Content uploaded by José Luis González-Castro
Author content
All content in this area was uploaded by José Luis González-Castro
Content may be subject to copyright.
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
35
Independent and Interdependent Self-construals
and Socio-cultural Factors in 29 Nations
Titre français
Itziar Fernández*
Dario Paez**
José Luis González***
Mots-clés
Self-construals,
Individualism,
Collectivism,
Hierarchy, Cultural
Femininity
Key-words
Representations ou
Concepts de Soi,
Individualisme,
Collectivisme,
Hierarchue, Femineité
Culturel
Résumé
5 688 étudiants de 29 nations ont
répondu à une version courte de
l’échelle de Singelis sur le concept
de soi indépendant – interdépen-
dant. Une analyse factorielle trans-
culturelle nous a permis de mettre
en évidence quatre facteurs du
concept de soi : loyauté groupale,
interdépendance relationnelle,
unicité et « bas contexte ». Une
analyse factorielle confirmatoire
multi-échantillons a confirmé la
stabilité transculturelle de ces résul-
tats ainsi qu’une structure compa-
rable de six items de l’échelle pour
l’interdépendance.
L’interdépendance est en relation
avec une moindre connaissance de
la langue anglaise, une forte identi-
fication nationale, un bas statut
social et un plus grand partage
social des sentiments positifs.
L’indépendance est associée à la
Abstract
5688 students from 29 nations
answered a shortened version of the
Singelis scales for Independent-
Interdependent self-construal. A
pan-cultural factor analysis found
four factors: Group Loyalty,
Relational Interdependence,
Uniqueness, and low Contextual
Self. Multisample Confirmatory
Factor Analysis indicated a cross-
culturally stable, and comparable,
six item scale for Interdependence.
Interdependence was related to low
English fluency and high national
identification, low social status and
higher social sharing of positive feel-
ings. Independence was associated
with high English fluency and high
social status. A multivariate analysis
controlling for national income,
national identification, English
fluency, students’ relative social
status, parent’s educational level
* Itziar Fernández, U.N.E.D., Madrid, Spain
** Dario Paez,University of the Basque Country, San Sebastián, Spain
*** José Luis González, University of Burgos, Burgos, Spain
Correspondence concerning this article should be addressed either to Dario Paez.
Departamento de Psicología Social y Metodología, Universidad del Pais Vasco, Avda. de Tolosa
70, San Sebastian 20009, Spain. Request for reprints should be directed to Dario Paez (e-
mail pspparod@ss.ehu.es).
This study was supported by the following Basque Country University Research Grants (UPV
109.231-G56/98; 109.231-136-45-2001).
RIPS / IRSP, 18 (1), 35-63 © 2005, Presses Universitaires de Grenoble
MEP 1/2005 18/04/05 17:33 Page 35
This study aims to test empirically relations between self-con-
cept and beliefs related to the social context and cultural
values. The use of value, attitudes and belief scales has been crit-
icized due to the fact that they refer only to declarative and
abstract verbal knowledge (Matsumoto, 2002; Fiske, 2002).
Nevertheless, studies conducted on the topic of values have
shown convergent results (Markus, Kitayama, Fiske & Nisbett,
1998; Hofstede, 2001; Smith & Schwartz, 1997). Empirical analy-
ses of large scale value surveys suggest that the so called
individualistic societies emphasize achieving status and contrac-
tual, negotiated social relations, based on trust and social
interactions with strangers (Hofstede, 2001; Smith, Dugan &
Trompenaars, 1996; Schwartz 1994). Personal responsibility, egal-
itarianism and freedom are also emphasized. Collectivist societies
emphasize ascribed status, loyalty to ascribed groups and duties
and obligations towards in-groups – family, clan, ethnic, religious
or other groups. Personal relationships, inequalities, respect and
self-discipline and mistrust towards out-group members are also
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
36
and other cultural dimensions,
showed that individualism
(Hofstede’s nation score) was asso-
ciated with high Uniqueness and
low Group Loyalty. Low Contextual
Self and Relational Interdependence
were unrelated to individualism.
Uniqueness and low Group Loyalty
were more typical of «masculine»
and hierarchical cultures. Relational
Interdependence was linked to
«feminine» cultures and partially to
individualism.
connaissance de l’anglais et à un
statut social élevé. Une analyse
multivariable, contrôlant les reve-
nus nationaux, l’identification natio-
nale, la connaissance de l’anglais, le
statut social familial des étudiants,
le niveau d’éducation des parents
et d’autres dimensions culturelles,
montre que l’individualisme (les
scores nationaux de Hofstede) est
associé à une Unicité élevée et à une
moindre loyauté groupale. L’auto-
concept de « bas contexte » et l’in-
terdépendance relationnelle ne sont
pas liés à l’individualisme. L’unicité
et la basse loyauté groupale sont
plus typiques dans les cultures
masculines et hiérarchiques.
L’interdépendance relationnelle est
associée aux cultures féminines et
partiellement à l’individualisme.
MEP 1/2005 18/04/05 17:33 Page 36
important in so called collectivist societies (Smith & Schwartz,
1997; Smith, Dugan & Trompenaars, 1996; Hofstede, 2001).
