Chapter 4

T he Political Psychology
of G roups

This chapter looks at Political Beings in their environment—that is, in the

presence of, and as a member of, groups. Groups have a prominent role in
politics. Small groups are often given the responsibility for making important
political decisions, creating political policies, and generally conducting pol­
itical business. Larger groups, such as the Senate, also hold a special place
in politics, and are responsible for larger-scale decisions and tasks, such
as passing legislation. Finally, large groups, such as states and countries,
carry with them their own dynamics, especially regarding how they view
each other and how they get along. Because so much political behavior is
performed by groups, it behooves us to learn more about the basic processes
that govern groups. Although groups are composed of individuals, group
behavior cannot be understood by studying individual behavior. Obviously,
understanding groups involves an understanding of the individuals who
compose a group, but there are dynamics of groups that cannot be observed
from examining individuals alone. Many observers (e.g., Durkheim, 1966;
LeBon, 1960) have noted that individuals often behave quite differently
when they are together compared with when they are alone. Consequently,
although the workings of the Political Being's mind are still operative, in this
chapter we are interested in the impact of the sociopolitical environment
on behavior.
The study of groups in social psychology has a short history, with some of
thefirststudiesbeingconductedjustbefore World War II (e.g., Lewin, Lippitt,
& White, 1939; Newcomb, 1943; Sherif, 1936; Whyte, 1943). Nonetheless, a
vast amount of information is available about group behavior, and most of
it can be applied to the study of groups in political settings. In this chapter,
we review a variety of information about groups. The first half of the chapter
focuses on the structural characteristics of groups, such as composition, for­
mation, and development. The second half of the chapter focuses on the
unique behaviors that take place in groups or because of groups, including
influence, performance, decision-making, and intergroup conflict.

DOI: 10.4324/9780429244643-4
 The Political Psychology of Groups 83

THE NATURE OF GROUPS

Definition of a Group
Imagine all the different types of collectives that exist in political settings.
People work together to solve problems, set political policies and agendas,
serve constituents, make legal decisions, run political campaigns, and
make decisions about world problems. Do all of these collectives constitute
groups? Group researchers were unable to answer that question. There is
litde consensus in the field about what characteristics of a collective make
a group. Although most social psychologists would agree that a group is a
collection of people who are perceived as belonging together and dependent
on one another, there are other ways to conceptualize groups. For example,
Moreland (1987) discussed “groupiness” or social integration as a quality
possessed by every collection of individuals to some degree. As the level of
social integration increases, people start to think and act more like a group
than a collection of individuals. Other social psychologists (Dasgupta, Banji,
& Abelson, 1999; Lickel et al., 2000) have pointed to entiativity, which refers
to the extent to which a collection of people is perceived as a coherent entity.
Some groups, such as people in line at a bank, are perceived as being low in
entiativity. Other groups, such as members of a family or members of a pro­
fessional sports team, are perceived as being high in entiativity.

Types of Groups
When thinking about groups, it is sometimes useful to consider the
various types. Several groups researchers provided typologies of groups.
For example, Prentice, Miller, and Lightdale (1994) investigated common-
bond and common-identity groups* Common-bond groups, such as social
groups, are based mostly on the attachments between group members. In
common-bond groups, the attachments to the group are based on such
things as member similarity, likability of fellow group members, and famil­
iarity with group members. Common-identity groups, on the other hand, are
based primarily on attachments to the group identity. Examples of common-
identity groups include music groups, sports teams, and performing groups.
Individuals in common-identity groups are far more attached to the group
identity than they are to individual members of the group.
Deaux, Reid,,Mizrahi, and Ethier (1995) provide another perspective on
types of groups. In research focusing on the dimensions of social identity,
they identified five types of social identities: personal relationships (e.g.,
friend, husband); vocations/avocations (e.g., student, gardener); political
affiliation (e.g., Democrat, Republican); stigmatized groups (e.g., alcoholic,
unemployed person); and ethnic/religious groups (e.g., Catholic, Hispanic).
Lickel et al. (2000) provided a more recent typology of groups. In
their study, participants were asked to rate a variety of groups on such
dimensions as perceived entiativity, interaction, common goals, similarity,
permeability, duration, size, and importance. The results of their investiga­
tion showed groups could be divided into three types. One type, intimacy
groups, consisted of small groups with frequent interactions, high similarity,
The Political Psychology of Groups

and importance to their members. Examples of intimacy groups include

families, friends, and fraternities. A second type, task groups, consists of
groups that are fairly small in size, but with high interaction, similarity,
and importance. Task groups are groups such as members of a jury, labor
unions, and student study groups. The third type of group, social categories,
consists of large groups that are usually low in interaction, importance, and
similarity of group members. These groups include women, Blacks, and
Jews, for example.
Why is it important to consider types of groups? One reason may be
the different functions groups serve or the types of needs they fulfill. For
example, Johnson etal. (2005) show that intimacy groups tend to fulfill needs
for affiliation, task groups tend to fulfill the need for achievement, and social
category groups are associated with the need for identity. Understanding
the relationship between group types and individual needs can be useful for
knowing what types of groups to look for to satisfy various social needs. In
addition, knowing what types of groups fulfill which needs might help to
explain dissatisfaction with a group. For example, if a person has a strong
need for affiliation but is not getting that need satisfied in their study group,
then it is likely because study groups usually do not help to fulfill a need for
affiliation.

Group Composition
Groups come in all shapes and sizes, and political groups are no excep­
tion. Groups can differ in size, composition, and type. Research suggests
naturally occurring groups are typically small, containing just two or three
persons (Desportes & Lemaine, 1988). People may prefer smaller groups
because they are confused by large groups (James, 1951) or because they
cannot easily control what happens to them in larger groups (Lawler, 1992).
Research examined some interesting effects of group size* For example, as
the size of the group increases, group members participate less (Patterson &
Schaeffer, 1977), display less commitment to the group (Widmeyer, Brawley,
& Carron, 1990), and show higher levels of tardiness, absenteeism, and
turnover (Durand, 1985; Spink & Carron, 1992), Other group dynamics are
also affected by group size. In larger groups, there tends to be more con­
flict (O'Dell, 1968), less cooperation (Brewer & Kramer, 1986), and less con­
formity to group norms (Olson & Caddell, 1994). Finally, group performance
can also be affected by the size of a group. In large groups, coordination is
more difficult (Diehl & Stroebe, 1987; Latane, Williams, & Harkins, 1979),
leading to decrements in performance, and it is easier to social loaf and free
ride, which can have harmful effects on the performance of a group (Karau
& Williams, 1993). In the following chapters, we examine large groups, such
as ethnic, national, and racial groups, as well as small groups involved in
political decision-making and political violence.
Groups can also differ in terms of their composition. The characteristics
of individual group members, such as sex, race, ethnicity, and physical
attractiveness, can be very important to the functioning of the group.
However, attention has also been focused on diversity within a group
(Levine & Moreland, 1998). Research examining the effects of diversity on
 The Political Psychology of Groups 85

communication suggests diversity can be harmful. As the degree of diversity

increases, group members tend to communicate with each other less and in
more formal ways (Zenger 8c Lawrence, 1989). When group members com­
municate less often, interpersonal conflicts become more likely (Maznevski,
1994). Diversity, however, can also be beneficial to group performance
(McLeod & Lobel, 1992) because it allows a group to be more flexible, foster
innovation, and improve the quantity and quality of relationships outside of
the group.
Groups can also be distinguished by their type. In a study, Lickel et al.
(2000) asked participants to categorize a large number of groups. Their
sorting resulted in four categories of groups: first, some groups, such as
families and romantic relationships, were categorized as intimacy groups',
second, task-oriented groups consisted of groups such as committees and
work groups; third, groups such as women and Americans were categorized
as social categories', and, finally, weak social relationships or associations
included such groups as those who enjoy a certain type of music or those
who live in the same neighborhood. Political groups certainly fall into the
task-oriented type, whether they are government working groups, juries,
political interest groups such as Greenpeace or Human Rights Watch, or
committees and subcommittees in Congress. Political groups can also be
social categories, such as ethnic groups, racial groups, or women, all with
particular political issues of concern.

Group Structure
Another important characteristic of a group is its structure. Every group
has a structure, and it tends to develop quickly and change slowly (Levine 8c
Moreland, 1998). Apparently group members need to know what the struc­
ture of a group is, and are reluctant to alter it once it is set. For example,
understanding the structure of a group, and how aspects of a group's struc­
ture can influence conflict and performance, is important. Aspects of group
structure include status, roles, norms, and cohesion.
Status in a group refers to how power is distributed among its members.
Indicators of high status include nonverbal behavior, such as standing
more erect, maintaining eye contact, and being more physically intru­
sive (Leffler, Gillespie, 8c Conaty, 1982), as well as verbal behavior, such as
speaking more, interrupting more, and being more likely to be spoken to
(Skovertz, 1988). The manner in which people acquire or are assigned status
is explained by two theories: expectation states theory (Berger, Rosenholtz,
8c Zelditch, 1980), suggested the expectations of a person, based on their
personal characteristics, contribute to group members' sense of the sorts
of accomplishments a person can achieve, and ethological theories (Mazur,
1985), which maintain that a group member acquires status when other
group members assess the person's strength by evaluating their demeanor
and appearance. However status is acquired, it is generally slow to change.
Because high status is associated with rewards, those high in status are reluc­
tant to give it up. And, because those high in status are usually evaluated
more favorably than those low in status, other group members are reluctant
to remove status (Messe, Kerr, 8c Sattler, 1992).
 86 The Political Psychology of Groups

The various roles group members hold constitute another important

component of group structure. Roles are expectations about how a person
ought to behave. Little is known about how roles in groups develop (Levine
& Moreland, 1998), except that task roles emerge before socioemotional
ones. Regardless of how roles develop, well-played roles can be beneficial
to a group (Barley & Bechky, 1994; Bastien & Hostager, 1998). Much of the
research on roles in groups focuses on the conflicts they create. Some role
conflicts occur as a result of role assignment, which refers to the decisions
made about who plays what role. Other conflicts center on role ambiguity
(uncertainty about how to behave in a role) or role strain (lacking knowledge
or ability to fulfill the role).
The norms of a group can be an important aspect of group structure.
Norms refer to expectations about how all group members should behave.^
Like roles, the formation of norms in a group can be difficult to identify.
Some argue that a group's initial behavior can be transformed into norms
(Feldman, 1984). Others argue that norms can arise from the expectations
for behavior people bring with them when they join a group (Bettenhausen
& Murnighan, 1991). Regardless of how norms are formed, there is strong
pressure to maintain them. Group members can impose strong sanctions
on members who violate the standards of behavior, and with good reason.
Research suggests adherence to norms improves the performance of a group
(Seashore, 1954). For example, in groups that have norms of productivity or
success, group members become more motivated to engage in behaviors or
tasks that ensure the success of the group. On the other hand, adherence
to norms can sometimes impede the performance of a group. If a norm of
laziness develops, for example, then group members might work less hard
to achieve their goals.
Cohesion refers to the factors that cause a group member to remain in
the group (Festinger, 1950). The importance of cohesion to a group's well­
being cannot be underestimated. It exerts powerful effects on a group's
longevity. As such, understanding how cohesion in a group develops is
important. There are several factors that affect the development of group
cohesion. First, the more time group members spend together, the more
cohesive they become (Griffith & Greenlees, 1993). Second, the more group
members like each other, the more cohesive the group is (Lott & Lott, 1965).
Third, groups that are more rewarding to their members are more cohesive
(Ruder & Gill, 1982). Fourth, external threats to a group can increase the
group's cohesiveness (Dion, 1979). Fifth, groups are more cohesive when
leaders encourage feelings of warmth among group members. Most studies
of the effects of cohesion on wellbeing and performance find a positive rela­
tionship. For example, members of cohesive groups are more likely to par­
ticipate in group activities and to remain in the group (Brawley, Carron,
& Widmeyer, 1988) and, in a meta-analysis of the effects of cohesion on
performance, Mullen and Copper (1994) found cohesive groups tend to per­
form better.
There are many studies of political decision-making groups, particularly
U.S. presidents and their close advisers, that show differences among those
groups in status, roles, norms, and cohesion. These studies are reviewed
extensively in Chapter 5. Here we provide a few examples, President John

i
 The Political Psychology of Groups 87

E Kennedy preferred a collegial advisory group. Although he was at the

top in terms of status, he saw the various advisers as colleagues. The group
formed at the outset of the administration, and members had their own
domain of expertise, which provided him with a particular role. In terms
of norms, conflicting viewpoints were encouraged, and all sides were taken
into account in searching for solutions to problems. President Nixon was
very different. His advisory group structure was hierarchical, with him on
top. Again, each adviser had a role to play, but conflict and brainstorming
were not encouraged. The emphasis in problem-solving was on technical
rather than political considerations. In the Clinton administration, role
assignments were ambiguous, which led to many delays and much turmoil
in policy-making.

