THE INFLUENCE OF DEMOGRAPHIC HETEROGENEITY ON
THE EMERGENCE AND CONSEQUENCES OF COOPERATIVE
NORMS IN WORK TEAMS
Drawing from social categorization theory, we found that greater demographic heterogeneity
led to group norms emphasizing lower cooperation among student teams and
officers from ten business units of a financial services firm. This effect faded over time.
Perceptions of team norms among those more demographically different from their
work group changed more, becoming more cooperative, as a function of contact with
other members. Finally, cooperative norms mediated the relationship between group
composition and work outcomes.
Increased demographic heterogeneity in organizations
has been expected to generate important
benefits, such as increasing the variance in perspectives
and approaches to work brought by
members of different identity groups. Given the
purported advantages, managers might eagerly
incorporate workforce diversity into organizational
problem-solving processes. Yet attempts to
capitalize on these advantages have met with
mixed success (e.g., Heilman, 1994). Likewise,
research on the effects of demographic heterogeneity
in organizational settings has been characterized
by mixed findings, leading researchers
to conclude that, in spite of the recent popularity
of demographic heterogeneity as a topic, there is
little consensus about either what constitutes diversity
or how it affects performance (Guzzo &
Dickson, 1996: 331).
We address these inconsistencies by suggesting
that past researchers have neglected to consider
whether demographic heterogeneity among work
group members led to the emergence of certain
norms that subsequently influenced work processes
and outcomes. Drawing on self-categorization
theory, we begin by exploring how demographic
heterogeneity influences the emergence
and stability of a group's emphasis on cooperative
norms. We then consider the relative impacts of
increased contact on cooperative norms for demographically
similar and different people. Finally,
we examine how inconsistencies in the relationship
between group heterogeneity and work outcomes
might be explained by considering the mediating
role of norms. Thus, this study may explain
the contradictory findings described above; the
negative effects of demographic heterogeneity may
diminish when norms that encourage a focus on
interdependent objectives develop.
THEORY AND HYPOTHESES
Group Norms: The Relative Emphasis on
Cooperation
Group norms, defined as legitimate, socially
shared standards against which the appropriateness
of behavior can be evaluated (Birenbaum &
Sagarin, 1976), influence how a group's members
perceive and interact with one another, approach
decisions, and solve problems. Norms are regular
behavior patterns that are relatively stable and expected
by group members (Bettenhausen & Murnighan,
1991: 21). Cooperative group norms, in particular,
reflect the degree of importance people
place on their personal interests and shared pursuits
(Wagner, 1995: 153), shared objectives, mutual
interests, and commonalties among members.
Emphasizing independence, rather than cooperation,
causes people to differentiate themselves from
others and focus on their own and others' unique
interests, abilities, and characteristics.
Group cooperation may be dictated by the characteristics
of a group's task but, more typically, a
work team's objectives are specified at its incepWe
thank Charles O'Reilly, Anne Tsui, Tom Lee, and
our AMJ reviewers for constructive comments on this
article. We thank the Citigroup Behavioral Science Research
Council and the Institute of Industrial Relations at
the University of California, Berkeley, for their financial
support.
956
2001 Chatman and Flynn 957
tion, and the means of accomplishing those objectives
are left to the team's discretion (Hackman,
1987). Further, outcome interdependence is distinct
from task interdependence (Wageman, 1995).
Thus, even when incentives and rewards are allocated
to a group, variations are likely to emerge in
the level of interdependence members exhibit in
accomplishing the task.
A Self-Categorization Approach to
Understanding the Effects of Demographic
Composition on the Emergence and Stability
of Cooperative Norms
Much is known about the behavioral consequences
of cooperative orientations, but researchers
know relatively little about the factors that influence
the emergence of cooperative norms. Given
the impact of cooperative orientations on processes
and outcomes in organizations and the importance
of matching a group's orientation to its task (e.g.,
Ancona & Caldwell, 1992), understanding the
emergence and stability of such norms over time is
critical. Examining group composition at the time
groups form and how members categorize themselves
and other members on the basis of their
demographic differences may shed light on variations
in cooperative orientations in different groups
at different times.
