Sage Publications, Inc. and Johnson Graduate School of Management, Cornell University
are collaborating with JSTOR
to digitize, preserve and extend access to
Administrative Science Quarterly.
http://www.jstor.org
Being Different Yet Feeling Similar: The Influence of Demographic Composition and
Organizational Culture on Work Processes and Outcomes
Author(s): Jennifer A. Chatman, Jeffrey T. Polzer, Sigal G. Barsade and Margaret A. Neale
Source:
Administrative Science Quarterly,
Vol. 43, No. 4 (Dec., 1998), pp. 749-780
Published by: on behalf of the Sage Publications, Inc. Johnson Graduate School of Management,
Cornell University
Stable URL: http://www.jstor.org/stable/2393615
Accessed: 26-02-2015 20:52 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
This content downloaded from 128.83.205.78 on Thu, 26 Feb 2015 20:52:52 UTC
All use subject to JSTOR Terms and Conditions
Being Different Yet
Feeling Similar: The
Influence of
Demographic
Composition and
Organizational
Culture
on Work Processes
and Outcomes
Jennifer
A.
Chatman
University
of
California,
Berkeley
Jeffrey
T.
Polzer
University
of
Texas,
Austin
Sigal
G.
Barsade
Yale
University
Margaret A. Neale
Stanford
University
?
1998
by
Cornell
University.
0001-8392/98/4304-0749/$1
.00.
This research was
supported by
a
Center
for Creative
Leadership grant
to the
first
author.
We
thank Dan
Brass,
Ben
Herma-
lin, Rod Kramer, and
three
anonymous
reviewers for
helpful
comments
on
ear-
lier
drafts
of the
paper,
and Linda Johan-
son for
working
her
usual editorial
magic.
We
also thank Zoe
Barsness,
Joe Bau-
mann, Mary Cusack,
Brenda
Ellington,
Tiffany Galvin,
Anne
Lytle,
Ann
Tenbrun-
sel,
Melissa
Thomas-Hunt,
and
Kim
Wade-Benzoni
for
help
in
administering
the study.
Drawing from self-categorization theory,
we tested hy-
potheses on the effects of an organization's
demographic
composition and cultural emphasis on
work processes
and outcomes. Using an organizational
simulation, we
found that
the
extent to which an organization
empha-
sized individualistic or collectivistic values
interacted with
demographic composition
to influence social interaction,
conflict, productivity,
and
perceptions
of creativity
among 258
MBA
students. Our findings
suggest
that
the
purported benefits of demographic
diversity are more
likely to emerge
in
organizations that,
through their cul-
ture, make organizational membership
salient and en-
courage people
to
categorize
one another as having
the
organization's interests
in
common,
rather than those
that
emphasize
individualism
and
distinctiveness among
members.'
Research generated
from
a variety of fields
predicts that im-
portant benefits
will
accrue from
demographic
heterogeneity
in
organizations by increasing the variance
in
perspectives
and
approaches
to work that
members
of different
identity
groups can bring (e.g.,
Thomas and
Ely,
1996). For example,
cognitive
and
experiential diversity may
add
to
the
perspec-
tives
available
within
an
organization
and facilitate clarifying,
organizing, and combining novel approaches
to
accomplish-
ing goals
(e.g.,
Hoffman and Maier, 1961).
Further,
work
units
composed
of
diverse members can
tap
into
broad net-
works
of
contacts, making
it
likely
that useful
new informa-
tion
will
be incorporated
into
decisions,
which can increase
commitment
to
choices
and
enhance
responsiveness
to
rap-
idly changing organizational
environments (Tushman, 1977;
Donnellon, 1993).