Group Formation
If you think about all the groups of which you are a member, do you know
how or why each of those groups formed? What were the circumstances
surrounding the formation of each of your groups? Some of the answers
may be easier than others. For example, the animal shelter at which you vol­
unteer formed because there was a need to care for stray dogs and cats, and
the group of people with whom you spend free time formed because the
members liked one another. But how did the church you attend get started?
Why did the Tuesday night intramural softball team you play on come to
be? Groups researchers have yet to develop a comprehensive theory to
explain how and why groups form, but there are two perspectives that
offer promise. The functional perspective suggests groups form because
they serve a useful function or fulfill a need for their individual members
(Mackie & Goethals, 1987). For example, your animal shelter formed to
fulfill the need created by so many homeless dogs and cats. The interper­
sonal attraction perspective suggests groups form because members like
one another and seek to spend time together. Thus, the group of friends
with whom you spend time formed because you all liked one another and
wanted to spend time together.

Functional Perspective
According to the functional perspective, groups satisfy many needs,
including survival, psychological, informational, interpersonal, and col­
lective. Groups can be functional, in that they can fulfill many of out sur­
vival needs, including feeding, defense, nurturance, and reproduction
(Bertram, 1978; Harvey & Greene, 1981; Scott, 1981). Many of these needs
were stronger during earlier periods in history, but we still rely on groups
to fulfill many of these functions today. For example, we rely heavily on our
military forces to defend our country. We depend on farmers to provide
some of our food. And, to the extent we have a need to defend our country,
we might decide to join the armed forces. From an evolutionary psychology
perspective, membership in groups can help groups to fulfill their evolu­
tionary goals of survival, for most of the reasons expressed above.
88 The Political Psychology of Groups

Political Action Groups and the Internet

The internet seems to affect everything, even group-formation patterns.
Consider, for example, a political action group called MoveOn. Two
Silicon Valley entrepreneurs, Joan Blades and Wes Boyd, organized
MoveOn in 1998 when they reached a level of frustration with the effort
to impeach President Clinton. Then, after 9/11, Eli Pariser started an
online petition for peace. They joined forces and formed the MoveOn.
org political action committee, creating an international association of
over two million online activists.
This is an example of a cyber group that formed to achieve func­
tional goals of affecting politics through interest group activity. Will the
opportunity to form and join cybergroups affect group psychology? Will
it affect group influence in politics? (To learn about MoveOn, naturally,
you should visit their website, [Link].)
Groups can also satisfy a host of psychological needs, some of which we
introduced in Chapter 2. For example, joining a group can satisfy the need
for affiliation. Those with a high need for affiliation join groups more often,
communicate with others frequently, and seek social approval (McClelland,
1985). Groups also satisfy the need for power. People with a high need
for power want to control others (Winter, 1973). Schutz's Fundamental
Interpersonal Relations Orientation (FIRO) also explains how joining
a group fulfills psychological needs (Schütz, 1958). According to this per­
spective, joining a group satisfies three basic needs: inclusion (the desire to
be part of a group), control (the need to organize an aspect of the group),
and affection (the desire to establish positive relations with others). Joining
a group offers individuals a way to fulfill these needs.
Another category of needs is informational needs. Festinger (1950,1954)
argued that people join groups to provide standards with which to compare
their own beliefs, opinions, and attitudes. People often need to determine
whether their own viewpoints are correct or accurate. One way to make
such determinations is to seek similar people with whom to compare our
views. This perspective suggests people join groups to better understand
social reality.
Groups also meet people's interpersonal needs. Many groups provide
social support, giving emotional sustenance, advice, and valuable feedback.
Social support can be a valuable function of groups. Groups protect us from
the harmful effects of stress (Barrera, 1986). The social support of groups
also protects us from loneliness. Research indicates that people who were
members of many groups reported less loneliness (Rubenstein & Shaver,
1980). College students who eat dinner with others and spend time with
their friends also report being less lonely (Russell, Peplau, & Cutrona, 1980).
Finally, groups fulfill important collective needs. Sometimes, groups
are more productive and efficient than individuals working alone. Groups
often form because individuals believe pooling the efforts of multiple people
will lead to better outcomes. Some of the collective goals sought by groups
include engaging in the performing arts, enriching the leisure time of their
members, changing the opinions of persons outside the group, and making
routine individual tasks more tolerable (Zander, 1985).
 The Political Psychology o f Groups 89

Interpersonal Attraction Perspective

Sometimes, groups form because individuals discover they like each other
and want to spend more time together. There are many factors that influ­
ence our liking of another. First, we tend to be attracted to those who are
most similar to us in attitudes, beliefs, socioeconomic status, physical
appearance, and so on (Newcomb, 1960). This suggests we prefer to form
or join groups with people who are most similar to us. Second, we tend
to form relationships with those who are physically closer to us (Festinger,
Schachter, & Back, 1950). Thus, we tend to make friends with those who
live next door, those we sit next to in class, and those with whom we work
closely. We are likely, then, to form or join groups with people who are phys­
ically close. Third, we like people who like us (Newcomb, 1979). We are thus
more likely to form or join groups with people who are fond of us. Fourth,
we are attracted to people who are physically attractive. With the exception
of those who are extremely attractive, physically attractive people are more
accepted than those less physically attractive.
In summary, people join groups for a variety of reasons. One reason
people join and form groups is to satisfy a number of important needs,
including survival, psychological, informational, interpersonal, and col­
lective needs. We are more likely to join groups that can effectively satisfy
our needs. Another reason people join groups is to spend more time with
the people they like. Such situations, especially when reciprocal, can be very
rewarding.

Group Development
Think again about the groups to which you belong. Have they remained
the same over time or have they changed somehow? Most likely, groups
of which you are a member have changed somewhat over time, but how?
Group development refers to the stages of growth and change that occur
in a group, from its formation to its dissolution (Forsyth, 1990). Of course,
there is disagreement among researchers about the number and types
of stages, but most models include the following basic stages: forming,
storming, norming, performing, and adjourning (Tuckman, 1965; Tuckman
& Jensen, 1977).
The first stage refers to the point during which the collection of indi­
viduals is forming. This stage is also referred to as the orientation stage
because prospective members are orienting themselves to the group.
During this stage, individuals are getting to know one another. The stage is
often characterized as one with a fair amount of tension—prospective group
members are on guard, reluctant to share much information or discuss their
personal views. Also, as you can imagine, group norms have not yet formed,
making this a difficult period of development. In fact, the tension can be
so high that those who believe they lack the skills necessary to effectively
handle such a situation try to avoid group membership (Cook, 1977; Leary,
1983). Over time, tensions lessen and group members begin to exchange
more information. Also, feelings of interdependence—one of the defining
features of a group—increase during this stage. In Chapters 10 and 12, we
look at a number of groups of political extremists, such as the Nazi SS and
90 The Political Psychology of Groups

terrorists. In such cases, careful attention is given to this stage to ensure

only people with particular characteristics are included.
The second stage of group development, storming, is characterized as
one of conflict. Many types of conflict exist. Some conflicts occur when a
person's position or action is misinterpreted (Deutsch, 1973). Other conflicts
arise when a group member's behavior is deemed to be distracting, such as
when a group member consistently arrives 15 minutes late for meetings.
Other types of conflicts can escalate, such as when minor disagreements
turn into major points of contention. Although conflicts, especially those
that escalate, can disrupt the group, they can serve as important catalysts
for group development. Conflicts can serve to promote group unity, inter­
dependence and stability, and cohesion (Bennis & Shepard, 1956; Coser,
1956; Deutsch, 1969).
Norming, the third stage of group development, is a phase in which con­
flict is replaced with cohesion and feelings of unity. When groups become
more cohesive, they have a heightened sense of unity. The relationships
among members become stronger, as do individual members' sense of
belonging. The degree of group members' identification is heightened
during this period. Another characteristic of groups in this stage of devel­
opment is stability. There is a low turnover of members, a low absentee rate,
and a high rate of involvement.

Urban Street Gangs as Groups

Urban street gangs in the United States, and elsewhere, provide
illustrations of the power of group demands for loyalty, conformity, and
obedience. In the book Monster: The Autobiography of an L.A. Gang
Member (Shakur, 1993), Sanyika Shakur, a.k.a. Monster Kody, describes
those group dynamics.
(1) Belonging to the group enhances self-esteem, and cohesive groups
demand strong loyalty:
Actually, I wasn't fully aware of the gang's strong gravitational
pull. I knew, for instance, that the total lawlessness was alluring,
and that the sense of importance, self-worth, and raw power was
exciting, stimulating, and intoxicating beyond any other high on
this planet. But still I could not explain what had happened to pull
me in so far that nothing outside of my set mattered (p. 70).
(2) Loyalty and solidarity are described in passages such as this:
I went to trial [for murder] three months later. The gang turnout
was surprising. Along with my family, at least fifteen of my
homeboys came. All were in full gear (gear as in gang clothes,
colors and hats—actually uniforms) (p. 23).
(3) He describes the norms of his gang as "You are your brother's keeper";
when war is declared, all members are expected to fight; reject family
and other agents, like the schools and teachers and police, to associate
{Continued)
 The Political Psychology of Groups 91

0Continued)
with the gang; respect and honor others according to their status; pro­
tect the gang turf; and retaliate against all perceived offenses.
(4) Shakur also outlines what the gang values, including trouble (fighting,
drinking, drugs, and sex), toughness, smartness (respect for streetwise
savvy), and fatalism (they did not believe they would grow old).

During this stage of development, group members also report a high

degree of satisfaction with the group. They enjoy the group more, note
increases in self-esteem and security, and experience lower levels of anx­
iety. Finally, the internal dynamics of the group begin to intensify. There is
greater acceptance of the group's goals by group members, a low tolerance
for disagreement, and increased pressures to conform.
The fourth stage of group development is characterized by performing.
Performance usually only occurs when groups mature and have success­
fully gone through the previous stages of development (Forsyth, 1990). In a
study of neighborhood action groups (Zurcher, 1969), only one of 12 groups
reached the performing stage. All others were stuck in the conflict or cohe­
sion stages.
A group's decision to dissolve (adjourning) can either be planned or
spontaneous. A planned dissolution occurs when the group accomplishes
its intended goals or exhausts its time and resources. Examples of groups
with planned dissolutions include a jury that has reached a verdict, a softball
team playing its last game of the season, or a class that dissolves because the
semester has ended. Spontaneous dissolutions occur when unanticipated
problems arise that prevent the group from continuing. Examples of groups
with unplanned dissolutions include those that repeatedly failed, or those
that fail to satisfy their members' needs.