Self-categorization is the process by which people
define their self-concepts in terms of membership
in social groups. Self-concepts are activated
and provoke specific behaviors depending on the
characteristics of the others who are present in a
situation (e.g., Markus & Cross, 1990). People often
use immediately apparent physical features, such
as race, sex, and national origin, to categorize others
and predict their behavior. Further, members of
demographically heterogeneous groups are more
likely to categorize one another in terms of demographic
characteristics than are members of homogeneous
groups (Stroessner, 1996).
The principle of functional antagonism describes
an inverse relationship between the salience of different
social categories: as one category becomes
more salient, others become less salient (e.g.,
Turner, Oakes, Haslam, & McGarty, 1994). This
principle implies that when demography is salient,
a group of people will focus more on their differences
than on their similarities; that is, they will be
less likely to acknowledge and act in accordance
with factors that tie them together. Research has
shown that demographic heterogeneity within
work groups is inversely related to members' focus
on organizational objectives (Chatman, Polzer, Barsade,
& Neale, 1998). We suggest that this focus on
differences will lead to the formation of norms that
highlight individual members' interests; such
norms would be independent, rather than interdependent,
norms.
One of the few studies that focused on norm
formation in work groups showed that their norms
formed early, often before groups' members adequately
understood their tasks (Bettenhausen &
Murnighan, 1985). Norms were subject to modification
over time, however. As group members interacted,
shared experiences formed the basis for
norms governing future interactions. Interestingly,
demographic heterogeneity may influence the stability
of cooperative norms as "social targets initially
activate primary or primitive generic categories
such as race, gender, and age" (Messick &
Mackie, 1989: 54; emphasis added). A negative relationship
between demographic heterogeneity and
cooperative norms may be most pronounced early
in a group member's tenure or in a group's formation.
It is during this early period that people will
have the fewest alternative, potentially competing,
categories on which to focus, given a lack of prior
knowledge about their colleagues (Brewer & Miller,
1984). The negative influence of demographic differences
on group members may weaken over time
as other social categories, which were not immediately
apparent, surface (Harrison, Price, & Bell,
1998).
The dominant paradigm in demography research,
similarity-attraction theory, cannot account
for such temporal changes in the demographybehavior
relationship. The similarity-attraction
model implies that a stable relationship exists between
demographic characteristics and behavior
because similarity remains constant (e.g., Pfeffer,
1983). Social categorization theory, in contrast,
provides a more dynamic explanation in which it is
recognized that attention to specific characteristics
in a given situation may change over time. At first,
demographically different team members may be
hesitant to cooperate with one another because
they categorize each other as out-group members.
However, if the salience of surface-level demographic
characteristics dissipates over time and demographically
dissimilar group members begin to
recategorize themselves as fellow in-group members,
they may be more inclined to cooperate with
one another. Drawing on social categorization theory
and invoking the functional antagonism principle,
we therefore predict the following:1
1 Predictions at both the individual and group levels of
analysis are relevant because a single member's perceptions
of group norms may depend on the extent to which
958 Academy of Management Journal October
Hypothesis la. The negative relationship between
being demographically different from
the other members of a group and perceptions
that group norms emphasize cooperation will
be strongest for members who are new to the
group.
Hypothesis lb. Group heterogeneity will be
most negatively related to cooperative norms
early in a group's existence.
Past researchers have found that recategorization
was facilitated by increased contact among group
members (Sherif, Harvey, White, Hood, & Sherif,
1954). Simple contact between people with different
backgrounds, here based on demographics, may
not, however, be enough to reduce biases or increase
trust. To induce group members' recategorization
of different people into a common in-group
identity, the contact situation must reflect certain
conditions, including, most importantly, an objective
that makes members' shared fate salient
(Dovidio, Gaertner, & Validzic, 1998). This should
influence members to perceive themselves as one
superordinate group rather than as individuals differentiated
by demographic characteristics. Interaction
under such conditions