and women who met regularly to discuss their messages would be rescued by a flying saucer before the December 21 deadline.

Festinger and his colleagues joined this group and recorded members’ growing commitment to the group and to Ms. Keech as the deadline loomed. On that date, the faithful gathered in readiness in Ms. Keech’s living room, with their bags packed, all metal zippers, buttons, snaps removed from their clothing (metal does strange things in flying saucers), and passwords all memorized. Unfortunately, midnight came and went with no sign of a flying saucer. The bad news was that Mrs. Keech was wrong when she predicted they would be rescued; the good news was that she was also wrong about the flood.

So how did her followers respond? Did they denounce Mrs. Keech as a fraud? Did they try to contact the Guardians themselves? Did they sew all their zippers back into their clothes? No, they became more committed to Mrs. Keech’s teachings. They accepted her rationalization that the tremendous faith and devotion of the group was so impressive that God had decided to call off the flood, and they spread this important message to the news media. Although the central beliefs of the group had been disconfirmed, they refused to abandon them. If they admitted that Ms. Keech was mistaken, they would have no justification for their actions over the previous months; many had quit their jobs, dropped out of school, moved hundreds of miles to be closer to Ms. Keech, and alienated their friends and relatives. If, however, they became more firmly committed to Ms. Keech, then their cognitions and behaviors would all be consistent with one another, and they could avoid the dissonance that arises from inconsistency (cf. Batson, 1975; Hardyck & Braden, 1962). Festinger concluded that the need for cognitive consistency is a powerful human motive.

Severe Initiations and Group Attraction

Social psychologists Elliot Aronson and Judson Mills (1959) tested dissonance theory in their classic experimental study of young women’s reactions to initiations. Their experimenter, who was a young man, greeted each subject individually and told her that she had been chosen to take part in a group discussion on the psychology of sex. However, because only women who could discuss sex without embarrassment would be welcome in the group, all prospective members had to pass an “embarrassment test.” At this point, the experimenter randomly assigned women to one of the three conditions. Those in severe initiation condition were required to read descriptions of sexual interludes and a series of obscene words aloud to the male experimenter. (In 1959, women were presumably embarrassed by such vulgarities.) Women in the mild initiation condition read some mildly provocative passages and words, and women in the control condition only agreed to be screened.

After the initiation, the researchers told the women that the group they would be joining was already meeting but that they could listen in using an intercom system. But instead of listening to an actual group, the researchers played a recording of a discussion that was contrived deliberately to be exceedingly boring and dull. After listening for a time, the participants rated the group they had listened to on a number of dimensions. Although a rational person should judge a group that was double costly—causing them both embarrassment and boredom—very negatively, the results instead confirmed cognitive dissonance theory: Women who experienced the severe initiation rated the group more positively than those who had experienced a mild initiation or no initiation at all.

Several replications of this unusual study have confirmed the basic finding that initiations influence attraction to and dependency on the group. In one study, researchers used electric shocks rather than obscene readings to manipulate the severity of the initiation into the group and found very similar results: People who received stronger shocks liked the group the best (Gerard & Mathewson, 1966). In another set of studies, participants completed a series of embarrassing, socially awkward activities (such as acting out demeaning situations or performing silly behaviors) or neutral activities before joining a group. In these studies, individuals who suffered through the more severe initiation were more likely to conform to the group’s decisions, rated the group more positively, and felt more comfortable when part of the group (Keating et al., 2005).

Aronson and Mills concluded that the initiation increased cohesion by creating cognitive dissonance, but other factors may also account for the initiation–cohesion relationship. Rather than attempt to reduce cognitive dissonance, their public expressions of liking for such groups may also stem more from a desire to save face after making a faulty decision than from the psychic discomfort of cognitive dissonance (Schlenker, 1975). Initiations also fail to heighten attraction if they frustrate new members or make them angry (Lodewijkx & Syroit, 1997).

5-4b. Hazing

Some groups do not just require members to meet certain criteria and pass tests of stamina or intellectual fitness before allowing entrance. They instead subject new members to cost-creating experiences that have nothing to do with the actual qualities needed to be a successful group member. Initiates in biker gangs, for example, must earn the right to wear the letters and emblems of their gang—their “colors”—by performing a variety of distasteful behaviors (Davis, 1982). Pledges to fraternities at some universities are ritually beaten, subjected to ridicule and embarrassment, and required to drink unhealthy amounts of alcohol (Nuwer, 1999). New members of sports teams are frequently subjected to ritualized physical, psychological, and sexual abuse (Finkel, 2002). These practices qualify as hazing because they expose the new member to significant risk of psychological and physical harm.

Hazing is an entrenched group practice and has been documented in ancient and modern societies and in all parts of the world. Newcomers to groups are routinely subjected to various abuses for reasons both rational and completely irrational (Cimino, 2011).

Bonding and dependence: As Festinger’s dissonance theory suggests, individuals who suffer to join a group value the group more and become more dependent on the group as a source of support and acceptance. Initiation of groups of newcomers, which is typical of certain groups (e.g., sororities, fraternities, and sports teams), increases feelings of unity, for they tend to affiliate more extensively as they deal with the threat and stress. The initiation process thus creates greater cohesion in the overall group, as the individuality of each newcomer is diminished and they learn to rely socially on others (Lodewijkx, van Zomeren, & Syroit, 2005). Hazing of novices, as a group, also increases the unity of the novices, creating a more committed subgroup within the larger group (Mann et al., 2016).

Dominance: Initiations serve to introduce new members to the hierarchical order of the group and the requirements to recognize and respect veteran members. The initiation process humbles the newcomers and signals to them their low status, which they can raise only by contributing in substantial ways to the group. The hazing rituals also provide the current members with the means to exercise their power over the newcomers.

Commitment: Hazing requires a substantial commitment from newcomers and serves to weed out individuals who are not willing to meet the group’s demands. Hazing provides newcomers with the means to prove their worth.

Tradition: Many groups haze new members because they feel that they must honor the group’s traditions, established by founding members of the society (Nuwer, 1999).

Newcomers continue to accept membership in groups that use hazing for many reasons, as well, including a desire to be accepted and to make a good impression with others. In many cases, newcomers are not made aware of the hazing demands and dangers until they are sequestered by the group, and they may fear that refusing to comply with the group’s demands will cause more problems and pose greater risks than compliance. Although victims of hazing appear to have voluntarily taken part in the initiation rites, group influence processes create extremely strong pressures that limit hazing victims’ capacities to act of their own free will. One emergency room physician, who has experience dealing with hazing-related injuries, recommends treating “hazing patients as victims of violent crime, rather than willing participants in their traumatic injuries” (Finkel, 2002, p. 231).

Is Hazing Effective?

Many members of groups defend their right to haze, citing the benefits of initiation for increasing the cohesion of the group. However, research does not offer very much support for this position. One team of investigators asked the members of a number of groups and teams to differentiate between appropriate and inappropriate activities on a list of 24 practices commonly used in initiations and hazing. Appropriate activities included requirements to take part in group activities, swearing an oath, taking part in skits and team functions, doing community service, and maintaining a specific grade point average. Inappropriate activities, in contrast, included kidnapping and abandonment, verbal abuse, physical punishment (spankings, whippings, and beatings), degradation and humiliation (such as eating disgusting things or drinking alcohol in excessive amounts), sleep deprivation, running errands, and exclusion. Somewhat unexpectedly, a number of the behaviors that the researchers felt belonged on the list of inappropriate hazing behaviors, such as wearing inappropriate clothing, head shaving, and sexual activities, were viewed as relatively innocuous by participants (van Raalte et al., 2007).

Did these experiences work to build a cohesive group? Some of these practices were rarely used by groups, but groups that did use inappropriate hazing methods were judged to be less cohesive rather than more cohesive. Hazing, and illicit hazing in particular, backfired, for it did not contribute to increased cohesion, whereas more positive forms of team-building did (van Raalte et al., 2007).

Should Groups Haze?

Hazing is illegal in a number of states, is aggressive in character, yields unhealthy consequences, and does not even work to increase cohesion, yet this practice continues unofficially. Some mild rites and rituals—as when new members must take a public oath of loyalty, memorize the group’s mission, or carry a distinctive object—cause little harm, but in other cases, new members must endure physical and psychological abuse before they are accepted into the group. Emergency room physicians report that they have treated victims of hazing for alcoholic coma, chest trauma, aspiration, 1st, 2nd, and 3rd degree burns, syncope, vomiting, organ damage, heart irregularities, gastrointestinal distress, brain damage, multiorgan system failure, spinal cord injury, exposure, depression, posttraumatic stress, anxiety, and anal, oral, and vaginal trauma (Finkel, 2002). Each year many students are killed or seriously injured in hazing incidents (Goldstein, 2002).

Herb Brooks, the coach of the U.S. Olympic hockey team, did not use hazing, but instead timeworn traditions in sports fitness and training to build the group into a cohesive unit. He was tough on his players, they played dozens of exhibition games leading up to the Olympics, and Brooks kept all of them on edge, threatening to send any one of them home who did not perform up to his standards. Through these experiences, the group reached a very high level of cohesiveness, without recourse to hazing. Given that groups can turn to a variety of equally effective but safe methods to build their group’s cohesiveness, the use of hazing is completely unjustified (Diamond et al., 2016).

Resources

Chapter Case: U.S. Olympic Hockey Team

“A Reminder of What We Can Be” by E. M. Swift (1980) provides many of the basic details about the 1980 U.S. Olympic hockey team and its coach.

Defining and Measuring Cohesion

“Group Cohesion: From ‘Field of Forces’ to Multidimensional Construct” by Kenneth L. Dion (2000) reviews key issues in the study of cohesion, with a focus on definitional debates and problems in measurement.

“Measuring Team Cohesion: Observations from the Science” by Eduardo Salas, Rebecca Grossman, Ashley M. Hughes, and Chris W. Coultas (2015) reviews the many and varied methods investigators have used to measure group cohesion, and offers researchers clear suggestions for methods that best suit their specific needs.

Group Development

Group Processes: A Developmental Perspective by Susan A. Wheelan (2005) reviews theory and research pertaining to the development of task-focused groups.

“Team Development” by Claire B. Halverson (2008) provides a practical guide to dealing with the changes groups experience over time.

Consequences of Cohesion

“Modeling Reciprocal Team Cohesion–performance Relationships, as Impacted by Shared Leadership and Members’ Competence” by John E. Mathieu, Michael R. Kukenberger, Lauren D’Innocenzo, and Greg Reilly (2015) investigates the cohesion–performance relationship by examining the unique empirical findings generated by studies that make use of cross-lagged panel designs.

“The Essence of Military Group Cohesion” by Guy L. Siebold (2007) provides a clear introduction to the standard model of cohesion that guides experts’ analyses of cohesion in combat groups.

Hazing

Wrongs of Passage: Fraternities, Sororities, Hazing, and Binge Drinking by Hank Nuwer (1999) remains one of the best researched analyses of hazing on college campuses.

Chapter Case: U.S. Olympic Hockey Team

“A Reminder of What We Can Be” by E. M. Swift (1980) provides many of the basic details about the 1980 U.S. Olympic hockey team and its coach.

Defining and Measuring Cohesion

“Group Cohesion: From ‘Field of Forces’ to Multidimensional Construct” by Kenneth L. Dion (2000) reviews key issues in the study of cohesion, with a focus on definitional debates and problems in measurement.

“Measuring Team Cohesion: Observations from the Science” by Eduardo Salas, Rebecca Grossman, Ashley M. Hughes, and Chris W. Coultas (2015) reviews the many and varied methods investigators have used to measure group cohesion, and offers researchers clear suggestions for methods that best suit their specific needs.

Group Development

Group Processes: A Developmental Perspective by Susan A. Wheelan (2005) reviews theory and research pertaining to the development of task-focused groups.

“Team Development” by Claire B. Halverson (2008) provides a practical guide to dealing with the changes groups experience over time.