INFLUENCE IN GROUPS
Groups can exert a great deal of influence over their members. When people
are in a group, there is a strong tendency to adhere to the group's norms.
When group members act in accord with group norms, they are conforming.
Conformity refers to the tendency to change one's beliefs or behaviors so
they are consistent with the standards set by the group. Americans tend to be
ambivalent about the notion of conformity. On the one hand, to conform is
to be "spineless" and "wishy-washy"; because Americans tend to value indi­
vidualism, being labeled a conformist can be a negative label. On the other
hand, conformity is valued because it leads to harmony and peace. Imagine
a world in which no one conformed. In this section, we examine some of
the early studies on norm formation and conformity. We also explore the
reasons why people conform, as well as when people conform.
One of the earliest studies of conformity was conducted by Sherif
(1936), who was interested in how group norms form. To understand norm
formation, he made use of the autokinetic effect, which refers to a per­
ceptual illusion that occurs when a single point of light in a darkened room
92 The Political Psychology of Groups

appears to be moving. In Sheriff experiments, he asked participants to

stare at the point of light and estimate how far it moved. In reality, the light
did not move at all, so there was no correct answer on this task. In his first
experiment, Sherif asked individual participants to estimate, over a series
of trials, how far the light moved. The pattern of responses was nearly
identical for all participants; initially, their estimates were quite variable,
but over time they settled on a single estimate, such as 3 inches. In the next
experiment, Sherif asked pairs of participants to estimate, over a series of
trials, how far the light moved. Again, the pattern of responses for each
pair was nearly the same—variability in their initial estimates, then con­
vergence on how far the light moved. These experiments were important
in showing how norms form. Eventually, individual or pairs of participants
formed a standard for how far the light moved. In Sherifs third experi­
ment, he sought to determine whether people could be persuaded to con­
form to the judgment of another person* Participants in this experiment
made judgments in groups of two. In reality, only one of those in each
pair was a real participant; the other was a confederate of the researcher.
The confederate was asked to make estimates either lower or higher than
the real participant. Over time, the participant began to make estimates
closer to the estimates of the confederate, suggesting participants were
conforming to the standards set by the confederate. These experiments
were important in demonstrating that, in ambiguous situations where
there is no correct answer, people tend to conform to a norm. Another
researcher, Asch (1955), wondered whether participants would be as likely
to conform when the situation was not so ambiguous—that is, when there
was a correct answer on a judgment task. To answer this query, Asch
asked five participants to take part in a perceptual judgment task. The
participants were shown a series of three lines, varying in length. Their
, task was to determine which of the three lines matched a target line. The
task was designed to be unambiguous: there was clearly a correct answer.
Each participant, in turn, was asked to indicate, aloud, his answer to the
experimenter. In reality, the first four participants were confederates of
the experimenter. The person sitting in the fifth position was the real
participant. On half of the trials, they instructed the four confederates to
give the (clearly) wrong answer. The question was whether the fifth (real)
participant would also give the wrong answer. The results showed 75% of
the participants went along with the group and gave the wrong answer at
least once. Apparently, the pressure to conform was so strong that, even
on this unambiguous task with a clearly correct answer, participants were
willing to give an answer they knew was wrong.
Both these experiments are important in showing that people conform.
But why do they? Research suggests people conform for two reasons: to be
liked, and to be correct (Cialdini & Trost, 1998). In Sherifs (1936) study,
people conformed because they wanted to be correct. One way to be correct
is to gather as much information as possible before acting or making a deci­
sion. For example, one of the authors was recently in London and had to
take a train to the airport. Not knowing where or how to buy a train ticket or
where to board the train, she spent time observing what other people were
doing. In doing so, she gathered enough information so that she was able to
successfully purchase a train ticket and board the correct train. Whenever
f
 The Political Psychology of Groups 93

we use other people's actions or opinions to define reality, we conform

because of informational social influence.
Conformity on the basis of informational social influence occurs when-
ever we are uncertain about the correct or appropriate action. In the Sherif
(1936) studies, for example, the task was novel and ambiguous. Under these
circumstances, the best course of action is to gather information from other
participants, in order to arrive at the best answer. If we have a great deal
of confidence in our knowledge or ability to make the right decision, then
there is little reason to rely on others for information. Research suggests that
when our motivation to be correct is high, we tend to conform more when
we are uncertain about the correct answer than when we are certain (Baron,
Vandello, & Brunsman, 1996).
In Asch's study (1955), people may have conformed not because they
wanted to be correct, but because they wanted to be liked; this is called nor­
mative social influence. Sometimes, as in the Asch study, people clearly
gave a wrong answer in order to be liked and accepted by the group. In
these situations, the group has a powerful, if unspoken, influence over group
members' behavior. In an interesting twist on normative social influence,
two social psychologists investigated "jeer pressure,'' or the tendency to con­
form in order to avoid rejection from peers (Janes & Olson, 2000). When we
observe another person being rejected by the group, there is a tendency to
conform even more strongly to the standards set by the group, presumably
to avoid similar rejection from group members.

Situational Conformity
If you think about your own behavior, there probably were times when you
conformed or felt the pressure to conform more than others. Some aspects
of a situation lead to more pressure to conform than others. These factors
include the size of the group, group unanimity, commitment to the group,
and individuation and deindividuation.
Intuitively, one would predict that the pressure to conform is greater as
the size of the group increases. Early research (Asch, 1956) suggested that as
group size increased, so did conformity - but only to a point. Once the size
of the group reached about three members, conformity seemed to level off.
But later research (Bond & Smith, 1996) suggests conformity increases up to
a group size of eight members. So, it seems the larger the group, the greater
our tendency to conform. Group unanimity is also important. Imagine being
in the Asch line study-—in which all the group members give the (clearly)
wrong answer. Now, it is your turn to give your answer. What do you do?
Asch's results suggest you would give the wrong answer at least once. But
now imagine just one other member of the group gives the correct answer,
one that disagrees with the other group members. Now, what answer would
you give? Research (Asch, 1955; Morris & Miller, 1975) suggests conformity
drops if there is even one dissenter in the group.
Groups with highly committed members are more likely to conform than
members with less commitment (Forsyth, 1990). Obviously, highly committed
group members want to be liked and accepted by other group members. One
way to ensure being liked and accepted is to go along with the group.
94 The Political Psychology of Groups

One individual difference variable that predicts the tendency to con­

form or not is individuation. Individuation refers to the desire to be dis­
tinguishable from others in some aspect (Maslach, Stapp, & Santee, 1985;
Whitney, Sagrestano, & Maslach, 1994). Some people have a greater desire
than others to differentiate themselves. Those high in the desire for indi­
viduation are less likely to conform than those low in individuation.
Conversely, deindividuation can increase conformity. When this occurs,
people attribute their behavior to being part of the group's behavior and
there is a diffusion of responsibility. People feel less responsible for their
actions when those actions take place in a group context than they would if
they committed those acts alone.

The Tulsa Race Riot, May 31-June 1,1921

Mobs and riots are one of the most frightening and destructive instances
of group behavior, resulting, in part, from situational conformity factors.
One example of mob behavior with racist motivations occurred in Tulsa,
Oklahoma, in 1921. At that time, Tulsa was home to the most prosperous
Black community in the United States, called Greenwood. About 10,000
people lived in this 34-block neighborhood. It was separated from the
White community by railroad tracks. Tensions between the Black and
White communities increased in May 1921, when a Black man was
accused of assaulting a White woman. Fighting ensued, and on May 31 a
White mob pushed the Blacks across the railroad track and proceeded to
burn down Greenwood. It soon became evident the Whites would setde
for nothing less than the complete destruction of the Black community
and every vestige of Black prosperity. They spread gasoline inside homes
and businesses and set them on fire. Blacks fled; some were shot down
while they ran, and some burned to death in the buildings. The Whites
arrested any Blacks they caught but didn't kill. Before they burned, they
looted and stole personal property. It still is not known whether the mob
acted spontaneously, or was organized by the Klu Klux Klan (KICK), the
police, or another entity. For the full story, read Tim Madigan's book, The
Burning (2001).

Power
Implicit in our discussion of influence in groups is power. Power is the cap­
acity to influence other people (French & Raven, 1959). In groups, power
can be advantageous. Powerful group members can resolve group conflicts
more efficiently than those with less power (Levine & Moreland, 1998),
and powerful members are better liked and are deferred to more than less
powerful group members (Shaw, 1981). Of course, having power can also
serve as a disadvantage; those with power are granted the responsibility
to be effective leaders (Hollander, 1985), exercising power can be stressful
(Fodor, 1985), and it can lead to faulty perceptions of oneself and others
(Kipnis, 1984). In this section, we examine the bases of power, as well as the
reactions of group members to the exercise of power.
{
 The Political Psychology of Groups 95

One of the most influential typologies of power is French and Raven's

(1959; Raven, 1965) critical bases of power. The typology assumes that a
group member's ability to exert power over another member or the entire
group can be derived from one or more of the following lands of power:
reward, coercive, legitimate, referent, and expert. Reward power is defined
as the ability to control the distribution of positive and negative reinforcers.
In groups, there are many rewards: praise for good performance, money for
work completed, and trophies for winning championships. Group members
who control the distribution of those rewards are granted the most power.
For example, teachers exert power over students to study hard because
they control the distribution of good grades. Of course, the group member
who controls the distribution of rewards is only powerful if the rewards are
valued by the group member, the group member depends on the power-
holder for the reward, and the power-holder's promises are sincere (Forsyth,
1990). When a power-holder is the only one in the group who can dis­
tribute rewards, their position as a power-holder becomes more secure.
Coercive power refers to the capacity to punish those who do not comply
with requests or demands. For example, if one country threatens another
with attacks or boycotts, then the country is using coercive power. Teachers
can use coercive power to get students to work harder by assigning extra
work. Research suggests that given the choice of using reward or coercive
power, most will choose reward (Molm, 1987,1988). Those with legitimate
power have a right, by virtue of their position, to require compliance. For
example, when a military officer orders troops into battle, that officer is
exerting legitimate power. With legitimate power, the power-holder has the
right to exercise power, and the target has a duty to obey the power-holder.
An interesting characteristic of legitimate power is the power-holder is typ­
ically chosen to occupy the position of power, granting them the support of
the majority.
When we identify with someone because they are similar to us or because
we want to be like that person, the person then possesses referent power.
When a child imitates an older sister because the child wants to be like her,
this is an example of referent power. Of course, advertisers make use of ref­
erent power when, for example, they encourage young people to purchase
cigarettes by implying they will make them look like the attractive models
in the advertisements. Special knowledge, skill, or ability one possesses
can serve as a basis for expert power. Physicians, for example, are often
afforded a great deal of power because of their knowledge and ability. Of
course, expert power can only be exerted if the target of power is aware
of the power-holder's special knowledge or talent (Foschi, Warriner, &
Hart, 1985).

Reactions to the Use of Power

One of the goals of power exertion is to affect change. When one country
threatens to attack another country if that country does not comply with
certain demands, there is an expectation that the target country will change.
Of course, other changes may occur in the target country as a result of the
use of power tactics, including compliance, attraction, conflict, rebellion,
96 The Political Psychology of Groups
i

motivation, and self-blame (Forsyth, 1990). Compliance occurs when a

powerful member of the group asks a less powerful member of the group
to do something, and the member does what is asked. This response is
consistent with the complementarity hypothesis (Carson, 1969; Gifford &
O'Connor, 1987; Kiesler, 1983), which suggests that when one person acts
in a powerful manner, the other person will become submissive. Such a
response also ensures the power-holder will retain their power. Of course,
the complying group member need not change his or her attitudes or
behaviors permanendy. In fact, although a group member agrees to the
demands or requests of the power-holder, this does not necessarily corres­
pond to a permanent change in behavior or attitude (Kelman, 1958,1961). It
is only when the target of power internalizes the power-holder's views that
a permanent change in behavior or belief occurs. Note that compliance is
different from conformity, although both types of social influence can result
in a change in behavior. Compliance involves behavior motivated by a par­
ticular request, and conformity involves behavior motivated by a need to be
liked or correct.
Attraction is also affected by having power. A potential consequence of
having power is not being liked by targets of power. In general, we tend not to
like those who use power in direct and irrational ways (Forsyth, 1990). This
is not to suggest we dislike all powerful people. Research indicates targets
of power tend to like those who influence them via discussion, persuasion, or
expertise more than those who influence them via manipulation, evasion, or
threat (Falbo, 1977). Regarding the bases of power discussed earlier, research
shows managers who use referent power are liked the most, and those who
use coercive power are liked the least (Shaw & Condelli, 1986).
Another consequence of the use of power in a group is conflict and tension
(Forsyth, 1990). Some types of power engender more conflict than others.
For example, group members often respond to coercive power with anger
and hostility (Johnson & Ewens, 1971), except in situations when the group is
successful (Michener & Lawler, 1975), they have a trusted leader (Friedland,
1976), or the use of coercive power is normative for the group (Michener
& Burt, 1975). One problem with responses that involve anger and conflict
is that the functioning of the group may be compromised (Forsyth, 1990).
One group member's anger can be fueled by another group member's anger,
which can result in an escalation of anger and hostility. Research suggests if
a group member abuses power, a typical response is rebellion on the part
of other group members (Lawler & Thompson, 1978). Abuses of power can
also lead to reactance, a feeling one's freedom was limited or taken away
(Brehm, 1976). When group members believe their freedom (of choice, for
example) has been removed, they respond by becoming defiant and refusing
to go along with the leader (Worchel & Brehm, 1971).
Motivation can also be influenced when power is exercised in a group.
Often, group members are motivated intrinsically—that is, they enjoy being
productive and doing good work because they derive personal satisfaction
from it. But if a leader uses reward or coercive power, which often involves
the use of extrinsic rewards (e.g., money, promises), group members may
become less motivated to work hard and do a good job.
In some circumstances, a leader might be so abusive that they cause
group members to suffer tremendously. If group members believe the world
 The Political Psychology of Groups 97

is just, then they are likely to think they got what they deserve (Lerner &
Miller, 1978). They might come to believe they deserve to suffer and engage
in self-blame. A belief such as this allows suffering group members to make
sense of their plight.
Group leaders exercise power in a number of ways. There are several
bases of power, some of which are more conducive to certain situations than
others. If group leaders can choose which bases of power to use, it behooves
them to carefully consider the consequences. As we have seen, the use of
power can engender many reactions, some of which can be good for the
functioning and wellbeing of the group, but others detrimental.