Some of the most important collectivist attributes included by
social psychologists in belief and attitude scales have been (in
consensus order): a) a sense of duty to in-group (i.e. I will stay in
a group if it needs me, even when I am not happy with the group;
It is important for me to respect decisions taken by the group);
b) relatedness to others (i.e. I often have the feeling that my rela-
tionships with others are more important than my own
accomplishments); seeking others advice, and sense of belong-
ing; c) harmony and working in groups (i.e. It is important for me
to maintain harmony within my group; I respect people who are
modest about themselves), d) a contextualised self and e) valuing
hierarchy (Singelis et al., 1995; Oyserman, Coon &
Kemmelmeier, 2002).
Individualistic attributes included by social psychologists in belief
and attitude scales have been (in consensus order): a) indepen-
dence; b) personal achievement (i.e. I am comfortable with being
singled out for praise); c) self-knowledge (i.e. My personal iden-
tity, different from others, is very important for me and I act in the
same way no matter who I am with); d) uniqueness (i.e. I enjoy
being unique and different; e) privacy; f) direct communication
(i.e. I would rather say “no” directly than risk being misunder-
stood, and I prefer to be direct and forthright when dealing with
people I have just met), and g) competition (Singelis et al., 1995;
Oyserman, Coon & Kemmelmeier, 2002).
With regard to validation of the Singelis scales, there has been a
tendency towards comparing Euro and Asian Americans, and
North Americans and Asians. Usually Euro-Americans score
higher on Independence and lower on Interdependence than
Asian-Americans (Singelis et al, 1995; Singelis, 1994). Similar
results were found in studies comparing Euro-Americans and
Hong Kong Chinese (Singelis et al, 1999; Lee, Aaker & Gardner,
2000). Kurman (2001) confirmed that respondents from
Singapore scored lower in independence than (supposedly) indi-
vidualistic Jews and collectivist Israeli druzes -the latter scored
higher than the other two groups on Interdependence. Three
studies have shown that in a similar fashion to druze Israelis,
African Americans scored higher than Euro-Americans in
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
37
MEP 1/2005 18/04/05 17:33 Page 37
Independence and Interdependence, while no difference was
found between the latter and Latino Americans in Independence
– Latinos scored slightly higher in Interdependence. Asian
Americans scored lower in interdependence than Euro-
Americans (Coon & Kemmelmeier, 2001). Japanese scored lower
in Interdependence and no differences were found for
Independence in two studies comparing Canadian and US
respondents with Japanese (Matsumoto et al., 2001; Sato &
Cameron, 1999). Finally, Filipinos scored higher than North
Americans and Japanese both in Interdependence and
Independence (Uchida et al. 2001, quoted in Kitayama, 2002).
As we can see, there is mixed evidence for the validity of Singelis’
scales, and results suggest problems with their content validity
and the actual core attributes of individualism and collectivism.
Evidence relating collectivist and individualist beliefs
and attitudes to self-construal
Oyserman et al’s (2002) meta-analysis confirmed that individual-
ist North Americans score lower than other samples on scales
emphasizing a sense of duty towards the in-group. However,
when scales include items related to sense of belonging and seek-
ing other people’s advice, North Americans report higher scores.
Cross, Bacon & Morris (2000) differentiate a collectivist group-
oriented Interdependence (i.e. a sense of duty and in-group
loyalty) from a relationship-oriented Interdependence (i.e. vol-
untary interpersonal relatedness) and validate a measure of
Relational Interdependence associated with self-disclosure.
Gabriel and Gardner (1999) found that women focused more on
the relational aspects of Interdependence, whilst men empha-
sized more the collective aspects (i.e. group membership) of
interdependence more. Kashima, Yamaguchi, Kim, Choi, Gelfand
& Yuki (1995) found that females from individualistic nations
(USA, Australia) scored higher on a relatedness self-construal
dimension. Japanese males scored lower on this dimension. In
the same vein, McClelland’s (1961) scores on need for affiliation
are associated with cultural indices of individualism. As Hofstede
(2001) argues, in contractual and individualistic societies social
relationships are less ascribed or socially predetermined and peo-
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
38
MEP 1/2005 18/04/05 17:33 Page 38
ple have a stronger need to acquire friends. In sum, individualists
are more socially attuned and relationally skilled than collectivists
(Hofstede, 2001) and are more likely to include interpersonal
relationships in their self-construal (Cross, Bacon & Morris,
2000). Waterman (1981) argues that Interdependence and indi-
vidualism are associated, because individualistic values reinforce
generalized trust and voluntary coordination towards compatible
and coordinated goals, factors which facilitate pro-social and
cooperative behavior.
Oyserman et al’s (2002) meta-analysis also confirmed that indi-
vidualist North Americans score higher than other samples on a
scale emphasizing personal independence and uniqueness. US
respondents scored higher in public self-consciousness than
Japanese and Korean respondents, confirming a relationship
between individualism and presenting oneself in public as indi-
vidual (Hofstede, 2001).
With regard to competition and personal achievement,
Oyserman et al. (2002) remark that when competition items
were included in the studies, the difference between North
Americans and Japanese disappeared, suggesting that competi-
tiveness is unrelated to cultural individualism – as Mead
concluded in an ethnographic review (Fiske, 2002). However, a
competitiveness scale was related to collectivism in a 42 nation
study (Van de Vliert, 1998), and Triandis (1995) posited that com-
petition is related to vertical individualism and not to horizontal
individualism (Oyserman, Coon & Kemmelmeier, 2002).