Managers' attempts
to
capitalize
on
these
advantages
have
met
with mixed success
(e.g., Heilman,
1994). Research
on
the effects
of
demographic diversity
in
organizations
has also
been characterized
by
mixed
findings (Watson,
Kumar,
and
Michaelsen, 1993), leading
Guzzo
and
Dickson
(1996: 331)
to
conclude
that,
in
spite
of
its recent
popularity,
there
is little
consensus about either what constitutes
diversity
or
how
it
affects
performance. According
to
some
researchers,
demo-
graphic heterogeneity
seems to
be
beneficial, supporting
a
"value-in-diversity" hypothesis (Cox,
Lobel,
and
McLeod,
1991: 827);
but
others
have found
diversity
to
be detrimental
to
work
effectiveness
(see
Williams
and
O'Reilly, 1998,
for
a
review).
For
example, heterogeneous
work
groups
have
been found
to be
less
socially integrated
and
to
have
experi-
enced
more
communication
problems,
more
conflict,
and
higher
turnover
rates
than
homogeneous
groups (O'Reilly,
Caldwell,
and
Barnett, 1989; Zenger
and
Lawrence, 1989).
Further, people
who were
more different
from their cowork-
ers
in
terms
of
age, tenure, education,
sex,
and
race
have
reported feeling
more uncomfortable
and less
attached
to
their
employing organization (Tsui, Egan,
and
O'Reilly, 1992).
We
attempt
to
resolve some
of these
contradictions
by
con-
sidering
how
differences
in
organizational
cultures moderate
how
demographically
diverse
people approach
and
solve
problems.
An
organization's
relative focus
on
individualism
or
collectivism
may
affect
the salience
of
organizational
mem-
749/Administrative Science Quarterly, 43 (1998):
749-780
This content downloaded from 128.83.205.78 on Thu, 26 Feb 2015 20:52:52 UTC
All use subject to JSTOR Terms and Conditions
bership
as
a social
category
and, as
a result,
coworkers'
per-
ceptions
of having
interchangeable
interests
and
a common
fate. We
begin
by exploring
how being
demographically
simi-
lar
to or
different
from
coworkers
and an
organization's
em-
phasis
on
individualism
or collectivism
uniquely
and jointly
influence
the
self-categorization
process.
Self-Categorization,
Demographic
Composition,
and
Individualistic
Versus
Collectivistic
Values
Self-categorization
refers
to
the process
by which
people
define
their self-concept
in terms
of
their
memberships
in
various
social
groups.
It can
be situationally
based,
because
different
aspects
of a
person's
self-concept
may become
salient
in
response
to
the
distribution
of
characteristics
of
others
who are
present
in a
situation
(Markus
and Cross,
1990).
A
salient
social
category is defined
as
one that
func-
tions
psychologically
to
influence
a
person's
perception
and
behavior
and how
others treat
the focal
individual (Turner
et
al.,
1987).
To the
extent
that
a particular
in-group
member-
ship is
salient,
one's
perceived
similarity
to
others
in the
in-
group
is increased
(Brewer,
1979).
Increasing
the salience
of
in-group
membership
causes
a depersonalization
of the
self,
defined
as perceiving
oneself
as an interchangeable
exem-
plar
of the
social
category
(Turner,
1985: 99).
Members
of a
salient
in-group
are
more
likely
to
cooperate
with in-group
members
and
to compete
against
out-group
members (e.g.,
Wagner,
Lampen,
and
Syllwasschy,
1986).
An
organization
relies
on members
to cooperate
with one
another
in
accomplishing
goals
to enhance its
survival
(Si-
mon,
1976: 72).
Since
one way
to enhance cooperation
is
to
increase
the extent
to
which
members
view
one another as
part
of their
in-group,
identifying
the factors
that
cause
people
to
categorize
some
people
as
in-group
members
and
others
as
out-group
members
becomes important.
Two of
these factors
may
be
the
presence
of
demographic
diversity
among
employees
and
an
organization's
cultural emphasis
on
individualism
or
collectivism.
Basing
salient
social categories
on
demographic
attrib-
utes. People
frequently
use demographic
characteristics
to
categorize
others
and
predict
their
likely
behaviors.
Allport
(1954)
observed
that
immediately
apparent
physical
features,
such as race
and
sex,
are
widely
used
to
form
impressions
of
others (see
also,
Messick and
Mackie,
1989).