Minority Influence
A final topic of interest for the study of social influence in groups is minority
influence. Sometimes there are lone dissenters or a small faction of the group
that will refuse to go along with the others. Of interest to social psychologists
is the success of minorities in exerting influence. Research suggests minor­
ities successfully influence majorities under specific circumstances (Kaarbo,
1998; Kaarbo & Beasley, 1998; Moscovici, 1985). First, for minorities to
be successful in exerting influence on majorities, they must be consistent
in their opposition (Wood, Lundgren, Ouellette, Busceme, & Blackstone,
1994). Members of a consistent minority are perceived as being more honest
and competent (Bassili & Provencal, 1988). If they are inconsistent or appear
divided in anyway, then their influence is greatly diminished. Second, minor­
ities are more successful if they are able to refute the majority's arguments
successfully (Clark, 1990). Third, minorities are more successful if the issue
is not of great personal relevance to members of the majority (Trost, Maass,
& Kendrick, 1992). Finally, minorities are likely to be successful when they
are similar to the majority groups in most respects, except for the disagree­
ment at hand (Volpato, Maass, Mucchi-Faina, & Vitti, 1990). For example, if
a member of the Republican Party were trying to convince other Republicans
to change their views on homeland security, that member would be more
successful than if the would-be persuader were a Democrat. In this case, the
Republican dissenter is more similar (in terms of party membership) to the
majority than is the Democrat.
Successful minorities may be able to change the position of the majority,
which in the political realm may amount to a policy change. Short of
affecting policy as a whole, they may be able to indirectly affect it through
pressuring the majority to move in a particular direction, or affecting the
information received by the majority. Finally, studies show minorities can
improve the quality of a group's decision-making (Nemeth, 1986). Given
the inter-agency nature of many government decision-making units, which
makes the presence of minorities a frequent occurrence, understanding the
role of minorities helps us to understand both change in policy and shifts in
policy. Kaarbo and Gruenfeld (1998) point to a number of examples; change
in Japan's foreign aid policy from one that emphasized Japan's self-interest to
one reflecting humanitarian interest occurred because of a small minority in
the foreign ministry, which was pitted against the large and powerful minis­
tries of finance and international trade. Soviet policy toward Czechoslovakia
98 The Political Psychology of Groups

in 1968 changed because of the pro-interventionist minority who, through

manipulation of information and the decision-making process, moved
themselves from the minority to the majority.

GROUP PERFORMANCE
One of the primary functions of a group is to perform a task, and one of the
unique characteristics of a group is that its tasks typically are performed
in the presence of others. For some groups, tasks are performed in the
presence of other people, such as in a factory. For other groups, tasks require
group members to depend on one another to successfully complete them,
such as in an assembly line. In this section, we examine research suggesting
that sometimes the presence of other people enhances performance (social
facilitation) and at other times it hinders performance (inhibition).
Groups are often assumed to accomplish more than individuals and to
perform better than individuals. Yet research suggests groups do not always
perform better than individuals. We examine the various productivity losses
in groups, including coordination and motivation losses. Finally, we explore
some of the techniques used to help groups function more effectively,

Social Facilitation and Inhibition

Have you ever noticed that when you run a five kilometer race your time
is always better than when you time yourself during training? Why is it the
speech you gave in your communications class was better than when you
practiced it at home by yourself? In some situations, we appear to perform
better in the presence of other people than when alone, which is an effect
known as social facilitation* One of the first experiments ever conducted
in social psychology was designed to examine the effects of the presence of
others on an individual's performance, Norman Triplett (1898) tested the
hypothesis that people perform better in the presence of others than when
alone. In his study, he had children play a game alone or with one other
person. His results confirmed his hypothesis: when paired with another
person, individual performance is better than when performing alone. This
and subsequent research suggested that, if given a choice between working
alone or in the presence of other people, we would be better off performing
a task in the presence of others.
Now imagine another situation. You are playing on a basketball team.
Your coach spends hours helping you learn to shoot a left-handed layup,
which is not an easy shot for a right-handed person. When by yourself,
you can shoot 20 left-handed layups easily, But what happens when you
are playing a game in front of a cheering audience? Evidence suggests you
would miss the layup. This effect, known as social inhibitionf occurs when
the presence of others inhibits performance. According to research in this
area, we would be better off working alone than in the presence of others.
These two effects—facilitation and inhibition—seem contradictory. One
suggests working in groups can enhance performance,* the other suggests
the opposite. Zajonc (1965) reconciled these two seemingly contradictory
 The Political Psychology o f Groups 99

findings by arguing that the presence of others enhances performance on

well-learned or simple tasks, but inhibits performance on difficult or novel
tasks. The presence of other people enhances the tendency to display the
dominant (well-learned) response and inhibits the tendency to suppress the
nondominant response. Because running is a fairly simple task, the presence
of others during a race should enhance performance. But shooting a left-
handed layup when you are right-handed is a difficult task, so the presence of
a cheering crowd or other teammates should hurt performance because our
tendency is to shoot the ball with our right hand. A comprehensive review
of research in this area basically confirmed Zajone’s perspective (Bond &
Titus, 1983). The presence of others improves the quantity of performance
on simple tasks and decreases the quality and quantity of performance on
difficult tasks.
Zajonc’s (1965) perspective explains when facilitation and inhibition
occur, but why do these effects occur? What is it about the social situation
that causes improvement in performance on simple tasks, but decreases
performance on difficult tasks? Researchers in the area developed three
explanations; arousal, evaluation apprehension, and distraction. Zajonc
(1965,1980b) argues that the mere presence of others increases the arousal
level of the performer. When individuals are in a heightened state of arousal,
the tendency to display a dominant response is increased. If the dominant
response (shooting the ball with your right hand) is the correct one, then
social facilitation occurs. If the dominant response is not the correct one,
then social inhibition occurs. Cottrell (1972) agrees that the presence of
others causes arousal, but he argues that the source of arousal is evaluation
apprehension, or the anxiety created by the fear one is being evaluated. In a
study to test this idea (Cottrell, Wack, Sekerak, & Rittle, 1968), participants
were asked to work on a task alone, in the presence of others who were
also working on the. task, or in the presence of others who were blindfolded
(and thus could not see what participants were doing). The results showed
social facilitation occurred only when the others present could see the par­
ticipant perform the task. When the possibility of evaluation was removed
(in the blindfolded participants’ condition), social facilitation did not occur.
Finally, according to the distraction explanation, the presence of others is
potentially distracting. When one is distracted, paying attention to the task
at hand can be difficult. Such distractions create conflict as to whether to
pay attention to the audience or to the task. When one is distracted, more
effort is required to focus attention on the task, thereby improving per­
formance on simple or well-learned tasks. When tasks are difficult, even the
increase in effort is not enough to improve performance and usually leads to
impaired performance (Baron, 1989; Groff, Baron, & Moore, 1983).

Productivity Losses
As mentioned previously, there is a belief that groups are more productive
than individuals. More than likely, you were in a group that seems not to
have lived up to its fullest potential. Clearly, groups are not always as pro­
ductive as they should be. According to Steiner (1972), there are two reasons
for process losses in groups. One is that the responses of individual group
100 The Political Psychology of Groups

members are not combined in a way that enhances group productivity

Decrements in performance caused by poor coordination are known as
coordination losses. In an operating room, for example, coordination losses
occur if the surgeon is not handed the correct surgical instruments. Another
reason for productivity losses is known as motivation losses. These occur
when individual group members fail to exert their maximum effort on a
task. The operating room team will not perform at its maximum level if one
of the team members does not complete his or her assignment effectively.
Although both coordination and motivation losses in groups are interesting,
most attention has been paid to motivation losses. One such motivation loss
that has received a great deal of attention is social loafing.
Social loafing refers to the tendency of group members to work less hard
when in a group than when working alone. One of the earliest studies of
social loafing was conducted by Ringelmann (1913), who found people exert
less effort when pulling a rope or pushing a cart if they work in a group than
if they work alone. In another interesting study (Latane et al., 1979), groups
of six participants were asked to wear a blindfold, sit in a semicircle and
listen (via headphones) to the noise of people shouting. Participants were
asked to shout as loudly as they could while listening to the noise through
their headphones. On some trials, participants believed they were shouting
alone or with one other person. On other trials, they believed everyone was
shouting. When participants thought they were shouting with one other
person, they shouted 82% as intensely as when they thought they were alone.
When they thought everyone was shouting, they shouted 75% as intensely.
Because social loafing can lead to severe performance decrements
in groups, efforts are made to reduce or eliminate it. First, social loafing
can be reduced if each group member's contributions are clearly identifi­
able (Hardy & Latane, 1986; Kerr & Bruun, 1981). When the possibility of
being evaluated is evident, group members appear to give maximum effort
(Harkins, 1987). Second, if group members find the work to be interesting
and involving, they are less likely to loaf (Brickner, Harkins, & Ostrom,
1986; Harkins & Petty, 1982; Zaccaro, 1984). Third, if group members take
personal responsibility for the group's outcome, then they are less likely to
loaf (Kerr & Bruun, 1983). Group members need to believe their individual
efforts will have an impact on the group's outcome.

Improving Productivity
In addition to efforts to reduce social loafing in groups, researchers developed
techniques to help groups function more effectively and avoid production
losses of any kind. One such technique is team development, which includes
a variety of techniques, such as sensitivity training, problem identifica­
tion, and role analysis (Dyer, 1987). Techniques such as these are designed
to improve both the task and interpersonal skills of group members. A
similar technique involves the use of quality circles (Marks, Mirvis, Hackett,
& Grady, 1986). If group members engage in regular meetings to discuss
problems with productivity and ways to solve the problems, then product­
ivity losses can often be reduced. Another technique involves the use of
autonomous workgroups (Pearson, 1992). This technique involves the use of
self-managed work teams who can control how tasks are performed.
 The Political Psychology of Groups 101

Many of these techniques require groups to change how they function«

There are also techniques that focus on individual group members. For
example, in participative goal-setting, individual group members are respon­
sible for setting the group's productivity goals (Pearson, 1987; Pritchard,
Jones, Roth, Stuebing, & Ekeberg, 1988). Another technique, task design,
involves changing the attributes of the task to make it more attractive to
group members (Hackman & Lawler, 1971). Both these techniques involve
changing group members' perceptions of the task, rather than the task
itself.

GROUP DECISION-MAKING
The discussion of group productivity attests to the fact that groups are fre­
quently called upon to perform a variety of activities. An important activity
that groups—-especially political groups—are often asked to do is to make
decisions. Political groups are often responsible for making decisions with
large-scale consequences, such as whether to send troops to a region in
conflict or to escalate an existing conflict. As in productivity tasks, groups
are often assumed to make better decisions than individuals. Groups can
pool all the best resources that individual group members can offer. In this
section, we examine the group decision-making process, including how
decisions are made, the stages of group decision-making, and how indi­
vidual resources are pooled; then we examine research on the effectiveness
of individual versus group decisions. We also look at research suggesting
that groups often make bad decisions, and finally, we explore some tactics
to improve the decisions made by groups.

The Decision-Making Process

Imagine that a group of people, such as a jury, were assembled to make an
important decision. A jury spends time listening to testimony and the pres­
entation of evidence. When all the evidence is presented, the jury meets to
discuss its verdict. At the end of their deliberations, which can last from
a couple of minutes to weeks and months, the jurors reach a final, typic­
ally unanimous decision. From the perspective of an observer, the jury
appears to leave the courtroom and magically return with a verdict. But
what happened between the time the jury left the courtroom to deliberate
and when they returned with a verdict? How did this group of people reach
a decision about what should happen to the defendant? The group decision­
making process has been studied extensively, and several models help us
understand how groups arrive at a decision.

Three-Stage M odel o f Group Decision-Making

According to Bales and Strodtbeck (1951), groups proceed through a three-
stage model of group decision-making before eventually arriving at
a decision. In the orientation stage, group members spend time defining
the problem and planning their strategy for solving the problem. Research
(Hackman & Morris, 1975) suggests most groups spend little time in this
102 The Political Psychology of Groups

phase, assuming planning is a waste of their time, but groups that spend a
fair amount of time in the orientation phase are more successful (Hackman,
Brousseau, & Weiss, 1976; Hirokawa, 1980), In the discussion stage, group
members spend their time gathering information, identifying and evaluating
alternatives. The amount of time groups spend in this stage is also related
to the quality of the group's decisions (Harper & Aslding, 1980; Laughlin,
1988). However, groups do not often make full use of this stage (Janis &
Mann, 1977; Stasser & Titus, 1987). In addition, the use of information by
groups at this stage is problematic, in that new information brought forth
by one member of the group, but unknown to other members, is not fully
considered in discussion. In fact, groups tend to "omit unshared information
from discussions while focusing on information that all members already
know” (Wittenbaum & Stasser, 1996, p, 5). In the decision-making stage,
groups choose a solution. How groups combine the individual preferences
to reach a group decision can be explained by understanding the group's
social-decision scheme.