Hofstede (2001) also emphasized that cooperation is unrelated
to collectivism.
In relation to the emphasis on internal attributes and a less con-
textualised self, studies do not confirm that individualistic
cultures reinforce low self-monitoring or that collectivist ones
reinforce the contextualised self (e.g. How I behave depends on
who I am with, where I am – Gudykunst, Yang & Nishida, 1987;
Goodwin, 1999). In fact, Anglo-Saxon respondents scored higher
on self-monitoring than Asian respondents. Collectivist persons
rely on norms and roles when deciding how to behave and not
on what they perceive that other people think and feel, and this
explains why they are low in self-monitoring (Gudykunst et al,
1987).
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
39
MEP 1/2005 18/04/05 17:33 Page 39
Hofstede (1998) posited that cooperation and nurturance are not
typical of collectivism, but of so called feminine societies, and
that competition and assertiveness are typical of masculine cul-
tures. Scandinavian individualistic and feminine cultures
emphasize personal autonomy and are also extremely non-com-
petitive, stressing modesty (Hofstede, 1998; Fiske, 2002). Due to
the fact that people in so called feminine cultures value warm
relationships, belongingness and relatedness may be more
salient in self-construal there than in competitive cultures.
Hofstede’s Cultural Masculinity scores correlates strongly with
Schwartz’s Mastery values (independent, successful, daring,
ambitious, choosing one’s own goals). Competitiveness as mea-
sured by a Spence-Helmreich scale showed a moderate
correlation with cultural masculinity and was strongly related to
Power Distance (Van de Vliert, 1998).
Hofstede’s power distance dimension deals with the amount of
deference and respect between superiors and subordinates and
the extent to which people accept status differences. In high
power distance cultures, high status people exert their authority
and status distinctions are more accepted as expected and legiti-
mate. In low power distance societies, status distinctions are
minimized and relations are more egalitarian. Hofstede posits
that people in low power distance scenarios are more interde-
pendent (i.e. equality is seen as the basis of society, so that
consultative and decentralized group dynamics reinforce the
sense of belonging). In high power distance cultures a few could
be independent and most should be dependent on high status
subjects. Because individuals rely on status and hierarchical rules
there is a lower need for interdependence.
The power distance dimension is related to the sharing of power
in society in general, including differential rewards between high
and low status people. An asymmetrical society should reinforce
competitiveness in order that one can rise up the social pyramid
(Hofstede, 1998; 2001). Various studies suggest that competitive-
ness, internality, protestant work ethic beliefs (PWE), and work
centrality are higher in less developed, collectivist and high
power distance cultures in which materialistic values are still
important (Smith, Dugan & Trompenaars, 1998; Van de Vliert,
1998; Inglehart, Basañez & Moreno, 1998; Furnham et al., 1993).
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
40
MEP 1/2005 18/04/05 17:33 Page 40
Higher competitiveness, internal locus of control and agreement
with PWE beliefs are found in both student and manager samples
in developing, collectivist and high power distance countries.
This suggests that these results characterize only elitist groups in
poor, developing and collectivist countries. In fact, perception of
life control as measured in Inglehart’s World Value Surveys was
related to high social development – as could be expected from
studies linking higher social status and internal control (Sastry &
Ross, 1998). However, using nationally representative samples,
the emphasis on hard work as a quality that children should be
encouraged to learn at home (World Values Survey, 1994;
Inglehart, Basañez & Moreno, 1998) was shown to be related to
collectivism and high power distance (Hofstede, 2001). In post-
industrial societies, individualistic and low power distance
cultures, post-materialistic values related to self-actualization and
quality of life are more important than work and material success.
For instance, work is more central in cultures high on hierarchy
and mastery (Schwartz, date), and probably also masculinity and
power distance, as Hofstede’s scores on masculinity and power
distance are strongly related to mastery and hierarchy respec-
tively.
To summarise, a core aspect of collectivism is a sense of duty and
obligation towards the group. To a lower extent in-group har-
mony and working in groups are also typical features. A sense of
belonging, relatedness and cooperation are unrelated or nega-
tively related to collectivism and are probably more characteristic
of “feminine” or cooperative and egalitarian cultures. Core
aspects of individualism are personal independence and unique-
ness. Competition, personal achievement, emphasis on internal
attributes in opposition to other people’s opinions and cues are
unrelated to or negatively related to individualism. Personal
achievement and competition are more valued in developing
nations, “masculine” or competitive and hierarchical cultures,
than in post-materialist, developed, egalitarian, cooperative or
“feminine” and contractual societies. In the former societies, rel-
ative scarcity of resources and a strong struggle for social survival
and acceptance of inequalities impose strong in-group solidarity,
generalized competitiveness and an emphasis on personal effort
and rewards. In the latter, material stability, lack of ascribed group
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
41
MEP 1/2005 18/04/05 17:33 Page 41
membership and expressive individualism reinforce the impor-
tance of social relationships.
Previous studies of these issues share a series of misconceptions
and limitations. Most of the data compares North Americans with
Asian collectivists and does not compare differences within
America or within Europe. Cross-cultural reliability and structural
validity are checked by means of confirmatory factor analysis only
within a minority of studies. Reliabilities tend to be low and only
some studies try to improve reliabilities and cross-cultural com-
parability by means of item selection.