Consequences of Cohesion

“Modeling Reciprocal Team Cohesion–performance Relationships, as Impacted by Shared Leadership and Members’ Competence” by John E. Mathieu, Michael R. Kukenberger, Lauren D’Innocenzo, and Greg Reilly (2015) investigates the cohesion–performance relationship by examining the unique empirical findings generated by studies that make use of cross-lagged panel designs.

“The Essence of Military Group Cohesion” by Guy L. Siebold (2007) provides a clear introduction to the standard model of cohesion that guides experts’ analyses of cohesion in combat groups.

Hazing

Wrongs of Passage: Fraternities, Sororities, Hazing, and Binge Drinking by Hank Nuwer (1999) remains one of the best researched analyses of hazing on college campuses.

Chapter 6. Structure

Groups are dynamic—characterized by continual change and adjustment—but these processes are patterned ones—shaped by the group’s social order. In all but the most ephemeral groups, members are distributed into roles, and their behavior when in these roles is regulated by norms that dictate what is and what is not proper conduct. The group structure also includes the relationships that join members to one another in an integrated network that regulates interdependencies and increases the group’s unity and durability.

What is group structure?

What are norms, how do they develop, and how do they work to regulate behavior?

What kinds of roles are common in groups, and how do they influence members?

How do social networks shape status, attraction, and communication processes in groups?

Andes Survivors: One Group’s Triumph Over Extraordinary Adversity

The Old Christian’s rugby team had chartered the Fairchild F-227 twin-engined turboprop to travel from Montevideo, Uruguay to Santiago, Chile. Most of the passengers on the flight were members of the team or their family and friends. But they never reached their destination. The pilots misjudged their course and began their descent too soon. The plane clipped the peak of Mt. Tinguiririca and crashed in the snow-covered Andes.

Those who survived the crash struggled to stay alive in the harsh, subzero temperatures of the barren Andes. During the first days of the ordeal, they argued intensely over the likelihood of a rescue. Some insisted that searchers would soon find them. Others wanted to climb down from the mountain. Some became so apathetic that they didn’t care. But the search planes never spotted them, and their hopes began to fade when a second tragedy struck the group: Early one morning an avalanche filled the wrecked fuselage where they slept with snow and many died before they could dig their way out.

A lone individual would have certainly perished in the harsh climate. But the group, by pooling its resources, survived. They organized their work, with some cleaning their sleeping quarters, some tending the injured, and others melting snow into drinking water. When their food ran out, they made the difficult decision to eat the frozen bodies of those who had died in the crash. And when starvation seemed imminent, they sent two men, Fernando Parrado and Roberto Canessa, down the mountain to seek help. After hiking for 14 days, the two explorers, running low on food and supplies, chanced on a farmer tending his cattle. Parrado himself guided the rescue helicopters back to the crash site. All of them, when asked how they survived, credited the unity of the group. When they read author Pier Paul Read’s book about their ordeal, Alive, they complained of only one inaccuracy: They felt that he failed to capture the “faith and friendship which inspired them” for 70 days in the frozen cordillera (Read, 1974, p. 10).

The group that came down from the Andes was not the same group that began the flight. Many members were lost to the group forever, and the trauma changed each one of the survivors permanently. But its group structure also changed. The structure of a thing is the relatively fixed arrangement of and relations among its constituent elements that links those elements together to form a single integrated whole. In groups, structure creates the social order, including the regulatory standards that define how members are supposed to behave (norms) given their position in the group (roles) and the connections among members (intermember relations). When the group was a rugby team, members paid heed to a very different set of norms than they did when they were transformed into a group fighting for its very survival. The group began the flight with one set of roles and positions—a captain, a coach, parents, supporters, and friends—but ended with an entirely different set of roles, including leaders, helpers, and explorers. The network of relationships linking members one to another, in terms of status, liking, and communication, also changed. Men who were at first afforded little respect or courtesy eventually earned considerable status within the group. Some who were well-liked before the crash became outcasts. Some who had hardly spoken to the others became active communicators within the group.

If you want to understand a group—whether one stranded in the Andes, sitting at a conference table, or working to manufacture some product—you need to understand its structure. Explaining group behavior in structural terms is analogous to tracing individuals’ actions back to their personalities. Personality traits and dispositions cannot be observed directly, yet they influence people’s actions across time and settings. Similarly, a structural analysis assumes that group interactions follow a predictable, organized pattern because they are regulated by influential interpersonal structures. This chapter examines three of these core determinants of a group’s “personality”: norms, roles, and relations (see Biddle, 2001; Hechter & Op, 2001 for reviews).

6-1. Norms

The survivors of the crash needed to coordinate their actions if they were to stay alive. With food, water, and shelter severely limited, they were forced to interact with and rely on each other continually, and any errant action on the part of one person would disturb and even endanger several other people. So, members began to abide by a set of unstated rules that defined how the group would sleep at night, what types of duties each healthy individual was expected to perform, and how food and water were to be apportioned. These consensual, regulatory standards are norms.

6-1a. The Nature of Social Norms

Norms are a fundamental element of social structure; the group’s rules of order (Fine, 2012). As group standards, they provide direction and motivation, organize social interactions, and make other people’s responses predictable and meaningful. Both simple behaviors, such as choice of clothing, manners, and conversational pleasantries, and more complex social processes, such as fairness, morality, and justice—are based on norms. Each group member is restrained to a degree by norms, but each member also benefits from the order that norms provide.

How people speak, dress, and interact are determined by norms that define conventional behavior within that society. But each group within that society may adopt its own unique normative standards. For example, in one group it may be appropriate to interrupt others when they are talking, to arrive late and leave meetings early, and to dress informally. In another group, such behaviors would be considered inappropriate violations of group norms of dress and decorum. These group-specific norms combine to form the group’s consensually accepted knowledge, beliefs, rituals, customs, rules, language, norms, and practices: the group’s culture (Schein, 1990).

Types of Norms

Norms prescribe the socially appropriate way to respond in the situation—the normal course of action. Just as a physician’s prescription recommends a medicine, so prescriptive norms define the socially appropriate way to respond in a situation. Proscriptive norms, in contrast, are prohibitions; they define the types of actions that should be avoided if at all possible (Sorrels & Kelley, 1984). For example, some of the prescriptive norms of the Andes group were “Food should be shared equally” and “Those who are not injured should work to help those who are injured,” whereas some proscriptive norms were “Do not urinate inside the airplane” and “Do not give up hope.”

Norms also differ in their evaluative implications. Descriptive norms describe what most people usually do, feel, or think in a particular situation. None of the survivors in the Andes groups had ever been in such a desperate situation before, and so they intuitively did what most of the others did: They listened to the radio for information about the rescue and did what the leader told them to do. Everyone did not perfectly match their behavior up to these norms, but most did. This high degree of similarity in everyone’s actions provided members with information about how they should think, feel, and act. Social psychologist Robert Cialdini’s (2009) calls the tendency for people to use other people’s responses as useful information about how they should themselves respond the principle of social proof: People assume that a behavior is the correct one when they see others performing it.

Injunctive norms are more evaluative—they describe the sorts of behaviors that people ought to perform—or else (Gibbs, 1965). People who do not conform to descriptive norms may be viewed as unusual, but people who violate injunctive norms are negatively evaluated and are open to sanction by the other group members (Rimal & Real, 2005). In the Andes group, for example, those who failed to do their fair share were criticized by the others, given distasteful chores, and sometimes even denied food and water. People who violate injunctive norms are disliked, assigned lower status jobs, pressured to conform, and, in some cases, excluded from the membership. As sociologist Gary Alan Fine (2012) explains, “norms are not merely behavioral regularities but involve a collective embrace of the propriety of this regularity” (p. 69).

Internalization of Norms

Norms are not simply external rules but internalized standards. Members comply with their group’s norms not because they have to but because they want to: The group’s norms are their own, personal norms. Women who are members of groups whose norms emphasize healthy eating habits personally endorse healthier dieting practices (Cruwys et al., 2015). People who live in communities that stress the importance of pro-environmental practices such as water conservation themselves express more positive attitudes about protecting the environment (White et al., 2009). Students who enroll in a school whose norms support liberal rather than conservative causes come to accept their school’s position on political issues as their own personal position on those issues (Newcomb, 1943). Students report significantly more positive emotions when they think they have done something that is consistent with the norms of their school, and more negative emotions when they have acted in ways that violate their school’s norms—especially if they identify with their school and if the norm they violated was an injunctive one (Christensen et al., 2004).

Acting in ways that run counter to norms is personally upsetting—as social psychologist Stanley Milgram discovered when he asked his students to violate a norm deliberately. He asked his students to board a New York City subway and to perform a simple counter-normative behavior: asking someone for their seat. In this situation, all interactants recognize and accept the rule “All seats are filled on a first-come, first-served basis,” so asking for people to give up their seat is a norm violation. Still, many people gave up their seats, apparently because the request took them by surprise and they wanted to avoid interaction, or because they normalized the situation by concluding that the requestor was ill. Milgram was particularly intrigued, however, by the reactions displayed by the norm violators. Even though they were volunteers who were deliberately breaking the situational norms in the name of research, all experienced severe emotional turmoil as they approached the situation. “Students reported that when standing in front of a subject, they felt anxious, tense, and embarrassed. Frequently, they were unable to vocalize the request for a seat and had to withdraw” (Milgram, 1992, p. 42). Milgram, who also performed the norm violation task, described the experience as wrenching, and concluded that there is an “enormous inhibitory anxiety that ordinarily prevents us from breaching social norms” (p. xxiv).

6-1b. The Development of Norms

Where do a group’s norms come from? Some are deliberately put in place when a group is established; the founders of a group may make explicit the dos and don’ts for a group and make acceptance of these standards a condition for members. In many cases, too, a group that weathers a difficult issue or experience may endorse new standards that will provide guidance in the future. But even if the group does not explicitly adopt standards norms will develop as members gradually align their behaviors with commonly accepted behavioral practices in the group. People do not, for example, spend a great deal of time wondering “Should I be quiet in the library?” “Should I nap during the group meeting?” or “Should I stop when the light turns red?” Most take these norms for granted so fully that their guiding influence is only noticed when someone violates them (Aarts, Dijksterhuis, & Custers, 2003). Like the members of a concert orchestra tuning their instruments before a performance, this social tuning results in the alignment of each individual’s action with the actions of others around him or her (Lun et al., 2007).

Do Your Facebook Friends Know the Norms?

Groups that once convened in physical locations, such as clubs, conference rooms, and bars, now meet virtually in social networking sites. Individuals who use the very popular Facebook social networking site, for example, connect with others by “friending” each other. One person sends another person a “friend request,” and the person who receives the request must accept this offer to create the friend link.

Just as the interactions that occur in face-to-face groups are guided by a set of rules of comportment, so norms have emerged on social networks that define what is appropriate and what is not (Bryant & Marmo, 2012).

Communication norms set expectations about how frequently one should communicate with friends and the types of messages that should be exchanged.

Relationship maintenance norms encourage using Facebook to strengthen one’s relationship with others, primarily by keeping up-to-date on what is happening in others’ lives and also by wishing them happy birthday via Facebook.

Negative self-consequences norms include taking steps to avoid causing oneself harm when using Facebook (e.g., “I should not let Facebook use with this person interfere with getting my work done”).

Negative friend-consequences norms warn against doing things that might harm one’s friends (e.g., “I should consider how a post might negatively impact this person’s relationships”).

Deception and control norms define what steps to take if one violates Facebook norms. For example, “I should delete or block this person if he/she compromises my Facebook image” and “I should intentionally control the level of access this person has to my profile.”

Not heeding these norms will not yield the same consequences that violating a Facebook official users rule will. As Facebook explains, “If you violate the letter or spirit of this Statement, or otherwise create risk or possible legal exposure for us, we can stop providing all or part of Facebook to you.” But violating these social norms may be injurious socially. No one is directly taught the norms of Facebook, but those who do not follow them will likely pay the ultimate Facebook price: They will get themselves unfriended.