This
ten-
dency
may
be
even more
pronounced
when
demographic
diversity
is
historically
or
normatively
uncommon
in a situa-
tion,
because novel,
infrequent,
or distinctive
stimuli are
likely
to increase
the salience
of
the
particular
category
that
the
stimuli
represents
(Kanter,
1977;
Taylor
and
Fiske,
1978).
Demographic
attributes
also tend
to be
used as a basis
for
social
categorization
because
demographically
similar
people
are
likely
to
share
similar
backgrounds
and
experiences.
They
are
more
likely
to have been
treated
similarly
by
others
in
the
past
and
may
therefore
expect
one
another to
under-
stand and
react
to
situations
similarly (e.g.,
Pfeffer,
1983).
As
a
result,
demographic
attributes
are often
assumed to
be
associated
with
underlying
attributes,
such
as
values, cogni-
tive
styles,
or
past
experience.
This
implies
that
people
will
be
more likely
to use demographic
attributes
as social
cat-
750/ASQ,
December
1998
This content downloaded from 128.83.205.78 on Thu, 26 Feb 2015 20:52:52 UTC
All use subject to JSTOR Terms and Conditions
Being Different
egories
when they are
demographically
different from
others
in
a situation
and
when that
situation
has
typically not
been
characterized
by
demographic
heterogeneity in
the
past, as
in
most U.S.
business
organizations.
We
therefore
propose:
Proposition 1:
People
who are
more
demographically
different
from
their coworkers will
view
demographic
attributes as
more sa-
lient than
will
people who are
more similar
to one
another.
Basing
social
categories
on
an
organization's
emphasis
on
individualism or
collectivism.
An
organization's
culture
may also influence
which
of
its
members' social
categories
are
activated.
Organizational
culture,
defined
as the observ-
able norms
and
values that characterize an
organization,
in-
fluences which
aspects of its
operations and
its
members
become
salient
and how
members
perceive and
interact
with
one
another,
approach
decisions,
and
solve
problems
(Trice and
Beyer,
1993;
O'Reilly and
Chatman,
1996).
The
organizational culture's
emphasis
on
either
individualism or
collectivism
may
particularly
influence
the social
categoriza-
tion
process.
Research has shown
that the
individualism-collectivism
di-
mension, while often
examined at the
societal
level, is
also
central to
characterizing
how work is
conducted at the
orga-
nizational level
(Earley, 1993;
Chatman
and
Barsade,
1995).
Individualistic
cultures focus
on and
reward
achievements
that
can
be
attributed
to
a
particular
person.
Emphasizing
individualism
causes
people to focus on
their own and oth-
ers'
unique
abilities and
characteristics-on
what
differenti-
ates them
from
others. In
contrast,
the focus
in
collectivistic
cultures is
on
shared
objectives,
interchangeable
interests,
and
commonalties
among
members.
Members of
collectivis-
tic
cultures are
more
likely
than those
in
individualistic cul-
tures to
agree about
what
constitutes
correct
action,
behave
according
to
the
norms of
the
culture,
and suffer
or offer
severe
criticism
for
even
slight
deviations from
norms
(Trian-
dis,
1995:
52).
Because of
the
heightened
priority placed
on
interdependence
and
cooperation,
people
in
collectivistic
or-
ganizational
cultures
may
be
more
likely
to
use
organizational
membership
as a
basis
for
social
categorization.
In
these col-
lectivistically oriented
organizations,
simply
being
a member
of
the
organization
may
be
enough
to
qualify
one
as an
in-
group
member
(e.g.,
Wagner, 1995).
In
contrast, more
flex-
ibility
exists in how
people
are allowed to behave
in
individu-
alistic
cultures,
reducing
the
salience
of
organizational
membership compared
with
collectivistic
cultures. We there-
fore
propose:
Proposition
2:
Members of collectivistic
organizational cultures
will
view
organizational
membership
as a more
salient
category
than
will
members of
individualistic
organizations.