This study will contrast the convergent validity and reliability of a
short form of the Singelis self-construal scale, using large student
samples from the Americas, Europe (including Russia), Asia
(Taiwan, China and Singapore) and to a lesser extent the Middle
East (Turkey and Lebanon) and Africa (Nigeria and Ghana). Pan-
cultural factor analysis and confirmatory factor analysis will
contrast the structural comparability of scales. Within-subject
standardized scores will be used in order to control for response
set tendencies.
Finally, self-concept and attitudes scores will be correlated with
English fluency, national identification and a scale of verbal
expression and social sharing of positive emotions. High English
fluency and low national identification, as indexes of contact with
foreign cultures and cosmopolitanism, should be negatively
related to interdependent or collectivist self-construal and posi-
tively to independent or individualist self-construal. Social
sharing of emotions is common (Rimé et al.1999) and sharing of
happiness scenarios has been found related to collectivism, low
power distance and cultural femininity in other studies
(Fernandez et al., 2000). An interdependent self-concept,
because of the importance of social relations, should be associ-
ated to social sharing of happiness. Latin American and African
collectivism values “sympathy” that implies an emphasis on posi-
tive interaction and particularly the verbalization of happiness. An
independent self-concept, because of the importance of auton-
omy and competition, probably should be negatively related to
the social sharing of happiness.
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
42
MEP 1/2005 18/04/05 17:33 Page 42
Method
Procedure and Participants
All participants were volunteers from psychology and social sci-
ence courses in universities in the Americas, Europe, Russia
(Moscow and Tula), China (Beijing), Iran, Turkey, Lebanon,
Nigeria and Ghana (see Table 2 and annex for the 29 nations stud-
ied). The sample consisted of 5,688 subjects (2,459 male and
3,229 female) with a mean age of 21.78 years (SD = 4.25). Table
2 gives details of the samples, percentage of females, and relia-
bility for Independent and Interdependent self-construals.
Materials
Three questionnaires were used in the present study. We used
versions in English, Spanish, German, Portuguese, French,
Italian, Russian, Turkish, Greek, Chinese and Farsi (see Annex). In
designing these versions, the authors followed the guidelines
proposed in the literature on cross – cultural methodology
(Brislin, 1986): independent, blind, back translation and review
by representative focus groups. The objective of the translation
process was to preserve the conceptual meaning of the original
form.
The first questionnaire asked for background information,
including fluency in English, parent’s education, age, and identi-
fication with one’s nation. The second questionnaire contained
the Singelis scales discussed below. A third questionnaire asked
for typical expressive reactions of positive mood.
Measures
Nation-level scores
Cultural dimensions. Hofstede (2001) reports Individualism-
Collectivism (IDV), Cultural Masculinity (MAS) and Power
Distance (PDI) scores for 74 nations and regions. These ratings
are mostly based on questionnaires answered by IBM employees
throughout the world in the 1970s. In spite of the fact that the
survey was conducted nearly 30 years ago, Hofstede’s scores
show high convergent validity with current surveys of values and
with recent cross-cultural studies (Schwartz, 1994; Smith & Bond,
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
43
MEP 1/2005 18/04/05 17:33 Page 43
1998). Schwartz (1994) reports Intellectual and Affective
Autonomy (IDV), Hierarchy (PDI) and Mastery (MAS) scores for
38 nations and regions. Inglehart (1997) reports Post-materialism
scores for 43 nations and Hofstede provides Inglehart’s scores for
the Traditional-Secular dimension in 27 countries. Smith, Dugan
& Trompenaars (1996) report means for 43 nations on two
dimensions found with an MDS of values and scripts based on
11.000 employees. Higher scores on the first dimension
(Egalitarian Commitment) implies a preference for universalistic
relations and status based on achievement. Low scores empha-
size preferences for personal and particularistic relations and
ascribed status. The second dimension (loyal versus utilitarian
involvement) opposes preferences for family loyalty and collec-
tive responsibility to an emphasis on personal responsibility and
negotiated social relations. Hofstede’s scores and Schwartz’s rat-
ing of culture-level value types show strong concurrent validity:
IDV correlated with affective and intellectual autonomy (rAA (23)
=.46 and rIA (23) =.53, p .05 respectively), low hierarchy (rH
(23) =-.51, p .05) and egalitarian commitment (rEC (23) =.51,
p .05). Hofstede’s PDI scores correlated positively with conser-
vatism and negatively with affective autonomy (r (23) =.45, and
r =-.45, p .05, respectively (Schwartz, 1994, p.109). IDV and
FEM correlate with post-materialism and PDI with traditional
authority (Hofstede, 2001). High scores mean high IDV, MAS, and
PDI (see Table 1 for the contextual social and cultural scores for
nations that are used in the present study).
Socio-economic factors. For each country we used both the
Gross National Product (GNP), and the Human Development
Index (HDI). The HDI measures national well-being and trends
by combining three basic components of human development:
longevity (mean life expectancy in the nation); knowledge (rate
of literacy and school population); and standard of living (Gross
National Product per person). HDI is the best known measure of
development. It is more useful than purchasing power and per
capita measures (Cordelier & Didiot, 1997). GNP in dollars and
HDI scores for each nation were obtained from the United
Nations Program for Development (UNPD) (Cordelier & Didiot,
1997; PNUD, 1999).