The Andes survivors, for example, grew up in a culture where cannibalism was taboo. But when the group grew weak from starvation, one member casually remarked that the only source of nourishment was the frozen bodies of the crash victims. The others ignored his remark until the tenth day when “the discussion spread as these boys cautiously mentioned it to their friends or those they thought would be sympathetic” (Read, 1974, p. 76). When the topic was discussed by the entire group, two cliques emerged: one favored eating the corpses and the other opposed breaking this norm. But when the group learned by radio that the air force had given up the search, most of the members ate a few pieces of meat; over time, cannibalism became the norm (Parrado, 2006).

Social psychologist Muzafer Sherif (1936) studied this social tuning process by taking advantage of the autokinetic (self-motion) effect. This visual illusion occurs when a person stares at a pinpoint of light in an otherwise dark room. Ordinarily the visual system compensates for naturally occurring motions of the eye, but when only a single light is visible with no frame of reference, the light appears to wander in unpredictable directions and at variable speeds. Sherif found that when individuals judged the dot’s movement repeatedly, they usually established their own idiosyncratic average estimates that varied from 1 to 10 inches. But when people made their judgments in groups, their personal estimates blended with those of other group members. One group, for example, included three people who had already been tested individually. During these initial tests, person A thought the light moved very little—about 1 inch. Person B estimated the movement at 2 inches, but C’s estimates were higher, averaging about 7 inches. When these three people made their estimates of the movement aloud when seated together, their judgments converged. It took three meetings, but by the third session, a norm had emerged: All the members felt the light was moving about 3 inches. Figure 6.1 graphs this convergence process: Over time, individuals with the highest and lowest estimates revise their judgments to match the group average.

6-1c. The Transmission of Norms

Sherif confirmed that norms emerge, gradually, as group members’ behaviors, judgments, and beliefs align over time (Echterhoff, Higgins, & Levine, 2009). But Sherif also arranged for people to make their judgments alone after taking part in the group sessions when a norm emerged. Did these individuals revert back to their original estimates of movement, or did they continue to base their estimates on the norm that emerged within their group? Sherif discovered that, even though the other group members were no longer present, the individuals’ judgments were still consistent with the group’s norms (Sherif, 1966). They had internalized the norm.

Norms, because they are both consensual (accepted by many group members) and internalized (personally accepted by each individual member), are social facts—taken-for-granted elements of the group’s stable structure. Even if the individuals who originally fostered the norms are no longer present, their normative innovations remain a part of the organization’s culture, and newcomers must change to adopt that tradition. Researchers have studied this norm transmission experimentally using a generational paradigm: They create a group and then add newcomers to it and retire old-timers until the entire membership of the group has turned over. Do these succeeding generations of members remain true to the group’s original norms, even if these norms are arbitrary or cause the group to make errors and mistakes? In one autokinetic effect study, researchers established an extreme norm by planting a confederate in each three-member group. The confederate steadfastly maintained that the dot of light was moving about 15 inches—an excessive estimate given that most estimates averaged about 3–4 inches. Once the confederate deflected the group’s distance norm upward, he was removed from the group and replaced by a naive participant. The remaining group members, however, still retained the large distance norm, and the newest addition to the group gradually adapted to the higher standard. The researchers continued to replace group members with new participants, but new members continued to shift their estimates in the direction of the group norm. This arbitrary group norm gradually disappeared as judgments of distance came back down to an average of 3.5 inches, but in most cases, the more reasonable norm did not develop until group membership had changed five or six times (Jacobs & Campbell, 1961; MacNeil & Sherif, 1976). In another generational study, researchers gave groups feedback that suggested that their norm about how decisions should be made was causing them to make errors, but this negative feedback did not reduce the norm’s longevity across generations (Nielsen & Miller, 1997).

Studies of the emergence and transmission of norms in a variety of settings—in workgroups, families, sports teams, and children’s groups—all demonstrate just how rapidly norms can emerge to structure group behavior (e.g., Bicchieri, 2006; Rossano, 2012). Even children as young as three years old learn norms quickly and respond negatively when newcomers violate these rules. In one study, for example, two- and three-year-olds played for a time with a familiar object—a sponge—in two areas of the day care center (Rakoczy et al., 2009). When in the rest area, they were taught to use the sponge to clean. But when in the games area, they were taught to roll the sponge as part of a game. Later, someone else (a puppet, actually) tried to use the sponge for cleaning in the game area or for playing in the rest area, and the researchers watched to see how the children reacted. As they expected, three-year-olds complained when the sponge was misused, saying things like “No! It does not go like this” or “No, you are not allowed to clean up here.”

6-1d. Application: Norms and Health

Because norms tend to resist revision, some group’s norms may seem pointless and arbitrary rather than reasonable and functional (see Table 6.1). But norms are essential to group functioning, for even odd or unusual norms organize interactions, increase predictability, and enhance solidarity (Collins, 2004). Why do people feel obligated to help others who help them (Chapter 3)? Why do members of cohesive groups tend to work only as hard as everyone else in the group (Chapter 5)? Why do people sometimes abandon their own beliefs and adopt the group’s position as their own (Chapter 7)? Why do group members feel justified in retaliating against other members who treat them harshly (Chapter 13)? Why do people who are part of social movements or large crowds sometimes engage in aberrant behavior (Chapter 17)? Because of the influence of norms on members’ thoughts, feelings, and actions.

Why Do Unhealthy Norms Persist?

Why do group members continue to conform to norms that are harmful rather than healthy? The answer lies, in part, in the tenacity of norms. Even norms that run counter to society’s general traditions can establish a life of their own in small groups within that society. In some cases, too, norms remain in place because of pluralistic ignorance: Members privately disagree with the group’s norm, but feel that their outlook is shared by few others in the group. So, the norm continues to regulate behavior due to misperception rather than shared consensus. College students, for example, often misperceive the extent to which other students drink excessive amounts of alcohol. Even though most of the students who participated in one study were personally opposed to overindulgence, they believed that their campus’s norms encouraged heavy alcohol consumption. The men responded to this norm by gradually internalizing the misperceived norm. They began to drink more the longer they stayed at the school. The women, in contrast, responded by distancing themselves from their university and its norms about drinking (Krieger et al., 2016; Miller & Prentice, 2016).

Norms about Drinking Alcohol

Norms regulate interactions in groups, facilitate productivity, and limit conflict, but such negative, unhealthy behaviors as alcohol abuse, overeating, and drug use can also be traced to normative processes. Consider, for example, the impact of social norms on the consumption of alcohol by young adults. Heavy use of alcohol is common on college campuses, even among students who are too young to possess alcohol legally. This excessive use of alcohol is associated with a number of negative outcomes, including lower performance, physical injury, and violence (Padon et al., 2016). Yet, drinking excessively is considered “normal” on many campuses. Polls of students suggest that most consider drinking five drinks in succession to be appropriate if one is partying, and more than 40% of students conform to that norm (Neighbors et al., 2008). Many students, when asked about the drinking norms endorsed by the groups to which they belong, such as their primary friendship groups, campus clubs, and sororities and fraternities, reported that these groups often approve of “drinking alcohol every weekend” and “drinking enough to pass out” (LaBrie et al., 2010, p. 345)—and the stronger the group’s endorsement of drinking, the more heavily students who belong to such groups drank. Many students, when asked about clearly abnormal consequences of drinking—such as substantial loss of memory (e.g., blackouts), loss of consciousness (“passing out”), physical impairment (e.g., dizziness and staggering gait), and illness (vomiting)—considered them to be “normal” rather than abnormal (e.g., Mallett et al., 2011).

Norms and Health

Normative processes contribute to many other unhealthy behaviors besides excessive use of alcohol. Obesity, for example, tends to spread among individuals who are linked together in a social network, in part because norms encourage lifestyle choices that promote weight gain rather than fitness (Christakis & Fowler, 2007). Interventions designed to help at-risk adolescents by placing them in special programs may actually contribute to increased violence, drug use, and other antisocial behaviors when these groups develop negative rather than positive norms (Dishion & Dodge, 2005). Individuals who are frequent users of “club drugs” (methamphetamine, cocaine, ketamine, ecstasy, GHB, and LSD) are more likely to report that they do so because of social pressure (e.g., “When I was out with friends and they kept suggesting we go somewhere to do drugs,” “When I felt pressured to use drugs and felt I couldn’t refuse”; Starks et al., 2010, p. 1067).

Eating disorders, too, have been linked to normative processes. Social psychologist Chris Crandall (1988), for example, documented the detrimental effects of norms on group members in a study of one particular unhealthy behavior: bulimia—a pernicious cycle of binge eating followed by self-induced vomiting or other forms of purging. Certain social groups, such as cheerleading squads, dance troupes, sports teams, and sororities, tend to have strikingly high rates of eating disorders (Petrie & Greenleaf, 2007). In explanation, Crandall noted that such groups adopt norms that encourage binging and purging. Rather than viewing these actions as abnormal and a threat to health, the sororities that Crandall studied accepted purging as a normal means of controlling one’s weight. The women who were popular in such groups were the ones who binged at the rate established by the group’s norms. Even worse, women who did not binge when they first joined the group were more likely to take up the practice the longer they remained in the group. Other studies suggest that unhealthy eating patterns increase with the perceived strength of peer pressure within the sorority and the longer the woman lives in the sorority house itself (e.g., Basow, Foran, & Bookwala, 2007).

Norms may, however, promote healthy actions as well as unhealthy ones. Individuals who wish to reduce their negative indulgences often find success by joining a group and accepting that group’s norms as their own. Many fitness, weight-loss, and anti-addiction programs, as noted in more detail in Chapter 16, take a group approach to change. Alcoholics Anonymous, for example, has clear norms about the types of behaviors members must enact in order to stay sober, and those individuals who become highly involved members are less likely to continue to drink heavily (Witbrodt et al., 2012). Groups have also been found to be effective in preventing the onset of eating disorders, such as bulimia, in young women (Stice et al., 2008). Groups, then, can either promote or threaten members’ health, depending on their norms. Some groups may put members at risk by encouraging unhealthy actions, whereas others keep members on the path to good health and wellness (Miller & Prentice, 2016).

6-2. Roles

On the day after the Andes crash, Marcelo, the captain of the rugby team, organized the efforts of those who could work. Two young men and one of the women administered first aid to the injured. One subgroup of boys melted snow for drinking water, and another team cleaned the cabin of the airplane. These various positions in the group—leader, doctor, snow melter, cabin cleaner—are all examples of roles: coherent sets of behaviors expected of people in specific positions (or statuses) within a group or social setting.

6-2a. The Nature of Social Roles

The concept of roles explains the changes that some people exhibit when they become members of a group. The quiet recluse, taciturn by nature, may become convivial when given responsibility for organizing the group’s annual fund-raising event. The otherwise mild-mannered colleague may become habitually critical of process when taking part in group discussions. The staffer with the messiest office may become methodical and precise when elected the group’s secretary. Groups may pressure members to conform to the group’s norms, but they do not require that members all act the same as one another. Groups instead require members to enact a specific set of behaviors consistently, depending on one’s role within the group. Roles define responsibilities and expectations and facilitate coordination by specifying who can be counted on to do what within the group. Over the course of repeated interactions, members not only learn what others in the group usually do, but also what everyone in the group expects of them. By enacting roles, individuals establish an exchange relationship with their fellow members, building the interdependence that is essential for the coordination of behavior, group cohesion, and productivity.

The concept of a social role is similar, in many respects, to a theatrical role. In a play or film, a role is the part the actors portray before the audience. To become Juliet in Shakespeare’s Romeo and Juliet, for example, an actor must perform certain actions and recite her dialogue accordingly. Similarly, roles in groups structure behavior by dictating the part that members take as they interact. Once cast in a role, such as leader, outcast, or questioner, members perform certain actions and interact with others in a particular way—but this consistency reflects the requirements of their role rather than their personal predilections or inclinations (Stets & Thai, 2010). But just as some variability is permitted in theatrical roles, group roles do not structure group members’ actions completely. An actor playing the role of Juliet must perform certain behaviors as part of her role—she would not be Shakespeare’s Juliet if she did not fall in love with Romeo. She can, however, recite her lines in an original way, change her stage behaviors, and even ad-lib. In social groups, too, people can fulfill the same role in somewhat different ways, and, so long as they do not stray too far from the role’s basic requirements, the group tolerates this variation. However, like the stage director who replaces an actor who presents an unsatisfactory Juliet, the group can replace members who repeatedly fail to play their part within the group. The role often supersedes any particular group member. When the role occupant departs, the role itself remains and is filled by a new member (Stryker & Burke, 2000).