Social-Decision Schemes
Social-decision schemes refer to the process by which groups combine
the preferences of all the members of the group to arrive at a single group
decision (Stasser, Kerr, & Davis, 1989). If groups use the majority-wins rule,
then they combine individual preferences by opting for whatever position
is supported by the greatest number of group members. For example, if 10
of 12 jury members believe they should deliver a "guilty” verdict, then the
group's final decision will be "guilty.” In the truth-wins rule, group members
tend to be persuaded by the truth of a particular position. This rule tends to
be adopted when group members are discussing facts rather than opinions.
Another decision scheme that groups use is the first-shift rule, by which
groups tend to adopt the decision that is consistent with the first shift in
group members' opinions.
Describing social-decision schemes from a research perspective may
leave you with a sense that the process occurs without pressures or emotions,
but pressures, such as conformity pressures, occur during this process—and
they can be extreme, A book describing jury deliberations in a murder case,
written by the foreman of the jury, describes the pressure put upon the only
person reluctant to vote "Not guilty”:
Without pausing, Ivtook the cards out of my pocket and passed them
around ... There was silence as the cards started to come back, each
folded in half. I counted them. Nine. We waited, and two more came
in. Eleven. We waited. Still eleven. At this point there was no confu­
sion about who still held a card. Adelle [the holdout] sat at the corner
of the table to my left... She was looking fixedly away, up, behind her,
out the window. No one spoke ... One sensed everyone in the room
concentrating on the blank card in rapt meditation. Adelle breathed
audibly, wrote something rapidly on the card, closed it on itself, and
pushed it into the middle of the table. I placed it, consciously and
more or less conspicuously, at the bottom of the pile. I wanted the full
dismay of the room to land on her if she had voted for a conviction.

f
 The Political Psychology of Groups 103

Then I began to open the cards and read them: not guilty, not guilty...
And the last one: Not guilty.
(Burnett, 2001, p. 166)

Emotions and Decisions

As you can imagine, people and groups make decisions for a variety of
reasons ranging from the mundane (e.g., choosing the background on a
print advertisement) to the serious (e.g., deciding whether or not to engage
in a terrorist activity). As such, some decisions are made with weak emotion;
others with quite strong emotions. One question that is useful to ponder is
what effect emotions have on decision-making, The answer to this question
is more complicated than it may appear on the surface. Research (Leith &
Baumeister, 1996) on the effects of emotion on risky decisions shows when
people are angry, embarrassed, or upset, they tend to make risky decisions,
presumably in an attempt to gain a positive outcome, thereby offsetting or
neutralizing the negative emotion. Why do people experiencing negative
emotions make riskier decisions? The answer is that when people are angry,
embarrassed or upset, their ability to cognitively process rational informa­
tion is diminished. They appear to be less able to think things through and
consider all possible outcomes (Leith & Baumeister, 1996),
Emotions also impact how people search for information when making
decisions. Often, the quality of a decision is impacted by the amount of
information a decision-maker uses. In addition, decision-makers typically
strive for information-search efficiency, which refers to the amount of infor­
mation gathered relative to the total amount of information available. As it
turns out, a decision-maker's mood can impact how risky they are in their
search for information. Blay, Kadous, and Sawers (2012) found that when
decision-makers are in a negative mood, their information search efficiency
improved when the risk is high. When the risk is low, however, being in a
good mood impairs efficiency because decision-makers are less focused in
their search.
So does this mean that people should not make decisions when they
are feeling emotional? Not necessarily. Baumeister, Vohs, and Tice (2006)
suggest that sometimes being emotional can lead to good decisions and
sometimes bad decisions. What matters is the type and nature of the
emotion. Some emotions are anticipatory in nature—that is, they are
emotions people expect to feel once they made a decision. Consider a deci­
sion such as deciding whether to break up with your significant other. How
you expect to feel once the break-up happened influences your decision
about whether to initiate the break-up or not. Baumeister, Vohs, and Tice
(2006) found anticipated emotion can have beneficial effects on decision­
making because people are likely to base their decision on trying to achieve
the best emotional outcome, such as feeling better or feeling good. On
the other hand, one's current (as opposed to anticipated) emotional state
can lead to poor decisions because people want to make a decision that
will improve their mood or affect, thus causing them to engage in risky
decisions. Current emotional states, if they are negative, motivate people
to make decisions that will make them feel better right now. Such â strategy
often causes people to make irrational and bad decisions.
104 The Political Psychology of Groups

Groups and Political Decision-Making Units

Political decisions are made in response to a perceived problem, and they
tend to occur sequentially—that is, frequently a set of decisions is made,
one after another, without pausing to evaluate the effect of each decision
along the way, Decisions are also made by different actors, agencies, and
coalitions. The type of group making authoritative decisions can impact pol­
icies. Hermann (2001) proposed a model of foreign policy decision-making
by groups, which can also be used in domestic political contexts, She argued
there are three types of decision-making groups, or units. The predominant
leader group has "a single individual who has the ability to stifle all oppos­
ition and dissent as well as the power to make a decision alone, if necessary”
(p. 56). The single group is a decision unit that includes "a set of individuals,
all of whom are members of a single body, who collectively select a course
of action in consultation with each other” (p. 57). This can be an ad hoc
group set up to respond to a crisis, such as the Office of Homeland Security
established by President Bush after the attacks of September 11, 2001, a
standing bureaucracy (which Homeland Security eventually became), or
an interagency committee. Finally, a coalition of autonomous actors is a
decision unit composed of multiple groups that can act independently. U.S.
trade policy, for example, is affected by a wide variety of domestic and inter­
national interest groups, multilateral organizations, government bureaucra­
cies, and so forth. Each can act independently, and each has some impact at
different times on decisions and policies.
Hermann (2001) maintains that each land of decision-making unit has
different decision processes and different behavioral patterns. The first two
lands of decision units can be analyzed with political psychological concepts.
In the predominant leader unit, the most important factors affecting how
the group behaves and makes decisions are the personality characteristics
of the leader, which are discussed in Chapter 2. The single group pattern is
determined by group psychology, particularly the techniques used by the
group to handle disagreements and conflict in the group. There are three
alternatives: groupthink (discussed in more detail later), wherein groups
attempt to minimize disagreement by promoting conformity; bureaucratic
politics, wherein group members acknowledge that disagreements will
occur and attempt "to resolve the conflict through debate and compromise”
(Hermann, 2001, p. 65); and, finally, the implementation of a social-decision
scheme, discussed earlier.

Individual vs. Group Decision-Making

Evidence indicates that groups are not necessarily better decision-makers
than individuals (Hill, 1982). According to Hastie (1986), whether groups
make better decisions than individuals often depends on the characteristics
of the task. On numerical estimation tasks, for example, group judgments
tend to be a little better than the average individual judgment. However, on
problem-solving tasks, such as logic problems, group solutions tend to be
much better than average individual judgments, but worse than the best
individual judgment. One of the keys to determining the superiority of group
 The Political Psychology of Groups 105

or individual judgments, according to Hastie (1986), is whether the task

involves a demonstrably correct solution. When there is, groups tend to per­
form better than individuals. Some recent research indicates groups make
better decisions than individuals when they have been working together
for a long time, and the task is important to group members (Michaelsen,
Watson, & Black, 1989; Watson, Michaelsen, & Sharp, 1991).
Individual solutions were also compared with group solutions in
brainstorming tasks, which require participants to generate as many
different suggestions as they can. Intuition suggests groups would per­
form better than individuals on brainstorming tasks (more people should
produce more ideas), but research suggests individuals often produce more
and better ideas when working alone than when working in brainstorming
groups (Mullen, Johnson, & Salas, 1991; Taylor, Berry, & Block, 1958).
Several explanations were offered for the failure of groups to perform as
well as individuals on brainstorming tasks. First, when one group member is
speaking, another is prevented from speaking at the same time, which often
causes other group members to forget what they were going to say (Brown
& Paulus, 1996). This situation might lead to a loss of ideas. Second, group
members may have evaluation anxiety and fear their ideas will be ridiculed
by other group members (Camacho & Paulus, 1995). As a consequence, they
might be reluctant to share new ideas.
Consistent with the idea that groups often perform worse than individ­
uals on problem-solving or brainstorming tasks is the notion that groups
often make worse decisions than individuals. In fact, many group decisions
in political history (e.g., the Bay of Pigs, the Vietnam War) suggest groups
often make bad decisions with serious consequences. Researchers iden­
tified several faulty decision-making processes to describe some of
the bad decisions made by groups, including groupthink, new group
syndrome, bureaucratic politics, group polarization, and the escalation
of commitment.

Groupthink
Groupthink refers to an irrational style of thinking that causes group
members to make poor decisions (Janis, 1972). Janis argues that many
major political decisions, such as the Bay of Pigs invasion, the United States'
failure to defend against the attack on Pearl Harbor, the Vietnam War, and
Watergate, provide evidence for groupthink. According to Janis (1982), there
are a number of observable features of these groups proving the existence
of groupthink. First, in all these decision-making groups, group members
felt a strong pressure to conform to the group. There were strong sanctions
for disagreeing with other group members or criticizing their opinions.
Second, self-censorship was present in most of the groups. Although many
group members may have disagreed with the decisions; they felt pressured
to not express these disagreements openly. Third, mindguards in the group
prevented group members from learning of new information that might
disrupt the flow of the group's proceedings. Fourth, there was an apparent
unanimity of opinion. All the group members seemed to agree with one
another. Fifth, illusions of invulnerability allowed group members to feel
106 The Political Psychology of Groups

confident in their decisions. Most group members believed their judgments

could not be wrong. Sixth, illusions of morality prevented group members
from ever questioning the morality of their decisions. Because this was an
elite decision-making group, they believed all their decisions were moral
and justified. Seventh, group members had a biased perception of the other
group. In the Bay of Pigs decision, a decision by the Kennedy adminis­
tration to sponsor a group seeking to overthrow Fidel Castro in Cuba in
1961, members of the President's advisory committee believed Castro was
a weak and evil leader. Derogatory comments about Castro were frequently
voiced during meetings. Finally, many of the decisions made by these groups
represented defective decision-making strategies. Decisions made in group-
think situations are often described as fiascos, blunders, and debacles.
In addition to specifying the characteristics of the group and the group
decision-making process, Janis (1972) specified the causes of group-
think. One cause is cohesiveness. When groups are very cohesive, as was
the case in the Bay of Pigs advisory committee, disagreements are typic­
ally held to a minimum, creating the perfect conditions for faulty decision­
making. Another cause of groupthink is isolation. When groups, such as
the president's advisory committee, discussed top-secret issues, they did so
in isolation, which prevented outsiders from entering the group to review
deliberations. Another cause of groupthink is the presence of a directive
leader, who controls the discussion and can prevent disagreements from
being voiced. Finally, stress can also create symptoms of groupthink.
't Hart (1990, 1994) expanded Janis's concept and found that, in add­
ition to groupthink being a product of high in-group cohesion under
stress, it may also emerge because of anticipatory compliance by group
members seeking to reach decisions they believe will meet the views or
desires of powerful leaders or peers. Further, 't Hart notes that the situ­
ational conditions in which groupthink becomes most likely include situ­
ations of threat and stress (the context emphasized originally by Janis), and
situations perceived by group members as major opportunities requiring
rapid and major commitment to a pet project or policy to achieve major
success (Fuller & Aldag, 1997).
Groupthink received mixed reviews (Levine & Moreland, 1998). Some
studies supported parts of the model. For example, Tetlock, Peterson,
McGuire, Chang, and Feld (1992) analyzed records of 12 different political
decisions and found it was possible to distinguish between groups whose
decisions were indicative of groupthink and those reflecting good decision­
making. However, the research was not especially successful in locating evi­
dence for all the factors thought to cause groupthink. Other work (Aldag &
Fuller, 1993; Fuller & Aldag, 1997) suggested that research failed to provide
convincing support for the existence of groupthink, and the model was an
unnecessary constraint upon researchers seeking to adequately examine
the true dynamics of group decision-making under the conditions explored
by Janis. Indeed, Fuller and Aldag (1997), along with other researchers on
political group dynamics (see't Hart, Stern, & Sundelius, 1997), argue that
scholars should unpack the various component parts from the model and
embrace a wider range of new research and literature on group function
and dynamics that better reflects the behavior of actual political decision
groups.
 The Political Psychology of Groups 107