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
44
MEP 1/2005 18/04/05 17:33 Page 44
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
45
TABLE: 1
Nation-level predictors
Country
Argentina
Belgium
Bolivia
Brazil
Chile
China
Colombia
France
Germany
Ghana
Greece
Guatemala
Iran
Italy
Lebanon
Mexico
Nigeria
Panama
Peru
Portugal
Russia
Salvador
Singapore
Spain
Switzerland
Taiwan
Turkey
USA
Venezuela
IDV
46
75
12
38
23
20
13
71
67
20
35
6
41
76
38
30
77
11
16
27
39
19
20
51
68
17
37
91
12
PDI
49
65
64
69
63
80
67
68
35
77
60
95
58
50
80
81
77
95
64
63
93
66
74
57
34
58
66
40
81
MAS
56
54
50
49
28
66
64
43
66
46
57
37
43
70
53
69
46
44
42
31
36
40
48
42
70
45
45
62
73
SUP
36
56
21
15
31.5
5.5
16.5
56
47
15
36.5
8
18.5
47
27
16
4
32
26
35
43
18
34
56.5
32.5
21
60
28
HDI
.885
.929
.584
.796
.882
.609
.840
.935
.920
.467
.909
.580
.754
.914
.664
.845
.400
.859
.694
.878
.804
.576
.881
.933
.926
.711
.940
.859
Egal
-.38
.84
.66
-2.20
1.11
1.38
.31
.39
.68
-.08
.91
-2.32
-.60
.03
-.03
1.96
Utili
-.04
-.12
.22
-.99
.09
-.54
.86
.01
.50
.40
.18
-1.31
1.93
.53
.75
-.15
GNP
8030
24710
800
3640
4160
620
1910
24990
27510
390
8210
1340
2000
19020
2660
3320
280
2750
2310
11000
2240
1610
26730
13580
40630
2780
26890
3020
Note. IDV=Individualism, PDI=Power-distance, MAS=Masculinity (Data= Hofstede, 1991,
2001), (Bolivia = estimated mean between Peru and Ecuador, Lebanon=Arab-countries,
Ghana=West Africa); GNP=Gross National Product in dollars, SUP= Percentage of people
with higher level studies, HDI=Human development index (Data=PNUD, 1999; Cordelier
and Didiot, 1997); Egal= Egalitarian Commitment and Utili=Utilitarian Involvement
(Data=Smith, Dugan & Troompenars, 1996). A high score for each variable denotes a high
degree of the construct in question.; n= 29.
MEP 1/2005 18/04/05 17:33 Page 45
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
46
Percentage of the Population receiving higher education. As an
indication of students’ elite social position we used the percent-
age of the population receiving higher education. This figure is
given in Table 1.
Individual Level Measures:
1.- Identification with the nation (ranging from 1 = none to 5
= very much)
2.- Self-reports of English Fluency (ranging from 0 = none to 6
= correctly)
3.- Social status: Parent’s studies: 1 = Illiteracy; 2 = Primary; 3
= Secondary; Technical Higher Education; 4 = University.
4.- Self-construal
Singelis Scale. A shortened and modified version of the
Singelis et al (1994) scale was used. All the original items related
to vertical collectivism or «respect» were excluded in order to
avoid content confounding between collectivism and power dis-
tance (original items 1,2,8, 20). The selection of items was
performed by a group of 12 multicultural American and
European social psychologists. A pilot study in Latin-America and
southern Europe was used to exclude items related to health and
well-being that were normative A large majority of people
answered affirmatively to original items 15, 18, 24 and 11. The
seven remaining interdependent self-construal items and the six
remaining independent items are shown in Table 3. Respondents
answered using 4-point Likert scales: 1 = totally disagree; 4
= Totally agree. This was preferred to a longer scale because
reducing scale points has been proposed in order to reduce
response bias. Within-subject standardization was also used
(Smith & Bond, 1998).
5.- Social sharing of happiness scale: Six scales asked how typi-
cal it is when feeling joy or happiness to share with others: a)
Being friendly; b) Not falling out with other people; c) doing nice
things for other people; sharing feelings so that other people
may feel good; d) communicating good feelings to other people;
e) saying positive things; f) talking a lot, being talkative.
Respondents answered on 4-point Likert scales: 1 = not typical;
4 = very typical.
MEP 1/2005 18/04/05 17:33 Page 46
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
47
Nation
Argentina
Belgium
Bolivia
Brazil
Chile
China
Colombia
El Salvador
France
Germany
Ghana
Greece
Guatemala
Iran
Italy
Mexico
Nigeria
Panama
Peru
Portugal
Russia
Singapore
Spain
Switzerland
Taiwan
The Lebanon
Turkey
USA
Venezuela
N
225
87
114
500
138
119
127
102
191
109
28
114
42
87
120
169
37
80
120
264
120
119
1273
175
36
118
105
120
226
%
Wom en
59.6
82.6
51.8
52.6
57.7
49.3
52.8
83.9
51.6
62.1
44.8
81.2
81.3
60.9
51.7
50.9
25.1
75.1
50.0
59.1
50.9
48.7
54.8
81.1
86.1
49.2
63.5
42.6
58.1
Alpha
interde-
pendent
.61
.61
.46
.57
.65
.70
.75
.64
.54
.52
.11
.69
.53
.65
.73
.64
.58
.47
.64
.68
.65
.69
.67
.68
.70
.63
.64
.71
.66
Alpha
indepen-
dent
.31
.51
.49
.39
.64
.34
.55
.44
.45
.54
-.50
.33
.52
.14
.59
.60
.63
.37
.50
.54
.52
.59
.64
.48
.54
.60
.45
.54
.49
Mean
interde-
pendent
3.06
2.89
3.10
3.15
2.95
2.88
2.94
3.26
2.89
2.95
2.76
3.00
2.97
2.91
2.87
2.72
3.10
2.95
3.04
3.14
2.82
3.09
3.06
2.92
2.87
3.05
2.74
2.96
2.94
Mean
indepen-
dent
2.84
2.75
3.20
2.96
2.99
2.87
3.15
2.98
2.81
3.04
2.78
2.98
3.12
2.79
2.97
3.28
3.20
3.20
3.07
2.85
2.90
2.80
2.79
2.83
3.00
3.12
2.99
2.96
2.98
Singelis: Interdependent self–construal (7 items) and Independent self-construal (6 items)
Mean with effect on gender as a covariate.