Role Differentiation

As with norms, groups sometimes deliberately create roles to organize the group and thereby facilitate the attainment of the group’s goals. A group may decide that its efficiency would be augmented if someone takes charge of the meetings and different tasks are assigned to subcommittees. In some cases, too, someone outside the group, such as the group’s supervisor, may mandate roles within the group (Kozlowski et al., 2015). But even without a deliberate attempt at creating a formal group structure, the group will probably develop an informal role structure. Members may initially consider themselves to be just members, basically similar to each other. But in time, some group members will begin to perform specific types of actions and interact with other group members in a particular way. As this role differentiation process unfolds, the number of roles in the group increases, whereas the roles themselves gradually become more narrowly defined and specialized. In the Andes survivors, for example, the roles of leader, doctor, and cleaner emerged first, soon followed by the inventor who created makeshift snowshoes, hammocks, and water-melting devices; explorer who was determined to hike down from the mountain; and complainer, pessimist, optimist, and encourager. This proliferation of roles is typical of groups facing difficult problems or emergencies (Bales, 1958).

Task and Relationship Roles

What roles emerge more frequently in groups? Certainly, the role of leader is a fundamental one in many groups, but other roles should not be overlooked. Many of these roles, such as solution seeker, the problem analyst, and procedural facilitator, are similar in that they revolve around the task the group is tackling (Fujimoto, 2016). People who fulfill a task role focus on the group’s goals and on the members’ attempts to support one another as they work. Marcelo, in the Andes group, was a task-oriented leader, for he organized work squads and controlled the rationing of the group’s food supplies; the rest of the members obeyed his orders. He did not, however, satisfy the group members’ interpersonal and emotional needs. As if to offset Marcelo’s inability to cheer up the survivors, several group members became more positive and friendly, actively trying to reduce conflicts and to keep morale high. Liliana Methol, in particular, provided a “unique source of solace” (Read, 1974, p. 74) to the young men. She came to fill a relationship role (also frequently termed socioemotional role) in the Andes group. Whereas the coordinator and energizer structure the group’s work, such roles as supporter, clown, and even the complainer help satisfy the emotional needs of the group members (Lehmann-Willenbrock, Beck, & Kauffeld, 2016).

Why Differentiation?

Why do task roles and relationship roles emerge in so many different groups? One answer, proposed by Robert Bales (1950, 1955, 1958), suggests that very few individuals can simultaneously fulfill both the task and the relationship needs of the group. When group members are task-oriented, they must direct others to act in certain ways, restrict others’ options, criticize other members, and prompt them into action. These actions may be necessary to reach the goal, but others may react negatively to these task-oriented activities, so they then look to others in the group for socioemotional, relational support. The peacekeeper who intercedes and tries to maintain harmony is the relationship specialist. Task and relationship roles, then, are a natural consequence of these two partly conflicting demands.

Bales identified these tendencies by tracking role differentiation in decision-making groups across four sessions. Bales used his interaction process analysis (IPA) system to identify certain specific types of behavior within the groups. As noted in Chapter 2 (see Figure 2.2), half of the categories in IPA focus on the task-oriented behaviors and the other half focus on the relationship behaviors. Bales found that individuals rarely performed both task and relationship behaviors: Most people gravitated toward either a task role or a relationship role. Those who took on a task role (labeled the “idea man”) offered mostly suggestions and expressed opinions. Those who gravitated to the relationship roles (labeled the “best-liked man”) showed solidarity, more tension release, and greater agreement with other group members. Moreover, this differentiation became more pronounced over time. During the first session, the same leader occupied both the task and the relationship roles in 56.5% of the groups. By the fourth session, only 8.5% of the leaders occupied both roles. In most cases, individuals dropped their role as task leader in favor of the relationship role (Bales, 1958; Bales & Slater, 1955).

Subsequent work suggests that this division of task and relationship roles is not an inevitable occurrence in groups (Turner & Colomy, 1988). Some individuals are the small-group equivalent of master leaders, for they are both well-liked and focused on the work to be done (Borgatta, Couch, & Bales, 1954). When players on football teams were asked to identify the best players on the team and those who contributed most to the group’s harmony, many named the same person—usually a senior- or first-string player—to both roles (Rees & Segal, 1984). When students in groups rated each other’s role activity, many slotted the same person into both task and relationship roles. Cohesive groups tend to have leaders who can fill both roles, whereas the roles tend to be separated in groups with high levels of conflict (Burke, 1967; Mudrack & Farrell, 1995). Differentiation of these two types of roles is more common than their combination, however, perhaps because few people have the interpersonal and cognitive skills needed to enact both roles successfully.

6-2b. Role Theories

The concept of role is a redoubtable one and has given rise to a number of alternative theories that describe roles and role-related processes. These conceptualizations agree on many points. Roles, they note, organize group interactions by creating a set of shared expectations that script the behavior of the individuals who occupy them. These theories, however, disagree on other points, including what roles are common to most groups and what functions roles serve. This section reviews a sample of theories that seek to explain roles and role-related processes, focusing on three models that are most relevant to understanding groups: functional role theory, interactionism, and dynamic role theory (Biddle, 1986; Turner, 2001b).

Functional Role Theories

A number of theorists, in seeking to explain why roles develop in groups, stress their functional utility. All groups must, for example, respond in adaptive ways to their environment by recognizing challenges and responding successfully. Most groups, too, usually exist for some purpose, so they must develop methods that facilitate goal attainment—all the while making certain that members are so satisfied with membership that they remain in the group and continue to meet their obligations (Parsons, Bales, & Shils, 1953). Roles exist in groups to fulfill, at least in part, these personal and interpersonal needs (Belbin, 2010; Blumberg et al., 2012).

Education theorists and practitioners Kenneth Benne and Paul Sheats (1948) developed their well-known functional theory of roles by observing the interactions of groups at the National Training Laboratories (NTL), an organization devoted to the improvement of groups. They noted that while much work had been done to train people to lead groups, little had been done to train people to work in groups—even though the “setting of goals and the marshaling of resources to move toward these goals is a group responsibility in which all members of a mature group come variously to share” (pp. 41–42). Leaders, they suggested, are responsible for making sure roles are filled, but members are responsible for fulfilling the demands of the roles.

Benne and Sheats suggested that a group, to survive, must meet two basic demands: The group must accomplish its tasks, and the relationships among members must be maintained. But they identified 19 specific roles within these two broad categories, such as initiator/contributor, opinion seeker, and energizer on the task side and encourager, harmonizer, and compromiser on the relationship side. They also identified a third set of eight individual roles enacted by group members who are more concerned with their own personal needs rather than the needs of the group. This category includes such roles as aggressor, blocker, and dominator.

Benne and Sheats theorize that individuals, given previous experiences in groups and differences in personality, naturally gravitate to a particular type of role across all the groups they join. However, because a group’s need for a particular role will vary depending on the type of task it is attempting and the group’s stage of development, the most skilled group member is one with role flexibility: the capacity to recognize the current requirements of the group and then enact the role-specific behaviors most appropriate in the given context. A group striving to be creative, for example, has less need of an evaluator/critic than a group examining a range of solutions after extensive deliberation. Individual roles will prove more problematic in the early stage of a group’s life.

Interactionist Theories

Some theoretical analyses of roles put more emphasis on the generative process of role-related actions, arguing that “the patterning of behavior that constitutes roles arises initially and recurrently out of the dynamics of interaction” (Turner, 2001b, p. 234). Group members share a basic sense of the requirements of the roles that are common in most group settings, but they work out the details of their roles and their demands as they interact with one another. Interactionist approaches recognize that group roles are analogous to theatrical roles, but the group setting is more like improv than a well-rehearsed stage play. Roles are negotiated by all group members through a reciprocal process of role enactment—displaying certain behaviors as part of one’s role in the group—and role sending—the transmission of one’s expectations about what kinds of behaviors are expected of people who occupy particular roles (Stryker & Vryan, 2006).

What Roles Are Filled and by Whom in Your Group?

Although the parts people play in their groups—their roles—may be as varied as the diverse situations where groups exist, some roles occur with great regularity across all kinds of groups. Most groups need someone who consistently keeps the group on course; someone who intervenes to soothe the group’s feelings; someone who cheers the group up with its mood turns dark. Groups have needs, just as their members do.

Instructions. Take a moment to reflect on the roles in a group to which you belong. The group can be one that meets regularly in a work or social setting, or it can be a smaller group that is part of a larger group or organization. After you generate the list of all the group’s members, can you identify a person in the group who regularly takes on one or more of the following roles (including yourself, of course)?

Group Task Roles

Initiator/contributor: Offers novel ideas about the problem at hand, new ways to approach the problem, or different ways of doing or organizing things

Information seeker: Emphasizes getting the facts clear or seeks out verifying data

Opinion seeker: Asks for more information about the attitudes and values of the group members on issues

Information giver: Provides objective facts and information, drawing on own expertise or claimed authority

Opinion giver: Expresses own opinions and values on issues

Elaborator: Gives additional information, uses examples to clarify suggestions, or explores implications

Coordinator: Organizes the various contributions of others and shows their relevance and relationship to the overall problem

Orienter: Offers summaries, guides the group back to its purposes, or reminds group to stay focused

Evaluator/critic: Appraises the quality of the group’s activities, methods, processes, or results

Energizer: Stimulates the group to continue working when discussion drags

Procedural technician: Cares for operational details, such as arranging the room, providing materials, and attending to technology

Recorder: Takes notes and maintains records

Group Building and Relationship Roles

Encourager: Reassures others by expressing agreement, praise, and solidarity

Harmonizer: Reconciles disagreements and conflicts among group members (often through humor)

Compromiser: Agrees to shift his or her own position on an issue to reduce conflict or move the group along

Gatekeeper/expediter: Facilitates communication by drawing out reticent members and proposing ways to improve the discussion

Standard setter: Calls for discussion of the group’s procedures and processes

Group observer and commentator: Tracks the group’s processes and provides data during the review of group’s procedures

Follower: Agrees with others in the group

Individual Roles

Aggressor: Expresses disapproval of acts, ideas, and feelings of others; denigrates the group, purposes, and procedures

Blocker: Resists the group’s influence, often by not dropping issues after the group has settled them

Recognition seeker: Protects and promotes own status, accomplishments

Self-confessor: Discloses excessively personal interests, feelings, and opinions unrelated to group goals

Playboy: Shows a lack of engagement by expressing cynicism, boredom, and so on

Dominator: Asserts authority or superiority; manipulative

Help seeker: Expresses insecurity, confusion, and self-deprecation

Special-interest pleader: Remains apart from the group by invoking an extra-group identity

Interpretation. The number of roles in your group, the type of roles, and the distribution of members in those roles is indicative of your group’s functioning. If, for example, your group has very few individuals in task roles, your group is either focused more on relational concerns or it might still be developing role structures that facilitate its performance (Benne & Sheats, 1948).

This view is consistent with the sociologist Erving Goffman’s (1959) dramaturgical approach to social interaction. He maintained that individuals engage in self-presentation (also termed impression management) in order to steer others’ impressions and expectations. Marcelo, for example, as the rugby team captain, considered himself well-suited for the role of group leader following the crash, so he exhibited the kinds of behaviors appropriate for that role: He assigned tasks to the others, set goals for the group, and sanctioned members who did not do their part. He did so because these actions were required by the role of leader, but also because he wished to define himself as the leader in the other group members’ eyes. Roles, then, are negotiated among members through a process that requires motivation, experience, and the ability to step out of one’s own role and mentally imagining how others in the groups are seeing you. This process is termed role-taking. It includes not only taking on a role but also a willingness to put oneself into others’ roles to see the group as they do (Mead, 1934).