New Group Syndrome

Another analysis of conformity problems in group decision-making is called
new group syndrome, which is part of a collection of articles seeking to
move beyond groupthink ('t Hart et al., 1997). Stern (1997; see also Stern &
Sundelius, 1994) uses social psychological findings regarding the life-cycle
of groups in a reexamination of the Bay of Pigs disaster, one of Janis's group­
think cases. Group cohesion, norms, status hierarchy, and strength of group
identity all change as the group ages. With good performance, cohesion
increases. With time, the status hierarchies and role responsibilities become
clear and routine. Norms and accepted decision rules are internalized. When
groups are new, Stern (1997) argues, members bring with them extragroup
baggage, in the form of values, beliefs, and past experiences that affect
decision-making. In addition, leaders are particularly important in the early
stages of a group's life, and that is particularly the case when the leader is
the president of the United States. At this stage, leaders can establish roles,
norms, and group decision-making processes that lead to effective and crit­
ical policy option deliberation, rather than group conformity. Some leaders
do this early on, but others do not, which leads to new group syndrome.
When a leader does not establish norms and decision-making patterns,
"there is a serious risk that group interaction will spontaneously evolve in
a fashion leading to excessive degrees of conformity or conflict (an abrupt
shift into the storming stage)" (Stern, 1997, p. 163). In this early forming
stage, the group members are uncertain about how they should behave, are
anxious to do a good job, and therefore are very vulnerable to conformity
pressures if group leaders do not encourage the opposite by establishing
roles, norms, and decision-making procedures. As an explanation of exces­
sive group conformity, new group syndrome differs from groupthink in that
it is not dependent upon situational pressures such as extreme stress. The
phenomenon can occur in any group, in any context.
The Bay of Pigs fiasco showed evidence of new group syndrome. Kennedy
was in office for only four months, and the plan came from the previous
administration. Kennedy was under pressure to do something about Castro,
and the advisory group he used was informal and interagency in nature.
Kennedy campaigned against Republicans, in part on the platform that they
were lackadaisical in confronting communism, and he swept away the pre­
vious administration's policy-making system. Stern describes the group cul­
ture in the decision-making group as follows:
A number of analysts have suggested that a norm of "boldness"
associated with the "New frontier" mentality permeated the
proceedings. Another important norm appears to have been "rally
to the President" when his "project" came under the criticism of
outsiders ... Another apparent norm that proved dysfunctional was
"deference to experts." Finally, an emergent norm of deference to the
leader is noticeable, a norm of which the president himself appears to
have been unaware ... Kennedy, having little previous management
experience, reportedly had a relatively simplistic view of small group
and organizational management. He placed a premium on talent,
believing that this quality was the key to achieving policy and political
108 The Political Psychology of Groups

success. In other words, he believed that it was enough to assemble a

number of talented people, throw them in a room together, and wait
for good things to happen.
(Stern, 1997, pp. 174,177)
What he got instead was failure. The Cuban exiles, sponsored by the United
States to invade Cuba and overthrow Castro, landed in a swamp, the Bay of
Pigs, and were quickly captured by the Cuban military. The popular uprising
against Castro, which they counted on in their plan to overthrow Castro,
never happened.

Bureaucratic Politics
Another set of group-related decision-making problems plaguing polit­
ical decisions consists of bureaucratic politics. Although political systems
differ widely, many political decisions are affected by the interactions of
groups based in governmental bureaucracies. Those groups have differing
perspectives and interests; they see issues and problems differently, and
compete for policy dominance and resources. At the same time, these
groups interact in a variety of policy contexts and need to work together.
Consequently, their interactions are often characterized as "pulling and
hauling”: bargaining, coalition formation, compromise, competition,
and the selective use and sharing of information to enhance the position
of the group or faction in question. The result is policy decision-making
based upon organizational and group interests rather than on an objective
assessment of the policy issue. The often-quoted phrase "Where you stand
depends on where you sit” reflects this pattern.
Early studies of bureaucratic politics focused primarily on the standard
operating procedures and conflicts among bureaucratic groups, both of
which can negatively affect decision-making. The seminal study by Allison
(1971) on the Cuban missile crisis of 1962 illustrated the impact of bureau­
cratic struggles in one of the most dangerous episodes of American foreign
policy, which nearly led to nuclear war. Rather than maintaining a focus on
the national interest, the bureaucracies fought continuously for control of
the policy. In August 1962, there were increasing concerns in the Central
Intelligence Agency (CIA) that the Soviet Union was placing, or would place,
offensive nuclear missiles in Cuba. During the next month, these concerns
spread, and the question of whether to send U-2 spy planes to take pictures
of Cuba in search of missile sites was discussed. This was considered to be
a risky enterprise because of the diplomatic fallout should a U-2 be shot
down. Bickering between the Air Force and CIA over which agency would
get to fly the U-2s over Cuba caused a 10-day delay in spotting the missiles.
Those 10 days were crucial for the installation and arming of the missiles
and made the conflict that followed, between the United States and Soviet
Union, much more dangerous.
Other studies of bureaucratic politics focused more precisely on the
group nature of decision-making in bureaucracies (Preston & 't Hart, 1999;
Stern Sc Sundelius, 1997; Vertzberger, 1990). This enabled analyses of the
whole range of decision-making patterns that can emerge from group
t
t
 The Political Psychology of Groups 109

Optimal process
Bureaucratic Bureaucratic
Consensus­ • confrontation
making
Bureaupolitical <- (1) Reality testing -> Bureaupolitical
oversimplification Bureaupolitical distortion
deliberation
Bureaupolitical <- (2) Acceptability -> Bureaupolitical
isolation Bureaupolitical paralysis
compromise
formation
Bureaupolitical haste (3) Efficiency Bureaupolitical
Bureaupolitical waste
economy
Figure 4*1 Bureaucratic Politics: The Normative Dimensions
interaction in bureaucracies—a range that spans from consensus seeking,
the most extreme form of which is groupthink, to extreme intergroup con­
flict verging on bureaucratic warfare. Consensus and cohesion occur within
groups, particularly when pressured by intense intergroup conflict. Hence,
bureaucracies can be the site of a long continuum of group-produced
behaviors (see Figure 4.1).
For example, Preston and \ Hart (1999) argue that the actual degree to
which bureaucratic politics pervades the policy-making process is variable,
and bureaucratic politics is a continuum in which such dynamics will have
varying degrees of impact (both positive and negative) upon the quality
of the decision-making process. They employ three criteria developed by
George (1980) for evaluating the quality of decision-making on that con­
tinuum. The three criteria are reality testing (Does information get to the
central decision-makers and are multiple options considered?), accept­
ability (Are relevant players involved in the decision-making group and are
they listened to?), and efficiency (What are the costs of the decision-making
process?). Preston an d 't Hart (1999) argue that, at the consensus end of
the spectrum, one sees the decision-making pathologies of bureaupolitical
oversimplification, when reality testing is poor; isolationism, when accept­
ability is poor; and hasty decision-making, when efficiency is poor. On the
extreme conflict end of the scale, one sees bureaupolitical distortion, when
reality testing is poor; paralysis, when acceptability is poor; and waste, when
efficiency is poor.

Manipulation
Manipulation occurs when a group member, often a leader, rigs decision­
making, and may get a group to "accept a commitment which would have
been rejected out of hand had the full implications and full extent of the
project been revealed from the start” (Stern & Sundelius, 1997, p. 131).
Manipulators use at least three strategies: they affect the group's structure
so their allies dominate decision-making; they manipulate the procedures
110 The Political Psychology of Groups

followed by the group by setting the agenda and framing issues in a par­
ticular way; and they manipulate their personal relationships with group
members, both formally and informally, to put themselves in a favorable
position to influence the decision's outcome (Hoyt & Garrison, 1997).

Group Polarization
Groups researchers are interested in whether groups make riskier decisions
than individuals. Janis's (1972) groupthink model suggests groups take
unjustified risky courses of action. Research on the risky shift phenomenon
suggests decisions made by groups are often riskier than those made by
individuals (Stoner, 1961j Wallach, Kogan, & Bern, 1962). But evidence also
exists suggesting that groups sometimes make more cautious decisions than
individuals (Wallach et al., 1962), and some evidence suggests that groups
make both more risky and more cautious decisions (Doise, 1969).
Groups make both very risky and very cautious decisions compared
with individuals. When in a group, there is a tendency to make extreme
decisions. Whether the decision is extremely risky or extremely cautious
depends on what position dominated at the outset of the discussion. Group
polarization refers to the tendency for individuals' opinions to become
more extreme after discussion than before discussion (Myers & Lamm,
1976). For example, if group member A's pre-group discussion opinion
tended to be moderately cautious, then their post-group discussion opinion
would probably be extremely cautious. Likewise, if group member B's pre­
group discussion opinion was moderately risky, it will become even more
risky after group discussion. Although there is a tendency to assume that
extreme decisions, in either direction, are bad decisions, this is not the
case. Extremely risky or cautious decisions can have positive or negative
outcomes.
A number of explanations were offered to account for polarization
effects. One explanation is based on the persuasive arguments perspective,
which assumes people are likely to be exposed to persuasive arguments
that favor their initial position (Burnstein & Vinokur, 1977). Although
group discussions are likely to contain some arguments for and against an
individual's initial position, there is a tendency to sample information that
is consistent with our own point of view. Such biased information sampling
is likely to shift a group member's opinion further in the direction of their
initial position. Additionally, a group member is likely to share their initial
position with the rest of the group. The mere expression and restatement
of ideas may increase the shift toward a more extreme view (Brauer, Judd,
& Gliner, 1995). Those members committed to a more risk-prone decision
may be more committed, more vocal, and hence more influential in per­
suading others. However, as Vertzberger (1997) notes, when more cautious
members are more committed, they can sway the group toward that pole.
Another explanation for group polarization is based on social com­
parison processes. According to this perspective, group members often
compare themselves with others in order to gain approval for their views.
Comparisons with other group members might lead to the realization that
others have similar opinions and still others have more extreme opinions.
 The Political Psychology of Groups 111

Motivated by a need to be viewed positively by other group members, indi­

viduals may shift their opinions to a more extreme position (Brown, 1974;
Myers, 1978). Social comparison processes can be so strong that polariza­
tion is produced by merely knowing of others' positions, in the absence of
exposure to supporting arguments (Isenberg, 1986).
A third explanation for group polarization is based on social identity
processes (Hogg, Turner, & Davidson, 1990; Mackie, 1986). According to
this perspective, group discussion causes individual group members to
focus on the group, which can often lead to pressures toward conformity.
Rather than perceiving the average opinion of the group, individual group
members often perceive the group's opinion to be more extreme. Pressures
to conform lead individuals to adopt a more extreme position than their
initial position.

