A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Personality and Social Psychology
This content is subject to copyright. Terms and conditions apply.
Journal of Personality and Social Psychology
1995,
Vol. 68, No. 2, 280-291Copyright 1995 by the American Psychological Association, Inc
0022-3514/95/S3.00
Parameters of Social Identity
Kay Deaux, Anne Reid, Kim Mizrahi, and Kathleen A. Ethier
City University of New "fork
Two
studies establish distinct types of social and collective identities (Study 1) and describe dimen-
sions that differentiate among identities (Studies
1
and 2). In Study
1,
individuals (N
=
50) sorted
64 social identities on the basis of perceived similarity; 259 respondents provided trait property
ratings of the identities. Cluster analysis indicated 5 types of social identity: personal relationships,
vocations/avocations, political affiliations, ethnic/religious
groups,
and stigmatized
groups.
Multi-
dimensional scaling analysis shows that identities differ on several trait properties, including desir-
ability and collectivity. In Study 2, 171 people rated the similarity of identities within a specific
cluster; 193 respondents provided trait property
ratings.
Results indicate that different trait proper-
ties are relevant to each cluster. The theoretical importance of distinguishing among forms of social
identification
is
stressed.
Membership in social groups or collectives provides an im-
portant basis for self-definition. Thus, in addition to whatever
traits or characteristics people use to describe themselves as
unique individuals (sometimes referred to as personal
identity), they also locate themselves in the social context
through their claims to social categories. These categories in-
clude demographic categories (e.g., gender, ethnicity), social
roles (e.g., parent, spouse), and membership organizations
(e.g., Democrat, soldier).
How similar or different are these various forms of social
identification? Initially, theorists of
social
identity tended to as-
sume equivalence among
categories.
Conceptualizing at a more
abstract level (and often operationalizing in the form of artifi-
cial groups that have no intrinsic meaning), social identity the-
orists did not feel compelled to consider how different types of
group membership might yield different patterns of behavior.
Tajfel (1981) discussed possible differences between categories
to which
one is
assigned and social groups to which
one
belongs.
He suggested the two terms may simply represent "the begin-
ning and the end of a long social psychological process"
(1981,
p.
311), but he never developed this analysis.
More recently, investigators within the social identity tradi-
tion have begun to question assumptions of homogeneity of
so-
Kay Deaux, Anne Reid, Kim Mizrahi, and Kathleen
A.
Ethier, Doc-
toral Program in Social-Personality Psychology, Graduate School and
University Center of the City University of New \brk. Kathleen A.
Ethier
is
now at the Department of Psychology, Yale University.
This research
was
supported in part by National Science Foundation
Grant BNS-9110130. Some of the
findings
reported here were initially
presented at the June 1991 meeting of the American Psychological So-
ciety in Washington, DC.
We
thank Judi Addelston for her help in col-
lecting data and Seymour Rosenberg for statistical advice. In addition,
we
are grateful to Rupert Brown, Seymour Rosenberg, and members of
the City University of New \ork Identity Research Seminar for their
comments and suggestions on an earlier version of this article.
Correspondence concerning this article should be addressed to Kay
Deaux, Graduate School and University Center of the City University
of New
York,
33
West 42nd Street,
New
York,
New York 10036-8099.
cial
groups.
Brown and Williams (1984) were among the
first
to
raise the question of whether group identification means the
same thing to all people. Analyzing the applicability of social
identity theory for gender
identity,
Skevington and
Baker
(1989)
also pointed out areas of diversity.
Other research traditions also differentiate among social cat-
egories. In symbolic interaction theory, the concept of role-
identity is sometimes defined so as to include some collective
definitions and exclude others. Thoits (1992), for example,
confined the concept of identity to positions or roles, such as
mother or tennis player, and excluded social categories such as
gender, race, mental illness, or what she termed
"ex-roles,"
such
as retired person or divorced person. Stryker (1987), another
theorist within the symbolic interaction tradition, considered
identities based on race, age, and gender to be master statuses
that subsume more specific role-related identities.
Which of these distinctions is valid, or indeed whether any
partition of collective identities is useful, depends on the un-
derlying psychological significance. Are there key psychological
dimensions that differentiate one identity from another? Do
different types of identities have distinct meanings or serve dis-
tinct functions for the individual? Only by understanding what
the basis for differentiation
is
can
we
begin to think about what
the consequences of different identity categories might be.
Attempts to identify possible dimensions have been relatively
few and have typically been analytically rather than empirically
based. Hinkle and Brown
(1990;
Brown et
al.,
1992), for exam-
ple,
proposed that social identities differ on
two
dimensions,
in-
dividualistic versus collective and relational versus nonrela-
tional. Although they have not yet attempted to demonstrate
where various social identities might be represented in their re-
sultant four-category system, they have reported data showing
that
the
predictions of social identity theory relating group iden-
tification to outgroup
bias
hold true primarily for identities that
are both collective and relational.
Investigators of stigmatized identities have been most active
in trying
to
identify dimensions that differentiate various forms
of stigma (Goffman,
1963;
Jones
etal.,
1984;Katz, 1981).Sug-
gested dimensions include visibility of the stigmatizing condi-
280
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