Beth A. Bechky (2006), an ethnographer of work and occupation, applied an interactionist approach to roles in her participant observation study of the interpersonal dynamics of temporary film-production teams. These production groups are common in the construction and film industries. The groups are assembled by production managers or contractors, and they set to work immediately on their task. Bechky, for example, observed crews ranging in size from 35 to 175 set up and film commercials, music videos, and movies. These group members may have worked together previously, but usually as individuals—they were not members of any larger production teams that moved as a unit from project to project. These groups can assemble quickly, however, because each group member’s role is relatively well-defined. Filming a commercial, video, or movie requires completion of a number of highly specialized tasks, and these tasks are assigned to specific roles. The production team includes, for example, camera assistants and operators, costume designers, crane and dolly operators, directors, film loaders, gaffers (electricians), grips (lighting and rigging technicians), hair stylists, lighting technicians, location scouts, makeup artists, photographers, producers, and prop managers. Those who occupy these roles work in a highly coordinated manner during the actual production process.

Bechky also discovered, however, that the coordination of these groups is further enhanced on a continual basis through the interactions among the group members on each set. Many of those who work in a particular role have also performed other roles in other projects. An electrician on one shoot, for example, may have worked as a camera loader on another, and so the level of consensual agreement regarding roles is substantial. The complexity of the work, however, required even the most experienced members to modify their actions to suit the needs of others. Throughout every production day, group members were regularly providing each other with corrective feedback about one another’s role-based performances, but they did so through “effusive thanking, polite admonishing, and role-oriented joking” (p. 11). Group members openly and routinely thanked each other for completing actions required by their roles, but they were also quick to intervene with suggestions for improvement, if needed, usually phrased as “don’t worry about it for now, but next time be sure to…”. Humor was often used to give a lighter touch to these suggestions, but also to ridicule someone who had displayed a more jarring role-related misstep. Her observations convinced Bechky (2006, p. 11) that within each project, the generalized role structure is instantiated by a set of crew members who negotiate and modify their particular roles. The generalized role structure and the role enactments mutually support one another, while at the same time establishing a means for almost immediate coordination on each new project.

Dynamic Role Theories

Sigmund Freud (1922) is best known for his insightful analyses of personality and adjustment, but he also analyzed group behavior. He suggested our actions when in a group are based, in part, on our rational plans, motives, and goals, but also on unconscious interpersonal and psychological processes that are largely unrecognized. He believed, for example, that groups psychologically replace our first, and most basic, group: our family. This replacement hypothesis suggests that in highly cohesive groups, the other group members come to take the place of our siblings, so the emotional ties that bind us to our groups are like the ties that bind children to families. We also unconsciously accept our leaders as parental figures, so they become the “mother” or “father” of our groups. Those leaders who identify with the mother role adopt more relational behaviors, whereas those who identify with the father role engage in more task-oriented, agentic actions. Freud believed, however, that children were not wholly accepting of their parents, and so this ambivalence is transferred to our leaders. We therefore consider some leaders to be persecutory, critical “bad mothers” or “bad fathers” rather than benevolent, supportive, ones (Klein, 1948).

Do Group Roles Have a Deeper Meaning?

Organizational psychologist and group therapist Paul Moxnes (1999) extended Freud’s psychodynamic perspective in his theory of “deep roles.” Moxnes theorized that group members’ early experiences in families, with their apportionment of roles based on biological sex and cultural practices, creates similar structures in groups that are not kin-based. Like functional theorists, Moxnes distinguishes between task and relationship roles, but links this division to the separation of mother and father roles in families. He also suggests that these roles are viewed with some ambivalence, due to the differences in power associated with each role. The occupants of some role positions are powerful, whereas the occupants of other roles are less influential. On the basis of these assumptions, Moxnes identified a number of deep roles that he creatively named the king (the “good father”), queen (the “good mother”), princess (the daughter), and so on.

When Moxnes asked members of groups to use these role labels to describe their fellow group members, he found that the primary deep roles—King, Queen, Princesses, Prince, and Wiseman—were used very frequently (Moxnes & Moxnes, 2016). Members’ agreement on who in their group fit which role exceeded random chance: members tend to place the same person in the “good father” (king) role, “bad mother” role, and so on. Moxnes’ intriguing theory awaits further empirical testing (see Hare, 1999; Robison, 1999).

Bale’s SYMLOG Model

Sociologist Robert Bales’s (1970, 1980, 1999) SYMLOG model provides a final example of a comprehensive explanation for the types of roles commonly observed in groups. Bales (1950), in his earliest studies of groups, distinguished between two basic categories of roles—task and relationship roles—and based his IPA system on this distinction. In time, however, he came to recognize the critical importance of two other aspects of roles and intermember relations: dominance and friendliness. So, he developed a new, more complex model—Systematic Multiple Level Observation of Groups (SYMLOG)—based on the following three dimensions: Dominance or submissiveness: Is this member active, outgoing, and talkative or passive, quiet, and introverted? (dominance is Up, submissive is Down.)

Friendliness or unfriendliness: Is this member warm, open, and positive or negative and irritable? (friendliness is Positive, unfriendliness is Negative.)

Acceptance or nonacceptance of task-oriented authority: Is this member analytic and task-oriented or emotional, nontraditional, and (in some cases) resentful? (acceptance of the task-orientation of established authority is Forward, nonacceptance is Backward.)

Observers, or the group members themselves, can rate each individual in the group using the 26 categories shown in Table 6.2. The group leader’s behaviors, for example, might be concentrated in the “active, dominant, talks a lot” category rather than the “passive, introverted, says little” category. A disillusioned group member, in contrast, might get high scores for “irritable, cynical, won’t cooperate.” These ratings can be used to chart the flow of a group’s interaction over time. When a group first begins to discuss a problem, most of the behaviors may be concentrated in the dominant, friendly, and accepting authority categories. But if the group is wracked by disagreement, then scores in the unfriendly, nonaccepting authority categories may begin to climb. SYMLOG can also be used to create a graph of the group profile based on dominance, friendliness, and authority dimensions (Hare & Hare, 2005; Isenberg & Ennis, 1981; Polley, 1989).

Although SYMLOG ratings were never completed for the Andes group, Figure 6.2 is a speculative mapping of the group’s structure into the three dimensions in Bales’ model. The vertical axis corresponds to the role-related behavior in the group. People like Fito Strauch and Fernandez rank near the task-oriented, accepting of authority end of this dimension, whereas Harley and Mangino are located near the opposing authority end of this dimension because they tended to resist group pressures and to express their feelings and emotions within the group. The horizontal axis pertains to attraction relations among the members. Parrado and Turcatti, for example, occupy positions at the friendly end of this dimension because they were both very popular within the group, whereas Delgado’s and Canessa’s low social standing places them at the unfriendly end. Bales uses circles of varying size to illustrate the third structural dimension: dominance/submission. The larger the circle, the greater the group member’s status in the group; hence, Fito Strauch is represented by a very large circle, whereas Harley (one of the malingerers) is represented by a very small circle.

Possible locations of a subset of the Andes group members in the three-dimensional space described by the SYMLOG rating system.

SYMLOG, by taking into account role, status, and attraction, yields an integrative and in-depth picture of the organization of groups (Hare et al., 2005). The task-oriented acceptance of authority/nonacceptance of authority dimension focuses on role structure, but distinguishes between roles that are higher and lower in status, and ones that exert a positive or negative influence on the group and its processes. Thus, SYMLOG is a powerful conceptual and methodological tool that provides a clearer understanding of the unseen group structures that underlie recurring patterns of interpersonal behaviors in groups.

6-2c. Group Socialization

An actor answering a casting call may hope to land the lead role of Juliet, but the director may instead offer her only a smaller part, such as the role of the nurse or Lady Capulet. She may decide that the role is too insubstantial for her talents and not accept it, or she may decide that any role in the production is better than no role at all. Similarly, individuals often seek particular roles in groups, but the group may not permit them to occupy these roles. In the Andes group, for example, many sought to be one of the “expeditionaries”—explorers who were selected to hike away from the crash site and seek help. But only three were chosen.

Group Socialization Theory

Richard Moreland and John Levine (1982) developed their theory of group socialization to explain how individuals negotiate their role assignments in groups. Their theory, which is summarized in Figure 6.3, recognizes that individuals are often asked to take on roles that they would prefer to avoid. Newcomers must “learn their place” in the group and acquire the behaviors required by the roles to which they have been assigned. Veteran group members must, in some cases, be ready to take on new roles within the group that force them to learn new skills and seek new challenges. But group members also feel that their groups should be flexible enough to change to meet their particular needs. So, individuals attempt to influence the group. Hence, group socialization is a mutual process: Through assimilation, the individual accepts the group’s norms, values, and perspectives, and through accommodation, the group adapts to fit the newcomer’s needs.

The Moreland and Levine theory of group socialization. The model identifies five types of roles (top of the figure), five stages and processes of socialization (bottom of the figure), and four transition points (identified as stars on the curve). The curved line represents the gradual increase (and eventual decrease) of a hypothetical member’s commitment to the group. Commitment increases as the member moves from prospective member to new member to full member, but then declines as the member moves to the role of marginal member and finally to ex-member.

Moreland and Levine’s theory distinguishes between five classes of roles—prospective member, new member, full member, marginal member, and ex-member. Prior to actually joining a group, individuals may study the group and the resources it offers, and part of this reconnaissance involves identifying the type of role they will be given should they join. The group, in contrast, seeks to recruit new members, often by promising them desirable roles and responsibilities (Kramer, 1998). Should the individuals choose to enter the group (entry), their commitment to the group increases, and their socialization by the full members begins in earnest (Tan, 2015). To the full members, the newcomers are inexperienced and cannot be completely trusted until they accept the group’s norms and role allocations.

The Newcomer Role

The role of newcomer can be a stressful one (Moreland & Levine, 2002). New to the group and its procedures, newcomers lack basic information about their place in the group and their responsibilities. Although the passage of time will eventually transform them into rank-and-file members, newcomers often prolong their assimilation into the group by remaining cautiously aloof or by misinterpreting other members’ reactions. Moreland (1985), to study this process, led some members of a newly formed group to think that they were newcomers surrounded by more senior members. He arranged for groups of five unacquainted individuals to meet for several weeks to discuss various topics. He told two of the five that the group had been meeting for some time and that they were the only newcomers. Although the role of newcomer existed only in the minds of these two participants, the people who thought themselves newcomers behaved differently from the others. They interacted more frequently and more positively with each other, they were less satisfied with the group discussion, and their descriptions of the group made reference to members’ seniority. Thus, the belief that one is a newcomer who will be treated differently by the old-timers can act as a self-fulfilling prophecy: Just thinking of oneself as a newcomer caused people to act in ways that isolated them from the rest of the group (Major et al., 1995). This “mistreatment,” which they themselves partially caused, may undermine their loyalty to the group (Levine, Moreland, & Choi, 2001).

Role Transitions

The socialization process does not end when individuals become full-fledged group members. Even seasoned group members must adjust as the group adds new members, adopts new goals in place of its old objectives, or modifies status and role relationships. Much of this maintenance phase is devoted to role negotiation. The group may, for example, require the services of a leader who can organize the group’s activities and motivate members. The individual, in contrast, may wish instead to remain a follower who is responsible for relatively routine matters. During this phase, the group and the individual negotiate the nature and quantity of the member’s expected contribution to the group.

Many group members remain in the maintenance period until their membership in the group reaches a scheduled conclusion. An employee who retires, a student who graduates from college, or an elected official whose term in office expires all leave the group after months or years of successful maintenance. In some cases, however, the maintenance process builds to a transition point that Moreland and Levine labeled divergence. The group may, for example, force individuals to take on roles that they do not find personally rewarding. Individuals, too, may fail to meet the group’s expectations concerning appropriate behavior, and role negotiation may reach an impasse.