Escalation of Commitment
In making political decisions, people sometimes decide on a course of
action that proves detrimental to the achievement of their goals. Both indi­
viduals and groups can become overly committed to these failing endeavors.
Situations such as these have been referred to as escalation situations (Staw
& Ross, 1989), or situations in which some course of action has led to losses,
but in which there is a possibility of achieving better outcomes by investing
further time, money, or effort (Brockner, 1992; Staw & Ross, 1987, 1989).
Thus, there is still a glimmer of hope that, by investing additional resources,
the project will become successful. Three characteristics define escalation
situations (Staw & Ross, 1987). First, escalation situations involve some
loss or cost. Second, there must be a lapse in time from the initial decision;
escalation situations do not refer to one-shot decisions, but rather refer to a
series of decisions made over time. Third, withdrawal from the situation is
not obvious or easy. Countless examples of these escalation situations exist
at both the individual and group levels (Ross & Staw, 1986). Individual-level
examples include a person deciding whether to invest more money in a
broken car or in a declining stock.
Decision-making during the Vietnam War illustrates the impact
of commitment. In his memoir, former Secretary of Defense Robert
McNamara reviews the debates within the policy advisory circles of the
Johnson administration. The administration, under the influence of the
domino theory prominent during the Cold War, believed that if South
Vietnam were overtaken by the communist government of North Vietnam,
regimes all over Asia would become communist as well, like dominoes
falling. Yet the government of the South was corrupt and illegitimate, and
the determination of the North, as well as of the Viet Cong (the guerrillas
operating in the South) was clear. Important voices in the military warned
that the administration's hopes in 1964—that the insurgency problem could
be solved by bombing the North, thereby eliminating the need for U.S.
ground troops—would not work. Bombing North Vietnam into oblivion
would still not stop the Viet Cong's efforts to overthrow the government
of South Vietnam. They also knew, from an intelligence report, that the
chances of a stable South Vietnamese government emerging—one with
112 The Political Psychology of Groups

popular support, which could pursue the war on its own terms—was very
unlikely. As McNamara recalls:
These two assessments should have led us to rethink our basic
objective and the likelihood of ever achieving it. We did not do so, in
large part because no one was willing to discuss getting o u t... We ...
wished to do nothing that might lead to a break in the "commitment
dike” as long as there appeared to be some alternative... It is clear
that disengagement was the course we should have chosen. We did
not. Instead we continued to be preoccupied by the question of which
military course to follow*
(McNamara, 1995, pp. 154,164)
They decided to pursue a course that led to a quagmire from which they
could and would not extract themselves, a situation resulting in thousands
of American and Vietnamese casualties, which would have been avoided by
an earlier withdrawal of U.S. military forces.
Project, psychological, social, and organizational factors could all affect
escalation behavior. Project factors are the most obvious determinants of
commitment to a failing course of action. The manner in which a failing
project is structured seems to influence whether an individual or group
withdraws from it or persists. One such factor is whether a setback is the
result of permanent or temporary causes (Leatherwood & Conlon, 1987).
Commitment is more likely to escalate when the setback results from a tem­
porary cause. In the Vietnam case, the focus on military options led the
decision-making group to think a change in military strategy would work,
and their inability to win the war was a temporary result of incorrect military
strategy, rather than the result of permanent irremediable political realities.
Similarly, when future costs required for the project's success are expected
to be small, commitment is more likely to escalate (Brockner, Rubin, &
Lang, 1981). Escalation of commitment can also depend on how often
previous commitments succeeded (Goltz, 1992; Hantula, 1992; McCain,
1986). When previous investments are successful, people are more likely to
escalate their commitment to a project, even when the project is currently
failing. Commitment to a failing project is also more likely to escalate if the
size of the initial investment is relatively large (Teger, 1980). Finally, escal­
ation of commitment is stronger when the size of the payoff from continued
investment is likely to be high (Rubin 8c Brockner, 1975).
There are also several psychological factors that can influence’persistence
in an escalation situation. Information-processing errors, for example, can be
very important. Individuals often misinterpret or seek data in a manner that
supports their beliefs (Frey, 1986), thus strengthening their commitment to
a failing course of action (Bazerman, Beekun, 8c Schoorman, 1982; Caldwell
8c O'Reilly, 1982; Conlon 8c Parks, 1987). A related factor is the type of goal
individuals set before initiating the project. If people do not set explicit goals
about the maximum size of their investment (Kernan 8c Lord, 1989) or the
extent of their commitment (Brockner, Shaw, 8c Rubin, 1979), then they are
likely to escalate their commitment to a failing project. This pattern was also
evident in Vietnam. The escalation of force was gradual and incremental,
with no set limit on size or the point at which there would be an evaluation
•w
 The Political Psychology of Groups 113

of the effort to determine whether it had failed, and no upper limit identified
on how many U.S. troops would be committed.
Self-justification is another psychological factor that has been shown
to influence commitment. Individuals often commit further resources to
a losing course of action to justify previous behavior, such as advocating
the project in the first place (Bazerman et al., 1982; Bazerman, Giuliano,
& Appelman, 1984; Staw, 1976; Staw & Fox, 1977). Other research
suggests conscientiousness also impacts whether individuals escalate their
commitment. When individuals felt a sense of duty, they escalated less
than did individuals motivated by an achievement obligation (Moon, 2001).
Finally, groups whose members identify strongly with their group are more
likely to escalate their commitment to a failing project than groups whose
members identify weakly with a project (Dietz-Uhler, 1996).
Another set of factors that can influence the escalation of commitment
is social in nature. One such factor is the need for external justification.
Individuals or groups may persist in order to save face or avoid losing cred­
ibility with others (Brockner et al., 1981; Fox & Staw, 1979). Another factor
that might influence persistence is external binding, which occurs when
individuals or groups become strongly linked with their actions related to a
project. For example, a project may become so associated with the primary
decision-maker (e.g., Reaganomics) that withdrawal is difficult or impos­
sible (Staw & Ross, 1989). Research on the “hero effect” found that, under
some conditions, people who remain committed to a failing project are
evaluated more favorably than people who withdraw (Staw & Ross, 1989).
Finally, structural or organizational factors can also influence
commitment to a failing project. One such determinant of persistence is
institutional inertia. Because change in an organization (especially a large
organization) is often difficult, it may seem easier to persist in a losing
course of action than to somehow mobilize the organization for change
(Staw & Ross, 1989). Another organizational determinant of persistence is
the operation of political forces. There may be strong political support for
the continuation of a project, even though it is not economically feasible.
Groups that are interdependent or politically aligned with a project may also
demand support for it (Staw & Ross, 1989). Finally, cultural norms can affect
the likelihood of escalating commitment (Geiger, Robertson, & Irwin, 1998;
Greer & Stephens, 2001).
Escalation of commitment to a failing project is a robust phenomenon.
Escalations of commitment were demonstrated in many laboratory and real-
life situations, and many factors were shown to account for the phenom­
enon in such situations; what is so intriguing is that the decisions appear to
be so irrational. From a rational point of view, it often seems the best choice
in these situations is to withdraw and avoid greater losses. However, sev­
eral researchers (Barton, Duchon, & Dunegan, 1989; Beeler, 1998; Beeler &
Hunton, 1997; Bowen, 1987; Northcraft & Neale, 1986; Northcraft & Wolf,
1984) note that decisions to continue investment in a failing project are not
necessarily irrational, at least from the perspective of the decision-maker(s).
For example, Northcraft and Wolf (1984) and Whyte (1986) argue from an
information-processing perspective that the manner in which decisions
are framed determines whether individuals escalate their commitment to
a failing project. If a decision is framed as a certain loss, then people tend
114 The Political Psychology of Groups

to abandon the project. However, if a decision is framed as an attempt to

recoup an investment, then people tend to escalate their commitment to the
project. In escalation situations, decisions are often framed as an attempt to
recoup an investment. Thus, to an outside observer, these decisions seem
to be irrational but, to the decision-maker(s), they can seem quite rational
because of the way in which they are framed.

Improving Group Decisions

Because the decisions made by groups are often disappointing, efforts
have been made to develop techniques to improve groups’ decisions. One
suggested technique is to appoint a group member to serve as a devil's advo­
cate (Hirt & Markman, 1995). The role of the devil’s advocate is to disagree
with and criticize whatever plan is being considered by the group. This tech­
nique can be effective because it encourages group members to think more
carefully about the decisions they are contemplating. A related approach
involves the use of authentic dissent, in which one or more members of
the group will actiyely disagree with the group’s initial plans without being
assigned to this role (Nemeth, Connell, Rogers, & Brown, 2001). This
technique can be effective because it encourages the group to consider
alternatives and often moves the group away from its initial preferences.
A technique relevant to political decision-making is multiple advo­
cacy (George, 1980; George & Stern, 2002). In this process, manipulation is
avoided by having the deliberation procedures managed by a neutral person,
a custodian manager, while the advocates of different positions are allowed
to fully develop their proposals and advocate the advantages. Mutual criti­
cism by the advocates of various proposals should, in theory, flesh out the
strengths and weaknesses of the different policy options. This is done for the
benefit of the final decision-maker, or chief executive (the president, prime
minister, etc.), who listens, evaluates the options, and makes an informed
decision. Many U.S. presidents tried to use this approach for improving
decision-making in their administrations. In fact, the National Security
' Council (NSC), and particularly the national security adviser, evolved into
the role of the custodian manager following its foundation in 1947. As
George (1980) describes it, the NSC took on a number of tasks in its role as
custodian manager:
1 balancing actor resources within the policymaking system
2 strengthening weaker advocates
3 bringing in new advisers to argue for impopular options
4 setting up new channels of information so that the president and other
advisors are not dependent upon a single channel
5 arranging for independent evaluation of decisional premises and options,
when necessary
6 monitoring the workings of the policy-making process to identify pos­
sibly dangerous malfunctions and institute appropriate corrective action.
(pp. 195-196)
Nevertheless, establishing and consistently using a multiple advocacy
system to improve group decision-making is difficult, and many presidents
 The Political Psychology o f Groups 115

fail to keep it alive. First, the custodian manager has to ensure a wide range
of views and proposals are heard and the appropriate people are involved in
deliberations. This is difficult to achieve, particularly given that this role is
typically held by someone from the administration and therefore a person
with their own political perspective and career, subject to pressure from
many different agencies and individuals (’t Hart, 1997). For similar reasons,
it is difficult for the chief executive to use the system. Choosing the best
option is often impossible because of domestic or international political
pressures and obstacles. Finally, some presidents, such as Ronald Reagan,
do not want to hear debate and discussion of multiple options.

CONFLICT IN GROUPS
When people work together to achieve a goal, there will inevitably be some
conflict, which occurs when group members believe their goals are not
compatible (Pruitt & Rubin, 1986). Group members can conflict with one
another in many ways. For example, conflict can arise if group members
have to compete for scarce resources. Groups can also experience conflict
when one member tries to exert influence or gain prestige in the group
(Levine & Moreland, 1998). In this section, we examine the various types of
conflict that can exist in a group, particularly in situations in which group
members are motivated to both compete and cooperate. A discussion of the
causes of conflict [Link] follows; we then briefly examine the formation
of coalitions in groups. Finally, we examine strategies designed to reduce
conflict in groups.

Types of Conflict: Social Dilemmas

Much of the research on conflict in groups examines mixed-motive situ­
ations in which the motivation to compete is mixed with the motivation to
cooperate. Perhaps the most famous mixed-motive game is the prisoner’s
dilemma (Luce & Raiffa, 1957; see Figure 4.2). Research of this type is used
to determine how tendencies to cooperate and compete can lead to various
outcomes for groups.
In this game, participants cannot communicate with one another, yet
the outcome of the game for each person is contingent on what the other
person decides. The game is set up so that; (1) if both players cooperate,
they receive a moderately favorable outcome; (2) if one cooperates
and the other competes, the cooperator receives an unfavorable outcome
and the competitor receives a favorable outcome; and (3) if both players
compete, they receive a moderately unfavorable outcome. In this situation,
the dilemma is whether to compete or cooperate. The situation is rigged
so both players benefit equally if they cooperate, but there is a tendency
to not trust the other player, so many people compete (Pruitt & Kimmell,
1977). More recently, research on mixed-motive interactions used an N-
person social dilemma, a social dilemma with more than two people (Levine
& Moreland, 1998). In these dilemmas, several outcomes are possible. First,
a player always benefits more from a noncooperative than a cooperative
choice. Second, a noncooperative choice is harmful to others in the group.
116 The Political Psychology of Groups

A Third, the amount of harm to others that results

from a noncooperative choice is larger than the
profit received as a result of any choice.
Both get A goes free There are several types of social dilemmas
B
Don’t confess 1 year
Bgets (Messick & Brewer, 1983). In a collective
15 years trap, behaviors that reward individual group
members can be harmful to the rest of the
group, especially if engaged in by enough
group members. For example, during a water
Confess A gets Both get shortage, individuals who use too much water
15 years 10 years harm everyone else by prolonging the shortage.
B goes free The best strategy for the collective is if each
individual takes a little. In collective fences, the
entire group is harmed if enough people avoid
Figure 4,2 The Prisoner's Dilemma. In this classic costly behaviors. For example, if each person
game, two prisoners, A and B, both of whom are does not donate money to medical research,
accused of a crime, have the options of confessing then everyone will be worse off. The best
or not confessing. If they maintain their alliance and strategy is for everyone to give a little. In either
neither confesses, both get short sentences. If each of situation, people are tempted to "free ride,” or
them confesses, they each get a heavy sentence. But if enjoy the group's resources without penalty.
one confesses and the other does not, the prisoner who Research using collective traps and collective
confesses is rewarded with freedom, while the one who fences can tell us much about human tenden­
does not confess gets a severely heavy sentence. The cies to be selfish or prosocial, as well as how a
dilemma for each prisoner is that, if he or she trusts the person's value orientation (e.g., cooperative or
other not to confess, the best option is to rat out his or competitive) can influence their behavior in
her partner in crime. social dilemmas.