TABLE 2:
Percentage of females,
means and Cronbach
alpha reliabilities for
interdependent and
independent self–cons-
trual, for each nation.
MEP 1/2005 18/04/05 17:33 Page 47
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
48
Results
Reliabilities and factor analysis of the Singelis items:
Individual level.
Means, Alpha coefficients and proportion of women for each
nation are shown in Table 2. Reliabilities were low, particularly in
African and Asian samples (see Table 2). A pan-cultural factor
analysis (Principal Components, Varimax rotation) was per-
TABLE 3:
Pancultural factor analy-
sis of Singelis’ Self-
Construals items
Item
Group Loyalty
7. I would stay in a group if they needed me,
even if I were not happy with the group.
4. I will sacrifice my self-interest for the benefit
of the group I am in.
5. I often have the feeling that my relationships
with others are more important than my own
accomplishments.
6. It is important for me to respect decisions
made by the group.
2. My happiness depends on the happiness of
those around me.
Uniqueness
11. I enjoy being unique and different from
others in many respects.
9. I am comfortable with being singled out for
praise or rewards.
12. My personal identity is independent of others,
is very important for me.
Low Context
10. I act the same way no matter who I am with.
13. I prefer to be direct and forthright when dea-
ling with people I’ve just met.
8. I’d rather say “no” directly, than risk being
misunderstood.
Relational Interdependence
3. I respect people who are modest about them-
selves.
1. It is important for me to maintain harmony
within my group.
Eigen Value
% explained variance
Cronbach alpha reliabilities
Factors
1
.69
.64
.59
.57
.51
.37
2.36
18.13
.61
2
.78
.66
.64
1.76
13.58
.51
3
.73
.69
.45
1.13
8.72
.37
4
.36
.77
.58
1.01
7.77
.39
MEP 1/2005 18/04/05 17:33 Page 48
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
49
formed both on raw scores and on the within-subject standard-
ized scores. Both analyses showed similar results. Additional data
analysis was performed with a random sample of 200 Spaniards
and 200 Brazilian participants, equalizing also the percentage of
males, in order to equalize the weight of nations and to control
for gender inequality. Factor analyses were performed also by cul-
tural areas (Latin-America, North-America, Germany, French
Nation
Argentina
Belgium
Bolivia
Brazil
Chile
China
Colombia
El Salvador
France
Germany
Ghana
Greece
Guatemala
Iran
Italy
Lebanon
Mexico
Nigeria
Panama
Peru
Portugal
Russia
Singapore
Spain
Switzerland
Taiwan
Turkey
USA
Venezuela
Groupe
loyalty
Mean
2.86
2.51
2.91
2.97
2.81
2.68
2.77
3.25
2.42
2.73
2.63
2.70
2.97
2.69
2.53
2.55
2.97
2.78
2.80
2.83
2.60
2.91
2.86
2.41
2.55
2.74
2.42
2.91
2.78
S.D.
0.04
0.06
0.06
0.03
0.05
0.05
0.05
0.05
0.04
0.06
0.11
0.05
0.11
0.07
0.05
0.05
0.10
0.06
0.05
0.04
0.04
0.05
0.02
0.04
0.10
0.05
0.06
0.06
0.04
Uniqueness
Mean
2.94
3.04
3.22
2.94
3.03
2.96
3.32
2.96
2.97
3.08
2.79
3.40
3.25
2.95
2.99
3.40
3.26
3.30
3.09
2.87
3.21
2.94
2.90
3.13
3.04
3.22
3.19
3.15
3.04
S.D.
0.04
0.07
0.06
0.03
0.05
0.05
0.06
0.06
0.05
0.06
0.12
0.06
0.11
0.07
0.06
0.05
0.10
0.07
0.06
0.04
0.04
0.06
0.02
0.05
0.10
0.06
0.06
0.06
0.04
Low Context
Mean
3.40
3.33
3.45
3.54
3.24
3.16
3.27
3.54
3.43
3.21
2.91
3.33
3.13
3.23
3.38
3.11
3.42
3.47
3.47
3.55
3.14
3.37
3.36
3.45
3.20
3.48
3.18
3.12
3.34
S.D.