Resocialization

When the divergence point is reached, the socialization process enters a new phase—resocialization. During resocialization, the former full member takes on the role of a marginal member, whose future in the group is uncertain. The individual sometimes precipitates this crisis point, often in response to increased costs and dwindling rewards, waning commitment to the group, and dissatisfaction with responsibilities and duties. The group, too, can be the instigator, reacting to a group member who is not contributing or is working against the group’s explicit and implicit purposes. Moreland and Levine identified two possible outcomes of resocialization. The group and the individual, through accommodation and assimilation, can resolve their differences. In this instance, convergence occurs, and the individual once more becomes a full member of the group. Alternatively, resocialization efforts can fail (see Figure 6.3). The group may conclude that the individual is no longer acceptable as a member and move to expel him or her. Similarly, the individual may reevaluate his or her commitment to the group and decide to leave. As a result, the divergence between the group and the individual becomes so great that a final role transition is reached: exit.

6-2d. Role Stress

Roles influence group members’ happiness and well-being in significant ways. Some roles are more satisfying than others; people prefer to occupy roles that are prestigious and significant rather than roles that are menial and unimportant. They also like roles that require specialized skills and talents more than unchallenging, uninvolving roles (Rentsch & Steel, 1998). The demands of a role can also be stressful for the occupants of that role. A player on a sports team, for example, may be called the spark plug, the comedian, or the mentor, but what are the duties associated with such an amorphous role (Cope et al., 2011)? And what if the leader of a group believes her role involves keeping members on track, but others in the group think that the leader’s role is to provide emotional support, encouragement, or advice (Junker & van Dick, 2014). When a role is ambiguously defined, internally inconsistent, or fits the occupant poorly, roles can be great challenges for group members (Kahn et al., 1964).

Role Ambiguity

The responsibilities and activities that are required of a person who occupies a role are not always clear either to the occupant of the role (the role enactor or the role taker) or to the rest of the group (the role senders). Even when a role has a long history in the group or the group deliberately creates the role for some specific purpose, the responsibilities of the role may be ill-defined. In such cases, role takers will likely experience role ambiguity—they wonder if they are acting appropriately, they perform behaviors that others in the group should be carrying out, and they question their ability to fulfill their responsibilities.

Role Conflict

In some instances, group members may find themselves occupying several roles at the same time with the requirements of each role making demands on their time and abilities. If the multiple activities required by one role mesh with those required by the other, role takers experience few problems. If, however, the expectations that define the appropriate activities associated with these roles are incompatible, role conflict may occur (Brief, Schuler, & Van Sell, 1981).

Interrole conflict develops when role takers discover that the behaviors associated with one of their roles are incompatible with those associated with another of their roles. When assembly line workers are promoted to managerial positions, for example, they often feel torn between the demands of their new supervisory role and their former roles as friend and workmate. Similarly, college students often find that their student role conflicts with other roles they occupy, such as spouse, parent, or employee. If the student role requires spending every free moment in the library studying for exams, other roles will be neglected.

Intrarole conflict results from contradictory demands within a single role. A supervisor in a factory, for example, may be held responsible for overseeing the quality of production, training new personnel, and providing feedback or goal-orienting information. At another level, however, supervisors become the supervised because they take directions from a higher level of management. Thus, the members of the team expect the manager to keep their secrets and support them in any disputes with the management, but the upper echelon expects obedience and loyalty (Katz & Kahn, 1978).

Role conflict also arises when role takers and role senders have different expectations. The newly appointed supervisor may assume that leadership means giving orders, maintaining strict supervision, and criticizing incompetence. The work group, however, may feel that leadership entails eliciting cooperation in the group, providing support and guidance, and delivering rewards. Because the demands of roles vary from one country to another, in multicultural groups, individuals often misunderstand what their roles require of them.

Person–Role Conflict

Sometimes, the behaviors associated with a particular role are completely congruent with the basic values, attitudes, personality, needs, or preferences of the person who must enact the role: A stickler for organization is asked to be in charge of organizing the group’s records; a relationship expert must take on a role that requires sensitivity and warmth. In other cases, though, role fit is poor. An easygoing, warm person must give performance appraisals to the unit’s employees. An individual with high ethical standards is asked to look the other way when the company uses illegal accounting practices.

When role fit is low, people do not feel that they can “be themselves” in their roles; they also question their capacity to enact the role’s demands competently (Talley et al., 2012). In one study, college students who held roles in campus groups were asked if they felt that their role “reflected their authentic self and how much they felt free and choiceful as they fulfilled their role” (Bettencourt & Sheldon, 2001, p. 1136). Those who felt more authentic when enacting their role reported more positive mood, less negative mood, and a higher level of satisfaction with life overall. Feeling competent when enacting one’s role was also a powerful predictor of well-being. In another study, students first rated themselves on 20 different traits (e.g., cooperative, outgoing, and imaginative). Later in the semester, they were given a list of five discussion roles (idea person, devil’s advocate, moderator, secretary, and announcer) and then asked to indicate how valuable these 20 traits were for enacting each role. For example, how important is it for the idea person to be cooperative? Outgoing? Imaginative? Then they were assigned to one of these roles in a class discussion. As the concept of role fit suggests, individuals assigned to roles that required the kinds of characteristics that they believed they possessed felt more authentic, and their moods were more positive (Bettencourt & Sheldon, 2001).

Roles and Well-Being

Uncertainty about one’s role, including role ambiguity, role conflict, and poor role fit, results in stress and tension, and the results are rarely positive for the group member or for the group itself. In study after study, increases in role ambiguity, role conflict, and person–role conflict are associated with a host of negative psychological and interpersonal outcomes, including heightened levels of tension, employee turnover, absenteeism, interpersonal conflict within the group, and declines in job satisfaction and performance quality (Ritter et al., 2016). Individuals who experience conflicts between the work roles and their family roles experience declines in family satisfaction and, to a lesser extent, their work satisfaction (Ford, Heinen, & Langkamer, 2007). Role stress is also associated with physical well-being. One meta-analytic review of 79 studies of the relationship between role stress and physical maladies concluded that role conflict predicts backaches, sleep disturbances, dizziness, and gastrointestinal problems, whereas role ambiguity predicted fatigue. Interpersonal conflict—not getting along with people at work—predicted increases in all of these negative physical symptoms (Nixon et al., 2011).

Was It You or Role Conflict?

Have you ever held a position in a group that you never managed to master? That despite your best efforts, the final result never quite met the group’s standards? Your experience in the group may have been due, in part, to inadequacies in the group’s structure rather than your own insufficiencies. Review that disappointing group experience using the items listed below.

Instructions: Circle the number of any items with which you agree.

Too often it was unclear what I was supposed to do for this group.

The various roles I took on in this group did not mesh well with each other.

The role I had in the group pulled me in too many different directions.

I was expected to do things that I was uncomfortable doing.

I often wondered “Who is supposed to do what?” in this group.

If I did what one of my roles required I could not take care of my other duties.

I had a mishmash of mismatched responsibilities in this group.

I could not be myself when I was in this group.

I did not know what I was supposed to do when I was in this group.

I had taken on too many commitments to too many groups.

People in my group misunderstood what I was supposed to do for this group.

I too frequently had to do things in this group that I did not agree with.

Scoring: Add up the number of items you checked for each type of role-related frustration—role ambiguity: Items 1, 5, and 9; interrole conflict: Items 2, 6, and 10; intrarole conflict: Items 3, 7, and 11; person–role conflict: Items 4, 8, and 12. A score of 2 or more in any category indicates that you may have experienced that form of role-related stress as a member of the group.

What can groups and organizations do to help their employees cope with role stress? One solution involves making role requirements explicit: Managers should write job descriptions for each role within the organization and provide employees with feedback about the behaviors expected of them (Schmidt et al., 2014). The workplace can also be designed so that potentially incompatible roles are performed in different locations and at different times. In such cases, however, the individual must be careful to engage in behaviors appropriate to the specific role, because slipping into the wrong role at the wrong time can lead to both embarrassment and a loss of coordination within the group (Goffman, 1959). Some companies, too, develop explicit guidelines regarding when one role should be sacrificed so that another can be enacted, or they may prevent employees from occupying positions that can create role conflict. Managers and the leaders of groups should also be mindful of the characteristics of the members of their groups and be careful to maximize role fit when selecting members for particular tasks. Work settings should also do what they can to maximize the salutary effects of positive, supportive member relationships. Antagonistic coworkers increase role stress, but people who work with others who are supportive and helpful experience fewer of the negative consequences of role stress, particularly when they work in settings that are more socially focused (Chiaburu & Harrison, 2008).

6-3. Intermember Relations

On the 17th day of their ordeal, an avalanche swept down on the Andes survivors as they slept, filling their makeshift shelter with snow. Many were killed, and soon a new order emerged in the group. Three young men stepped forward to take over control of the group. They were cousins, and their kinship bonds securely connected them to one another, but they also were friends with many of the remaining group members.

Connections among the members of a group provide the basis for the third component of group structure—the network of intermember relations. Which one of the three cousins had the most authority? Who in a group is most liked by others, and who is an isolate? How does information flow through a group from one person to the next? The answers depend on the group’s status, attraction, and communication networks.

6-3a. Status Relations

The roles that emerged in the Andes group following the crash defined who would lead, explore, and care for the injured. The individuals who took on these roles, however, were not equal in terms of authority in the group. In the Andes group, Fito Strauch, Eduardo Strauch, and Fernandez formed a coalition that controlled most of the group’s activities (see Figure 6.4). Below this top level was a second stratum of members who had less power than the leaders but more prestige than the occupants of lower echelons. The explorers (“expeditionaries”) occupied a special niche. These individuals had been chosen to hike down the mountain in search of help. In preparing for their journey, they were given special privileges, including better sleeping arrangements and more clothing, food, and water. The rank-and-file members included the youngest men in the group, the injured, and those thought to be malingering.

Status Differentiation

These stable variations in members’ relative status have many names—authority, power, status network, pecking orders, chain of command, or prestige ranking—but whatever their label they result in elevated authority for some and less for others. Group members may start off on an equal footing, but over time status differentiation takes place: Certain individuals are granted, or they acquire, more authority than others. Although the effects of status vary from one group to the next, in general those with higher status are afforded more attention by others in the group, they are held in higher regard, and they are more influential; they exert more control over the group’s processes (Anderson et al., 2001).

Status differentiation turns groups with flat, undifferentiated structures into ones with centralized, hierarchical structures, rather like the pyramid-shaped organizational charts of businesses and military organizations. This pattern is ubiquitous across groups, organizations, and cultures, and it will develop quickly in both informal, socioemotional groups and in those that are more task-focused (Tiedens & Fragale, 2003). Furthermore, even though individuals often express a preference for more egalitarian structures where each person is equal to every other person in terms of influence and control of resources, people are generally more comfortable when they are members of hierarchical groups (Zitek & Tiedens, 2012).

Status Rank

Who gains status in a group is the complex result of individual, group, and cultural factors (Piazza & Castellucci, 2014). Some individuals, given their personalities, skills, behavioral tendencies, and levels of experience, are more readily granted authority than others. Extraverts, for example, tend to more quickly gain status in groups, relative to introverts, as do individuals who are particularly skilled at the kinds of tasks the group must attempt (Anderson et al., 2001). But a person’s rise to a position of status in a group cannot be predicted solely on the basis of their individual qualities, for different groups value different attributes. The qualities that earn a person status in a boardroom differ from those that predict one’s prestige in a rugby team or a biker gang. Thus, predictions of status must take into account the degree to which individuals’ attributes match the qualities valued by the groups to which they belong: the person-group fit. In another group, Delgado might have been highly influential, for he was quite articulate and socially skilled. In the Andes group, however, the fit between his personal qualities and the group was poor (see, too, Chapter 8’s analysis of the sources of social power in groups).

6-3b. Attraction Relations

Some of the Andes survivors rose to positions of authority, while others remained relatively powerless. Yet, to describe the group in just these terms would be to miss a vital part of the group’s structure. The individuals were not just leaders and followers but also friends and enemies. This network of likes and dislikes among the members is called by many names, including attraction relations, social status, or sociometric structure.