Causes of Conflict
Conflicts such as social dilemmas typically arise when group members
have competing goals or see their goals as being incompatible. There are
many factors that can contribute to the origination of conflict, as well as
to its escalation. In the previous chapter, the concept of attributions was
introduced. They play a role in group conflict, as well as in individual
perceptions. Attributions refer to the explanations generated for the causes
of our own and others' behavior. Imagine playing a prisoner's dilemma game
in which you realize that the best strategy is for you to cooperate, but your
partner always seems to make the competitive choice. Why? You may gen­
erate many reasons why your partner makes the competitive choice: perhaps
he does not understand the game; perhaps he was told by the experimenter
to make consistently competitive choices; maybe he is just an evil person.
Attributing the cause of another's behavior to dispositional, rather than
situational, factors is the fundamental attribution error (Ross, 1977). If we
blame another group member for conflict, it is likely to escalate rather than
be resolved (Forsyth, 1990). Thus, if you blame your partner's personality
for the competitive choices they make, then you are likely to also make com"
petitive choices. People also tend to perceive their own views as correct
and objective, but to perceive others' views as biased (Keltner & Robinson,
1997; Robinson, Keltner, Ward, & Ross, 1995). A consequence of this bias in
 The Political Psychology o f Groups 117

perception is that we are likely to exaggerate the difference in perspective

between ourselves and another group member, which is likely to serve as
fuel for conflict.
Second, when in potential conflict situations, communicating effectively
can be difficult. Sometimes group members criticize one another harshly.
If you have ever been on the receiving end of harsh criticism, then you
realize it can be unpleasant and uncomfortable. Such discomfort can often
instigate revenge, which only serves to escalate the conflict (Cropanzano,
1993). If group members do not communicate reasonably and effectively,
then conflict will likely occur and may even be escalated. One particu­
larly destructive variant of faulty communication is nay-saying, a pattern
in which group discussions are crippled and paralyzed by negativism and
bickering over everything, down to the smallest details of a decision (Stern
& Sundelius, 1997). Whenever conflict becomes stronger, so do anxiety and
tension (Blascovich, Nash, & Ginsburg, 1978; van Egeren, 1979). According
to the arousal/aggression hypothesis (Berkowitz, 1989), group members
become frustrated when they are unable to attain their goals. Frustration
can lead to aggression, which is often displayed by lashing out at other group
members. If group members are aggressive, then conflict occurs and prob­
ably escalates.
Third, and more recent, are theories and research about the gendered
nature of group conflict. Much of this thinking is based on principles of
evolutionary psychology, which we discussed in earlier chapters. The pri­
mary idea of the gendered nature of group conflict is the experience and
reasons for group conflict differ for males and females, and they do so
because of different reproductive motivations for males and females. One
of the hypotheses derived from this perspective is referred to as the "out­
group male target hypothesis” (Sidanius & Kurzba, 2013). According to
this hypothesis, in-group bias is more prominent for males than females. In
other words, discrimination against out-group males is more pervasive and
severe than discrimination against out-group females. For example, arch­
ival evidence shows Black males are six times more likely to be imprisoned
than White males, while the disparity in imprisonment rates for Black and
White females is about half of that rate (McDonald, Navarrete, & Sidanius,
2011). This and other evidence (see Sidanius & Kurzba, 2013 for a review)
suggests that much larger gaps exist in discrimination against male than
female out-groups.
Of course, this disparity in out-group discrimination against males and
females begs the question "Why?” The answer lies in evolutionary psych­
ology. Male aggression against out-group males appears to be motivated by
aggressive tendencies and the desire to gather resources and status to pre­
sent as ideal mates. The picture for females, however, is quite a bit different.
Female motivation to discriminate against the out-group is rooted primarily
in fear of sexual coercion and assault. An interesting study that led to this
conclusion found females displayed more bias against out-group males
when their risk of conceiving was at its highest level (Navarrete et al„ 2009).
These ideas are consistent with the male warrior hypothesis that males
form coalitions that allow them to plan and execute out-group aggression.
McDonald et al. (2011) found males have a stronger desire than females to
control and dominate out-groups, while female out-group bias is predicted
118 The Political Psychology of Groups

primarily from their self-assessed vulnerability to sexual coercion. There

seems to be ample support for the male warrior hypothesis, including
findings that: (1) men discriminate against out-groups more than women;
(2) men prefer group-based hierarchies more than do women; (3) males are
more likely than females to defend their in-group; and (4) males are more
likely than females to engage in intergroup competition and aggression
(McDonald, Navarrete, & Van Vugt, 2012). Consistent with the male warrior
hypothesis, the rationale for this hypothesis lies in the idea that group con­
flict affords males the opportunity to attract reproductive mates, property,
and status, while for females, group conflict is largely avoided because it
increases the opportunity and risk of being sexually assaulted by out-group
males (McDonald, Navarrete, & Van Vugt, 2012).
Finally, in the review of the research on escalation of commitment, we
learned that group members easily become committed to a course of action,
even if it is a failing one. Group members can also become committed to
their viewpoints, especially when they are under attack (Staw & Ross, 1987).
There are several reasons. First, we tend to seek information to confirm,
rather than refute, our beliefs (Petty & Cacioppo, 1986). Such action tends to
make us even more committed to our beliefs. Second, in a public situation,
there is often a desire to appear strong and have conviction in our beliefs.
Third, once an individual commits to a belief, they rationalize their choice
by overestimating its favorableness and increasing their dedication to it
(Batson, 1975). Fourth, attacks from other group members can create react­
ance (Brehm, 1976), which occurs whenever we sense a loss of freedom.
The consequence is that we become even more committed to our belief or
position.

Coalitions
Conflicts exist between more than two group members. Sometimes, group
members persuade other group members to join forces by forming a
coalition, a small collection of group members who cooperate in order to
achieve a mutually desired goal. Coalitions have a number of characteristics
in common (Forsyth, 1990). First, they all typically involve group members
who disagree on fundamental issues, but decide to set aside those
differences and focus on the problem at hand. Second, they form for the
purpose of achieving certain goals. Third, coalitions tend to be temporary,
and there is often little commitment on the part of the participants, except
to the current goal. Fourth, coalitions typically form in mixed motive situ­
ations, so group members who formerly competed with one another must
cooperate to achieve the current goal. Fifth, coalitions are adversaries. The
goal is to make sure that, in the end, they are better off and another coali­
tion is worse off.
There are a number of theories put forth to explain when and why cer­
tain coalitions are likely to form. According to minimum-resource theory
(Gamson, 1961, 1964), group members form coalitions on the basis of
equal input-equal output; the most likely grouping of people is one that
 The Political Psychology of Groups 119

involves the fewest number of people with the fewest number of resources,
yet is most likely to win. The theory makes two assumptions: (1) people in
groups are primarily motivated by the need to maximize power and payoffs
and believe forming coalitions will satisfy this goal; and (2) members of
coalitions believe the distribution of power and rewards should be divided
equally among the members.
Another theory that explains when and why coalitions form is
minimum-power theory (Shapley, 1953). According to this theory, coali­
tion members expect payoffs directly proportional to their ability to turn a
losing coalition into a winning one. This type of power is referred to as piv­
otal power (Miller, 1980). In this theory, power, not resources, is the most
important determinant of coalition formation, The pivotal power of any
group member is determined by the number of times a member could turn
a winning coalition into a losing one by withdrawing from the coalition.
Thus, coalitions form on the basis of the highest chances of winning with
the lowest amount of pivotal power.
According to bargaining theory (Komorita & Nagao, 1983), coalitions
form on the basis of considering expected payoffs, which are based on
norms of equity and equality, and group members will appeal to whichever
norm provides them with the largest payoff. This theory assumes group
members prefer to form coalitions with those who will not withdraw. It
also assumes the amount of payoff may change over time, to compensate
for extra rewards given to coalition members who are being tempted to join
another coalition.
In addition to these theories of coalition formation, research identi­
fied other factors that influence the formation of coalitions, including the
number and size of existing coalitions (Komorita & Miller, 1986; Kravitz,
1987), expectations of each group member in forming coalitions (Miller &
Komorita, 1986), and the availability of other influence strategies that do not
require the formation of coalitions (Komorita, Hamilton, & Kravitz, 1984).

CONCLUSION
This chapter has reviewed some of the central findings from psychological
research on groups and their behavior. We also reviewed some of the key
patterns of group behavior in politics, discussed how and why groups form,
how they make decisions, and what problems arise in group decision­
making. We examined intra- and intergroup conflict dynamics, and some
techniques for conflict resolution. Several of the chapters that follow pro­
vide additional information and illustrations of group behavior. Chapter
5 provides examples of small-grpup dynamics in leadership management
styles. Chapters 8,9, and 10 provide examples of group behavior in cases of
race, ethnic, and nationalist group conflicts. Chapters 9 and 12 look at the
behavior of extremist groups, such as terrorist organizations, perpetrators
of genocide, and others. They provide several illustrations of obedience to,
and compliance with, group demands.
120 The Political Psychology of Groups

Topics, Theories/Explanations, and Concepts in Chapter 4

Topics Theories/Explanations Concepts
Definition of groups; Entiativity
central characteristics
size, composition,
type
Group structure: Expectation states
status, roles, norms, theory; ethological
cohesion; group theories; functional
formation perspective;
interpersonal
attraction
perspective
Group development Stages: forming,
storming,
norming,
performing,
adjourning
Influence in groups Conformity
Conformity Informational social
influence; normative
social influence
Situational conformity Group size; group
unanimity;
commitment to the
group; individuation
Power: reward, coercive,
legitimate, referent,
expert; minority
influence
Reaction to power: Complementarity
compliance, hypothesis
attraction, conflict,
rebellion, motivation,
self-blame
Group performance; Arousal; evaluation
social facilitation and apprehension;
inhibition distraction
Productivity losses; group Social loafing; three
decision-making stage model
Groups and political Predominant leader;
decision-making units single group;
coalitions
{Continued)
 The Political Psychology of Groups 121

(Continued)
Topics, Iheories/Explanations, and Concepts in Chapter 4
Group decision-making Groupthink; new
group syndrome;
bureaucratic politics;
manipulation;
group polarization;
escalation of
commitment .
Improving group Social dilemmas;
decisions; conflict in collective traps;
groups collective fences
Causes of conflict Faulty attributions;
faidty
communications;
biased perceptions;
personality;
commitment;
arousal and
aggression *
Coalitions Minimum resource
theory; minimum
* power theory;
bargaining theory

Key Terms
adjourning forming
attributions Fundamental Interpersonal Relations
autokinetic effect Orientation (FIRO)
bargaining theory group
coalition group development
coalition of autonomous actors group polarization
coercive power groupthink
cohesion individuation
collective fences informational social influence
collective trap legitimate power
common-bond groups minimum-power theory
common-identity groups minimum-resource theory
compliance motivation
conformity multiple advocacy
critical bases of power normative social influence
deindividuation norming
entiativity norms
escalation of commitment performing
expected payoffs predominant leader
expert power prisoner's dilemma
122 The Political Psychology of Groups

rebellion social loafing

referent power status
reward power storming
roles three-stage model of group
single group decision-making
social-decision schemes

Suggestions for Further Reading

Brown, R. (2010s). Prejudice: Its social psychology. Oxford: John Wiley & Sons.
Forsyth, D. R. (1990). Group dynamics. Pacific Grove, CA: Brooks/Cole.
George, A., & Stern, E. (2002). Harnessing conflict in foreign policy making:
From devil's advocate to multiple advocacy. Presidential Studies Quarterly,
32,484-508.
Janis, I. L. (1982). Groupthink: Psychological studies o f policy decisions and
hascoes (2nd ed.). Boston, MA: Houghton Mifflin.
LeBon, G. (1960 [1895]). The crowd: A study of the popular mind. New York:
Viking.
Levine, J. M., & Moreland, R. L. (1998). Small groups. In D.T. Gilbert, S.T. Fiske,
& G. Lindzey (Eds.), The handbook o f social psychology (4nd ed„ pp. 4 1 5 -
469). New York: McGraw-Hill.
Moscovici, S., & Doise, S. (1994). Conflict and consensus: A general theory of
collective decisions. London: Sage.
Prentice, D., & Miller, D. (Eds.). (1999). Cultural divides: Understanding and
overcoming group conflict. New York; Sage.
Preston, T., & 't Hart, P. (1999). Understanding and evaluating bureaucratic
politics: The nexus between political leaders and advisory systems.
Political Psychology, 20, 49-98.
't Hart, P. (1994 [1990]). Groupthink in government: A study o f small groups and
policy failure. Baltimore, MD: Johns Hopkins University Press.
't Hart, P., Stern, E. K., & Sundelius, B. (1997). Beyond groupthink: Political
group dynamics and foreign policy-making. Ann Arbor, Ml: University of
Michigan Press.