0.03
0.05
0.05
0.02
0.04
0.04
0.04
0.05
0.04
0.05
0.09
0.05
0.09
0.06
0.05
0.04
0.08
0.06
0.05
0.03
0.03
0.05
0.01
0.04
0.08
0.05
0.05
0.05
0.03
Relational
Interdependence
Mean
2.73
2.46
3.18
2.98
2.96
2.78
2.99
3.00
2.65
3.00
2.77
2.55
2.99
2.63
2.94
3.15
3.14
3.09
3.05
2.84
2.60
2.66
2.69
2.54
2.95
3.02
2.79
2.77
2.92
S.D.
0.04
0.07
0.06
0.03
0.05
0.05
0.06
0.06
0.05
0.06
0.12
0.06
0.11
0.07
0.06
0.05
0.10
0.07
0.06
0.04
0.04
0.06
0.02
0.05
0.10
0.06
0.06
0.06
0.04
TABLE 4:
Means for pancultural
factors by nation with
gender as covariate.
MEP 1/2005 18/04/05 17:33 Page 49
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
50
speaking Europe, Southern Europe, Arab and Asian countries).
The factor structure obtained was stable.
The first factor explained 15% of the variance and shows a clear
meaning of group loyalty, clustering items related to sense of
duty and obligation towards in-groups and relatedness. The sec-
ond factor explains 12% of the variance, and independent items
related to uniqueness and personal reward load in this dimen-
sion. The third factor explains 11% of the variance and was
formed by independent items associated with direct communica-
tion and a “less contextualised self ”. Finally the fourth factor
explains 10.5% of the variance and clusters interdependent items
with a meaning of in-group harmony and relation. Only the first
interdependent factor shows a moderately satisfactory reliability
coefficient.
Multisample confirmatory analysis found a six item interdepen-
dence factor with satisfactory goodness of fit (CFI =.94), and
transcultural validity. This factor includes group oriented items of
the first factor found by the exploratory pan-cultural analysis plus
an item emphasizing in-group harmony: a) It is important for me
to maintain harmony within my group, b) My happiness depend
on the happiness of those around me, c) I will sacrifice my self-
interest for the benefit of the group I am in, d) I often have the
feeling that my relationships with others are more important
than my own accomplishments, e) It is important for me to
TABLE 5:
Intercorrelations of
original and pancultural
Self-Construals
measures
Measures
1. Original-Singelis-Interdependent
2. Original-Singelis-Independent
3. CFA-Interdependent
4. Group Loyalty
5. Uniqueness
6. Low Context
7. Relational Interdependence
1
.07*
.94*
.86*
-.09*
.08*
.73*
2
-.01
-.02
.73*
.74*
.04*
3
.92*
-.11*
.08*
.67*
4
-.11*
.07*
.40*
5
.17*
-.03
6
.08*
Note. Pearson product-moment coefficients. A high number on each variable denotes a high
score on that same variable
Originals Self-Concepts: Interdependent (7 items) Independent (6 items)
CFA (Confirmatory Factor Analysis)-Interdependent (6 items)
Factors (exploratory factor anaysis): 4 dimensions (1=Group Loyalty; 2= Uniqueness, 3=Low
Contextual Communication; 4=Relational interdependence and group harmony)
n= 5138-87, 29 countries.
* p.01 (two-tailed).
MEP 1/2005 18/04/05 17:33 Page 50
respect decisions taken by the group, f) I will stay in a group if it
needs me, even when I am not happy with the group (one inter-
dependence item was dropped). With respect to independence
multisample confirmatory analysis (CFA) was unable to define a
factor with satisfactory goodness of fit and cross-cultural validity.
Concurrent validity of the Singelis sub-scales and factor
scores: individual level
To check for the construct validity of the Singelis items, correla-
tions were computed between the scores of the two original
Singelis scales and the four pan-cultural factors. Gender, parent’s
educational level as an index of social status, English fluency and
national identification as an index of acculturation were also cor-
related with self-construal scores. Separate analyses were
performed using raw scores and within-subjects standardized
scores. Both analyses show similar results. It is important to recall
that Group Loyalty and the Interdependent dimension found by
CFA are the most stables factors.
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
51
TABLE 6:
Predictors of factors
derived from Singelis
Self-Construals items
(Multiple Regression
Analysis-).
Nation
GNP
SUP
IDV
PDI
MAS
SEX
STATUS
ENGLISH
IDENTITY
R (R2)
Groupe
loyalty
r
.03
.12*
.05
-.14*
-.13*
-.03
-.12*
-.05
.02
β
.02
.06*
-.06*
-.14*
-.11*
-.04*
-.08*
-.07*
.03
.23(.054)
Uniqueness
r
.02
-.03
.01
.13*
.10*
-.05
.15*
.11*
-.07*
β
-.05
.04
.07*
.17*
.11*
-.05*
.12*
.12*
-.07*
.27(.07)
Low Context
R
-.13*
-.15*
-.12*
.05
.08*
.02
.03
-.03
.06*
β
-.09*
-.07*
-.04
-.04
.09*
.03
.01
.01
.03
.18(.03)
Relational
Interdependence
R
.10*
.06
.07*
-.04
-.07
.09*
-.08
-.06*
.05*
β
.15*
-.05*
.02
.00
-.11*
.07*
-.07*
-.07*
.07 *
.21(.04)
Note. Pearson product moment coefficients and standardized beta coefficients. A high number
on each variable denotes a high score on the variable as named., Collective variables:
GNP=Gross National Product, SUP= percentage of a nation’s people with higher level
studies, IDV= Individualism, PDI= Power distance, MAS=Masculinity; Individual variables:
Sex (1=men,2=women), Status= parent’s education level; English= English fluency,
Identity= identification with the nation. Individual level analysis n= 4023-85, 29 nations.