Sociometric Differentiation

Just as status differentiation results in variations in status, so, too, sociometric differentiation results in a stable ordering of members from least liked to most liked. Consider, for example, the relationships among the rank-and-file group members and the four designated explorers in the Andes group, Turcatti, Parrado, Vizintin, and Canessa. Nearly everyone admired Turcatti and Parrado; their warmth, optimism, and physical strength buoyed the sagging spirits of the others. Vizintin and Canessa, in contrast, “did not inspire the same affection” (Read, 1974, p. 141). They liked each other but had few other friends within the group. Mangino, one of the younger men, was an exception; he liked them both. Most of the others, however, quarreled with them constantly (see Figure 6.5).

Patterns of attraction and friendship in the Andes survivors groups. Each member is represented as a circle, and the lines (or edges) connecting each person to other individuals indicate who is linked to whom. (The relationships are for illustration only; social network data were not collected for this group.)

Attraction patterns like those in the Andes group are not a disorganized jumble of likes and dislikes but a network of stable social relationships. Popular individuals (stars) receive the most positive sociometric nominations within the group; rejected group members (outcasts) get picked on the most when group members identify whom they dislike; neglected isolates receive few nominations of any kind; and the average members are liked by several others in the group. In the Andes group, for example, Parrado was admired by all; he was, sociometrically, the star of the group. Delgado, in contrast, was the group’s outcast; he had few friends in the group, and the young men ridiculed him constantly for not doing his share of the work.

Like most groups, the Andes survivors’ attraction relations showed signs of reciprocity, transitivity, and homophily. Vizintin liked Canessa and Canessa liked Vizintin in return. Such reciprocity, as noted in Chapter 4, is a powerful tendency in most settings; it has been documented repeatedly in a variety of groups, including football teams, police squads, psychotherapy groups, sororities, and classroom groups (Kandel, 1978; Newcomb, 1979; Wright, Ingraham, & Blackmer, 1984). Exceptions to reciprocity sometimes occur, and some forms of attraction tend to be less reciprocal than other forms of attraction, but these exceptions to the reciprocity principle are relatively rare (Segal, 1979). The Andes group also showed signs of network transitivity: Canessa liked Mangino, Mangino liked Vizintin, and in confirmation of transitivity, Canessa liked Vizintin (A likes B, B likes C, so A likes C).

Clusters, or cliques, also existed in the Andes group, for Vizintin, Canessa, and Mangino formed a unified coalition within the larger group. Others rarely hesitated to show their disdain for the members of this subgroup, but these three were joined by strong bonds of attraction. Subgroups, in general, display homophily: members of cliques tend to be more similar to one another than they are to the members of the total group. Members of the same racial category, for example, may join to form a coalition, or the group may separate naturally into all-male and all-female cliques (Lau & Murnighan, 1998). Group members also often deliberately form and manipulate cliques within larger groups by systematically including some individuals and excluding others (Adler & Adler, 1995).

Is It Better to Be Liked by Many or to Be Well Liked by Just a Few?

As groups increase in size, subgroups—cliques—often form, creating groups within groups, and these cliques can substantially influence the cohesiveness of the overall group. Sociologists Pamela Paxton and James Moody (2003), for example, searched for and discovered cliques in their analysis of a particularly cohesive group: a sorority in a university located in the southern United States that they gave the fictitious name of Alpha Beta Chi, or ABX. ABX appeared to be a highly cohesive group with strong relations among all members, but their analysis revealed the existence of four cliques within the overall group, which Paxton and Moody labeled the Separatists, the Middles, the Random Chapter Members, and the Small Clique. The Separatists, for example, were relatively isolated from the other members of the sorority but they were closely linked to one another. The Middles, in contrast, had more ties to people outside of their clique. Also, some women in the sorority were sociometric stars—liked by many—but some were rejected members with only a single tie keeping them connected to the group (“hangers-on”).

Paxton and Moody discovered a member’s commitment to her sorority could be predicted by studying her place in the group’s attraction network. Women in the Middles, for example, had a stronger sense of belonging to the group, particularly in comparison to the Separatists. But it was the women who had close ties to other well-liked women who were the most satisfied with their group—even more so than women with higher overall popularity scores, as indexed by how many times a woman was picked as a friend by others. Also, within any particular clique, those women with more central locations within the clique tended to be less committed to their sorority as a whole. Devoting one’s relational energies to a small subset of the group may leave little time for the interpersonal work required to maintain good relations with the entire group.

Balance Theory

Why do most groups’ attraction relations tend to be reciprocal, transitive, and homophilous? According to Fritz Heider’s (1958) balance theory, some patterns of relationships in groups are more structurally sound, or balanced, than others, so groups naturally tend to gravitate toward these rather than toward unbalanced states. In the Andes survivors, the triad of Vizintin, Canessa, and Mangino was balanced: everyone in it liked one another, so all bonds were positive. What would happen, however, if Mangino came to dislike Canessa? According to Heider, this group would be unbalanced. Such a group pattern is considered so unstable that it has been given the ominous name “the forbidden triad” (Granovetter, 1973). In general, a group is balanced if

(1)

all the relationships are positive, or

(2)

an even number of negative relationships occurs in the group.

Conversely, groups are unbalanced if they contain an odd number of negative relations (Newcomb, 1963, 1981).

Because unbalanced sociometric structures generate tension among group members, people are motivated to correct the imbalance and restore the group’s equilibrium. Heider noted that this restoration of balance can be achieved either through psychological changes in the individual members or through interpersonal changes in the group. If Mangino initially likes only Vizintin and not Canessa, he may change his attitude toward Canessa when he recognizes the strong bond between Vizintin and Canessa. Alternatively, group members who are disliked by the other group members may be ostracized, as in the case of Delgado (Taylor, 1970). Finally, because the occurrence of a single negative relationship within a group can cause the entire group to become unbalanced, large groups tend to include a number of smaller, better balanced cliques (Newcomb, 1981). The Andes group, for example, was somewhat unbalanced overall, but its subgroups tended to be very harmonious. As a result, the group was high in cohesiveness (Rambaran et al., 2015).

6-3c. Communication Relations

In the Andes group, the three leaders stayed in close communication, discussing any problems among themselves before relaying their interpretations to the other group members. The other members usually routed all information to the threesome, who then informed the rest of the group. In contrast, the injured members were virtually cut off from communication with the others during the day, and they occasionally complained that they were the last to know of any significant developments. These regular patterns of information exchange among members of a group are called communication networks.

Patterns of communication among group members, like other structural features of groups, are sometimes deliberately set in place when the group is organized. Many companies, for example, adopt a centralized, hierarchical communication network that prescribes how information is passed up to superiors, down to subordinates, and horizontally to one’s equals. Even when no formal attempt is made to organize communication, an informal communication network will usually take shape over time. Moreover, this network tends to parallel status and attraction patterns. Take the Andes group as a case in point: Individuals who occupied high-status roles—the explorers, the food preparers, and the lieutenants—communicated at much higher rates and with more individuals than individuals who occupied the malingerer and injured roles (Shelly et al., 1999).

Communication networks become more complex and varied as groups increase in size, but some of their basic forms are graphed in Figure 6.6. In a wheel network, for example, most group members communicate with just one person. In a comcon, all members can and do communicate with all other members. In a chain, communication flows from one person to the next in a line. A circle is a closed chain, and a pinwheel is a circle where information flows in only one direction (Shaw, 1964).

Examples of common communication networks in small groups. These networks are a sample of the various kinds of communication networks that can be created by opening and closing lines of communication among members. In most of these examples, the lines are undirected ones, with information flowing back and forth between members. Only the pinwheel has directed, one-way communication links. The Y, Kite, and Wheel are centralized networks; the others are decentralized.

Centrality is a particularly important feature of communication networks. With centralized networks, one of the positions in the group has a very high degree of centrality—it is located at the crossroads (the hub) of communications—relative to the other positions in the group (e.g., the wheel, the kite, or the Y in Figure 6.6). Groups with this type of structure tend to use the hub position as the data-processing center, and its occupant typically collects information, synthesizes it, and then sends it back to others. In decentralized structures, like the circle or comcon, the number of channels at each position is roughly equal, so no one position is more “central” than another. These groups tend to use a variety of organizational structures when solving their problems, including the so-called each to all pattern, in which everyone sends messages in all directions until someone gets the correct answer (Shaw, 1964, 1978).

Network Centralization and Performance

Early studies of communication networks suggested that groups with centralized networks outperformed decentralized networks (Bavelas, 1948, 1950; Bavelas & Barrett, 1951; Leavitt, 1951). A group with a wheel structure, for example, took less time to solve problems, sent fewer messages, detected and corrected more errors, and improved more with practice than a group with a decentralized structure, such as a circle or comcon (Shaw, 1964, 1978). The only exceptions occurred when the groups were working on complicated tasks such as arithmetic, sentence construction, problem solving, and discussions. When the task was more complex, the decentralized networks outperformed the centralized ones.

These results led social psychologist Marvin E. Shaw to propose that network efficiency is related to information saturation. When a group is working on a problem, exchanging information, and making a decision, the central position in the network can best manage the inputs and interactions of the group. As work progresses and the number of communications being routed through the central member increases; however, a saturation point can be reached at which the individual can no longer efficiently monitor, collate, or route incoming and outgoing messages. Shaw noted that saturation can occur in a decentralized network, but it becomes more likely when a group with a centralized structure is working on complex problems. Because the “greater the saturation the less efficient the group’s performance” (Shaw, 1964, p. 126), when the task is simple, centralized networks are more efficient than decentralized networks; when the task is complex, decentralized networks are superior. In consequence, groups tend to gravitate naturally to more decentralized network structures when the tasks they must accomplish become more complex and multifaceted (Brown & Miller, 2000).

These different types of centrality also influence role allocations, overall commitment, and satisfaction with membership in the group (Krackhardt & Porter, 1986; Lovaglia & Houser, 1996). Individuals who occupy centralized positions in centralized networks, such as a wheel or a Y (see Figure 6.6), are nearly always thought to be the leader of their group, even when they are randomly assigned to this position (Leavitt, 1951). In studies of employees in work groups, those who are more central in their network are less likely to quit than are employees at the periphery of the company’s communication network (Feeley, 2000). Peripheral members are also more likely to quit in clumps. Because individuals in decentralized positions are connected to very few of the other members, when one peripheral member leaves the group, the individuals located near that person in the network also tend to leave the group (Krackhardt & Porter, 1986). Finally, centralized networks, by definition, have fewer centralized positions than decentralized positions. In consequence, the overall level of satisfaction in a centralized group is almost always lower than the level of satisfaction in a decentralized group (Shaw, 1964).

Directional (Up–Down) Effects

Only small groups with decentralized communication networks outperform groups with centralized networks. Once the group becomes too large, members can no longer keep up with the high rate and quantity of information they are receiving. Therefore, most organizations manage information flow by adopting hierarchical communication networks (Goetsch & McFarland, 1980). In such networks, information can pass either horizontally between members on the same rung of the communication ladder or vertically up and down from followers to leaders and back (Jablin, 1979).

Upward communications tend to be very different from downward communications (Sias, Krone, & Jablin, 2002). Downward-flowing information moves from the leaders to the followers of the group, and generally includes explanations of actions to be taken, the reasons for actions, suggestions to act in a certain manner, and feedback concerning performance. In some cases, too, up–down messages are urgent, sent using more immediate channels of communication, such as e mail, rather than face-to-face meetings (Byrne & LeMay, 2006). Upward communications from subordinates to superiors, in contrast, include information on performance, insinuations about a peer’s performance, requests for information, expressions of distrust, factual information, or grievances concerning the group’s policies. These upward communications, moreover, tend to be fewer in number, briefer, and more guarded than downward communications. In larger organizations, the upward flow of information may be much impeded by the mechanics of the transfer process and by the low-status members’ reluctance to send information that might reflect unfavorably on their performance, abilities, and skills (Sias, 2009). This reticence of low-status members means that good news travels quickly up the hierarchy, whereas the top of the ladder will be the last to learn bad news.