Since
multiple
social
categories
exist,
one
question
is
whether
they
can
be
activated
simultaneously.
Research
supports
an
inverse
relationship,
labeled
functional
antago-
nism,
between the
salience
of
different
categories.
Func-
tional
antagonism
predicts
that as
one
category
becomes
more
salient,
others become less salient
(e.g.,
Turner
et
al.,
1994).
In
an
organizational
context,
the notion of
functional
antagonism
implies
that
when
organizational
membership
is
salient,
demographic
attributes
will be
less salient.
Specifi-
cally,
a collectivistic
culture
may increase
the salience
of
or-
751/ASQ,
December
1998
This content downloaded from 128.83.205.78 on Thu, 26 Feb 2015 20:52:52 UTC
All use subject to JSTOR Terms and Conditions
ganizational
membership as
a social
category,
causing a
cor-
responding
decrease
in the
salience
of
demographic
categories.
Overt
connectedness,
attending to
others,
and
harmonious
interdependence are
neither
assumed nor
valued
in
individu-
alistic
cultures.
Individuals
maintain
their
independence
from
others
by
attending
to
the self and
by
discovering
and ex-
pressing their
unique
inner
attributes
(Markus and
Kitayama,
1991:
224).
Because
the
organizational
boundary
connecting
members to
one
another
will not
be so
salient in
individualis-
tic
cultures as
in
collectivistic
cultures, an
organizational
em-
phasis
on
individualism
will
not
cause
demographic
catego-
ries to be
replaced as
the
primary
basis for
self-
categorization.
An
individualistic
culture
will
not
lead
people
to
ignore the
organizational
boundary
intentionally,
but an
individualistic
focus
may
make the
organizational
boundary a
less
likely
candidate for
categorizing self and
others. In
the
absence of a
salient
organization
boundary,
easily observed
and
distinctive
demographic
categories
may play
a
more
prominent
role
in
determining
category
salience,
as sug-
gested
in
the
following proposition:
Proposition
3:
The
salience of
demographic
attributes
as social
cat-
egories
will
be
inversely related to the salience of
organizational
attributes
as
social
categories.
The above
propositions
can
be
translated
into testable
hy-
potheses
by
considering
how
the
extent to which
people
are
demographically
different from one
another and an
organiza-
tion's
emphasis
on
individualism or
collectivism
can affect
work
processes.
Work
Processes:
Social Interaction and
Conflict
Social interaction.
Being
in
a situation in
which demo-
graphic
diversity
exists and
individual
achievement is
valued
increases one's
tendency to
categorize those who
are demo-
graphically
similar into
an
in-group
and
those who are
demo-
graphically
different into
out-groups.
People
trust and
interact
with
in-group
members
significantly
more than
out-group
members
(e.g.,
Brewer,
1979). But,
since the
pooling
of
di-
verse information
yields
many
of
the benefits of
diversity
(e.g., Gruenfeld et
al.,
1996),
one
of
the
fundamental chal-
lenges
to
increasing
these
benefits
in
organizations
is to
get
people
to
recategorize
demographically
dissimilar
people
as
in-group
members to increase
interaction and mutual
trust.
Research has
shown that
people
of
the same
nationality,
racial
background,
or
sex are more
likely
to
associate with
one another
within
organizations.
Ibarra
(1992),
for
example,
found that
men and women tended
to form
friendship
and
support
ties
with others of
the same sex
in
an
advertising
firm.
Even
though
people
who
are
demographically
different
may
have
multiple
perspectives
that are
valuable
in
ap-
proaching
and
solving
problems
(e.g.,
Jackson,
1992),
these
perspectives
are
less
likely
to
emerge
because of
people's
reluctance to
interact with others who are
demographically
different
and
difficulties
in
communicating
different
perspec-
tives when
interaction does take
place.
This
suggests
the
following
hypothesis:
752/ASQ, December
1998
This content downloaded from 128.83.205.78 on Thu, 26 Feb 2015 20:52:52 UTC
All use subject to JSTOR Terms and Conditions