6-4. Application: Social Network Analysis

In the fall of 1932, the staff of the Hudson School for Girls asked psychiatrist Jacob Moreno to help them solve a problem. In the preceding two weeks no fewer than 14 residents had run away from the school. The girls were housed in various residences across the facility, and the staff could not identify why these girls, at this time, decided to leave Hudson. Moreno could, however. He discovered something hidden beneath the surface of the groups at the Hudson School that the staff had overlooked. Using the sociometric methods that he pioneered, he discovered that the girls who ran away were joined together in an unnoticed web of social connections and that these connections facilitated the transmission of influence and ideas—some of which included the notion of leaving Hudson School behind (Moreno, 1934; see Chapter 2).

Social network analysis (SNA) is particularly useful in making what is often unseen and unnoticed evident. Dyadic relations—status differences, likes and dislikes, and patterns of communication—may be clearly known by some in the group, but often only a SNA will reveal the actual patterns and processes that sustain these relationships.

6-4a. Mapping Social Networks

Social network analysts are the geographers of the human terrain. They seek to map the connections that link individuals to one another and use that information to determine precisely where people are located relative to each other in interpersonal space. This approach dates back to some of the earliest work in sociology and psychology, for these fields’ founders all sought to make social relations tangible (Borgatti et al., 2009). These efforts, which included Moreno’s (1934) sociometric studies of attraction in groups, experimental studies carried out at the Group Networks Laboratory at the Massachusetts Institute of Technology (e.g., Bavelas, 1948; Leavitt, 1951), and studies of community-level patterns of informational influence (e.g., Granovetter, 1973), laid the foundation for social network analysis: A set of procedures defined by

(a)

a focus on the structures of social groups and on linkages among group members in particular;

(b)

the systematic measurement of these structures;

(c)

the use of graphics to represent these structures; and

(d)

the application of statistical and mathematic procedures to quantify these structures (Freeman, 2004).

The attraction patterns of the Andes survivors in Figure 6.5 illustrate an application of SNA. Each member, or node, is represented as a circle, and the lines, or edges, connecting nodes indicate who is linked to whom. The arrows indicate the direction of the relationship. An edge with a single arrow indicates the relationship is a directed one, linking the sender to the receiver. For example, the links between Zerbino and Fito Strauch, Paez, and Delgado go out from Zerbino and are received by Strauch, Paez, and Delgado. An edge with arrows at both ends indicates a symmetric, reciprocal relationship (for example, Zerbino and Fernandez). Distance, in social networks, is defined by relationships rather than physical distance. As noted in Chapter 3, two people who are directly linked to one another, such as Zerbino and Fito Strauch, are separated by a distance of 1: one degree of separation. But it would take four steps or links for Zerbino to reach Vizintin—Zerbino → Fito Strauch → Turcatti → Canessa → Vizintin—hence, four degrees of separation.

Individuals in Networks

SNA is a multilevel method. It yields information about each member of the network—the egocentric network—as well as insights into the group as a whole—the sociocentric network. Starting with the individual-level indexes, SNA describes each member’s location in the network relative to the others. Where, for example, were the cousins, Fito and Eduardo Strauch, located in the network? Were any members located on the fringes or isolated from the group altogether? Who was in a position to communicate most easily with the most members of the group? SNA answers these questions by calculating and organizing relational information about all the members, including how central they are in the group, how many people link to them, and their location relative to other people in the network. Some key indexes are centrality, betweenness, and closeness.

Degree centrality is the number of connections or ties to a node. Fito Strauch, for example, was connected by 10 ties to others and to him, whereas Delgado was linked to only two others.

Outdegree and indegree centrality can be calculated when ties are directed rather than undirected. Outdegree is the number of links directed out from the node, whereas indegree is the number of links directed in. Zerbino’s outdegree centrality, for example, is four, since he connects out to Delgado, Paez, Fito, and Fernandez, but his indegree centrality is only one because only Fernandez directs a relationship to Zerbino. Outdegree and indegree centrality are equivalent when the relationships linking members are undirected, such as the flow of communication in a back-and-forth conversation or friends on Facebook (Wasserman & Faust, 1994; see Borgatti, 2005, for more information about centrality indexes).

Betweenness takes into account ties to more distant actors in the network (Freeman, 1979). A position with a high degree of betweenness is one that is located between many of the other individuals in the network. Turcatti, for example, has a much lower degree centrality than Fito Strauch, but higher betweenness since he joins the subgroup of Canessa, Vizintin, and Mangino to the rest of the group. An individual in such a position often acts as the go-between or gatekeeper, linking people in the network who could otherwise not contact one another.

Closeness is determined by the distance to all other members of the group. Fito Strauch, for example, can reach all other members through short paths, whereas other group members (such as Delgado or Mangino) are separated from others by greater distances.

Groups as Networks

Unlike egocentric indexes that yield a value for every individual in the network, sociocentric, or group-level, network indexes describe the entire network—or, at least a portion of it. Some common group-level features of networks include size, density, cliques, and holes.

Density is determined by how many people are linked to one another out of the total possible number of links. The density of the Andes group, for example, would be 1.0—the maximium—if every one of its 13 members was linked to every other member: 156 ties in all. (The formula for calculating the number of possible ties in any group, mentioned in Chapter 1, is n(n–1) if relationships are directed and n(n–1)/2 if the relationships are not directed.) However, for this group, density is much lower than 1.0 because many members are only linked to one or two others and not to all 12—in fact, only 29 ties are present in this group. Therefore, the density of the group is .19 (29/156).

Cliques, or clusters, of subgroups often form in larger networks. In the Andes group, Vizintin, Canessa, and Mangino formed a unified coalition within the larger group based on friendship. Others rarely hesitated to show their disdain for the members of this subgroup, but these three were joined by strong bonds of attraction. Also, Fito Strauch, Eduardo Strauch, and Fernandez formed a second clique, in this case based on authority—these three formed a leadership triumvirate within the group.

Holes are “disconnections between nonredundant contacts in a network” (Burt, 1997, p. 339) or the gaps in a network that separate clusters or cliques. Holes may have a positive effect on group members if they buffer them from unwanted or too frequent contact with others, but they can also isolate members from the rest of the network. Turcatti spanned the hole in the Andes group.

6-4b. Applying Social Network Analysis

Who influences, likes, dislikes, trusts, admires, or talks to whom in any group? A simple question, but one that may not be easily answered with out first examining the group’s social network objectively. Even when members’ interdependencies are defined by the group’s organizational chart and the defined roles of each member, this explicit, formal structure may not correspond to the group’s informal actual structure. Consider, for example, the group of twelve individuals diagrammed in Figure 6.7a. Michael is the designated leader of this hypothetical group. He is responsible for three smaller teams within the group, and these teams are managed by Brandon, Ryan, and Jessica, respectively. At least, that is the way the group is intended to function. The group’s actual structure, however, is very different from its formal, mandated structure. Michael is the ostensible group leader, but Jessica (who is the group’s staff coordinator) holds the more centralized position within the group (see Figure 6.7b). Most of the group members trust Jessica—and seek her out for advice and information—rather than Michael.

The informal (actual) structure of a group does not always match the group’s mandated (formal) structure.

Groups with unrecognized oddities in their connections among their members are at risk: communication flow within the group could be disrupted, decisions might be made by the wrong people, and individuals may act in ways that are not consistent with their roles in the group. When, for example, Michael—the leader of the group charted in Figure 6.7b, wanted to introduce changes to the group’s procedures to enhance its performance, he first met with Brandon, Ryan, and Jessica to explain what he had in mind. Unfortunately, Jessica did not agree with the changes, and so the plan’s implementation did not go very smoothly. It was not until Michael, recognizing Jessica’s status in the group, met with her and adapted his plan so that it met with her approval, did the group adopt the changes he proposed (Krackhardt, 1996).

Social network analyst David Krackhardt, in his studies of the informal social networks in groups and organizations, documented group after group whose processes were disrupted and performance hobbled by members’ misunderstanding of the group’s social network. He concluded, “managers often pride themselves on understanding how these networks operate. They will readily tell you who confers on technical matters and who discusses office politics over lunch. What’s startling is how often they are wrong” (Krackhardt & Hanson, 1993, p. 104).

What’s the cure? Use social network analysis to gather, integrate, and evaluate the web of relationship ties that determine how the group gets it work done. A group’s dynamics cannot be understood if the relationships linking each member to one another and to the group are not understood. In group meetings, the opinions of those with higher status carry more weight than those of the rank-and-file members. When several members form a subgroup within the larger group, they exert more influence on the rest of the group. And when people manage to gain the top of the group’s status hierarchy, their influence over others also increases. When working in a group, its best to remain mindful of—and possibly openly discuss and clarify—the norms that guide members’ actions, the roles needed within the group, and the way the members are connected one to another.

Resources

Chapter Case: The Andes Survivors

Alive by Piers Paul Read (1974) is the best-selling account of the young men who crashed in the Andes and survived by creating a potent group.

Miracle in the Andes by Nando Parrado (2006), with Vince Rause, is a first-person account of the collective spirit of the rugby team. Parrado, the author, was one of the men who hiked down from the mountain to bring back help.

Norms

“Managing Normative Influences in Organizations” by Noah J. Goldstein and Robert B. Cialdini (2011) reviews the basic tenets of the focus theory of normative conduct and the intriguing empirical studies that support it.

Social Norms, edited by Michael Hechter and Karl-Dieter Op (2001), is a collection of solid theoretical and empirical reviews of the nature of norms and their influence in groups.

Roles

“Role Theory” by Bruce J. Biddle (2001) provides a concise summary of the history of role theory in the social sciences, as well as a review of current applications and trends.

Analysis of Social Interaction Systems, edited by A. Paul Hare, Endre Sjøvold, Herbert G. Baker, and Joseph Powers (2005), includes 26 chapters dealing with a variety of aspects of the SYMLOG method of group analysis with sections pertaining to leadership, organizational development, cross-cultural implications, and methodology.

Intermember Relations

“Network Analysis in the Social Sciences” by Stephen P. Borgatti, Ajay Mehra, Daniel J. Brass, and Giuseppe Labianca (2009) provides a concise, but comprehensive, overview of the uses of social network analysis in the social sciences, in general, and in the study of groups, in particular.

“Social Network Analysis in the Science of Groups: Cross-Sectional and Longitudinal Applications for Studying Intra- and Intergroup Behavior” by Ralf Wölfer, Nadira S. Faber, and Miles Hewstone (2015) provides an overview of the use of SNA methods with cross-sectional and longitudinal data.

Chapter Case: The Andes Survivors

Alive by Piers Paul Read (1974) is the best-selling account of the young men who crashed in the Andes and survived by creating a potent group.

Miracle in the Andes by Nando Parrado (2006), with Vince Rause, is a first-person account of the collective spirit of the rugby team. Parrado, the author, was one of the men who hiked down from the mountain to bring back help.

Norms

“Managing Normative Influences in Organizations” by Noah J. Goldstein and Robert B. Cialdini (2011) reviews the basic tenets of the focus theory of normative conduct and the intriguing empirical studies that support it.

Social Norms, edited by Michael Hechter and Karl-Dieter Op (2001), is a collection of solid theoretical and empirical reviews of the nature of norms and their influence in groups.

Roles

“Role Theory” by Bruce J. Biddle (2001) provides a concise summary of the history of role theory in the social sciences, as well as a review of current applications and trends.

Analysis of Social Interaction Systems, edited by A. Paul Hare, Endre Sjøvold, Herbert G. Baker, and Joseph Powers (2005), includes 26 chapters dealing with a variety of aspects of the SYMLOG method of group analysis with sections pertaining to leadership, organizational development, cross-cultural implications, and methodology.

Intermember Relations

“Network Analysis in the Social Sciences” by Stephen P. Borgatti, Ajay Mehra, Daniel J. Brass, and Giuseppe Labianca (2009) provides a concise, but comprehensive, overview of the uses of social network analysis in the social sciences, in general, and in the study of groups, in particular.

“Social Network Analysis in the Science of Groups: Cross-Sectional and Longitudinal Applications for Studying Intra- and Intergroup Behavior” by Ralf Wölfer, Nadira S. Faber, and Miles Hewstone (2015) provides an overview of the use of SNA methods with cross-sectional and longitudinal data.