Method enter. R(R2) = Multiple R (Adjusted R square).
Pancultural Factors (exploratory factor anaysis): 4 dimensions (1=Group Loyalty; 2=
Uniqueness, 3=Low Contextual Communication; 4=Relational interdependence and group
harmony)
*p.01 (two-tailed).
MEP 1/2005 18/04/05 17:33 Page 51
Table 5 shows that Group Loyalty and Relational
Interdependence are correlated significantly. Group Loyalty also
correlated negatively with Uniqueness. Uniqueness was positively
related to the Low Context factor.
Gender (being female) was associated with the original Singelis
interdependent scale, r =.04, p.001 (with raw scores), low con-
text self, r =.04, p.001 (with raw scores), and with Relational
Interdependence, r =.09, p.001 with raw scores and r =.10, p
<.001 with standardized scores).
Group Loyalty also correlated with low fluency in English and low
family status. Relational Interdependence correlated with low
family status, high national identification, and low English fluency.
Finally, Uniqueness factor correlated with low national identity,
high fluency in English and high family status.
Females show a relational and low contextual self profile while
males share a Group Loyalty and Uniqueness profile. Finally, the
original interdependent scale and the CFA interdependent scale
correlate with social sharing of positive emotions, both r =.10,
p.001 (with within subject standardized scores), Group Loyalty
also correlates with verbal expression of joy, (r =.05, p.001 and
r =.10, p.001 with standardized scores). The Low contextual
self and Uniqueness correlate negatively with social sharing of
happiness, r = –.06, and r = –.07, both p.001.
Regressions predicting Singelis factor scores at the indivi-
dual level
In order to control for possible interrelations or multicollinearity,
multiple regressions were computed and standardized?? coeffi-
cients were used as indices of specific influence. To control for
the relative status of students within each nation, the percentage
of the population with a high education level was also included
in the regression. Sex (dummy variable, Male = 0, Female = 1),
English fluency, national identification, and family status (parent’s
educational level) were also included with the aim of controlling
for individual factors. Singelis’ factor scores were regressed on
sex, Gross National Product per capita, percentage of population
with higher education, and Hofstede’s nation scores. Individual
level variables were entered in a first step and then in a second
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
52
MEP 1/2005 18/04/05 17:33 Page 52
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
53
step the more distal cultural values scores were entered (see
table 6).
Group Loyalty is predicted by cultural femininity, low power dis-
tance, low English fluency and high national identification, high
national educational level, collectivism, low family status and
male gender. Results were similar for the CFA Interdependent
score.
Relational interdependence is predicted by cultural femininity,
high income, lower acculturation, low family status and by being
female. Results suggest that interdependence, particularly in-
group loyalty, is more typical of traditional and low status
respondents in developed, feminine, low power distance and col-
lectivist societies.
Uniqueness was predicted by high power distance, masculinity,
individualism, by being male, lower identification with nation and
higher English fluency.
Low contextual self was predicted by low income, and nation’s
low educational level. Results suggest that independence, partic-
ularly uniqueness, is more typical of less traditional and high
status respondents in competitive, high power distance and indi-
vidualistic cultures, and less developed societies in which
university students belong to a social elite.
Discussion
First of all, we should note that effect sizes tend to be small, as
could be expected using large samples including large uncon-
trolled sources of variance (Scherer, 1997) and declarative
decontextualized items, as the Oyserman et al. (2002) meta-
analysis shows. However, in spite of using distal contextual
socio-cultural factors, the results do offer a coherent profile.
The Singelis interdependence items cluster into two strongly
related factors (Group loyalty and Relational interdependence) in
the pan-cultural factor analysis. Confirmatory factor analysis
shows that a cross-culturally stable and comparable six item inter-
dependence scale could be used combining five items of the
Group Loyalty factor and one Relational interdependence item,
in-group harmony.
MEP 1/2005 18/04/05 17:33 Page 53
Singelis’ Group Loyalty, Relational Interdependence and the CFA
interdependence factor score were all associated with lower
English fluency, confirming that low contact with other cultures is
related to collectivist self-construal. Moreover, interdependent
self-construal was related to national identification (see also in
this number the Smith et al paper for similar results), suggesting
that people with an interdependent self-concept, because of their
emphasis on in-group loyalty, are more likely to identify with their
nation. Simultaneously, interdependent self-construal was related
to the social sharing of positive feelings, confirming that people
with interdependent self-construals report communicating posi-
tive feelings more with others. Social sharing of emotions and
sharing in happiness scenarios has been related to collectivism,
low power distance and cultural femininity in other studies
(Fernández et al., 2000).
In the multivariate analysis, the Interdependence factor scores
were partially associated not only with low English fluency and
high national identification but also with cultural indices of col-
lectivism. This in some way confirms that beliefs of loyalty and
obligation to in-groups are a core attribute of collectivism. This is
congruent with Oyserman, Coon and Kemmelmeier’s (2002)
meta-analysis which found that Euro-Americans scored lower in
collectivism than African Americans and to a lesser extent than
Asians, Europeans, Latin Americans and people from the Middle
East. However, the interdependence scales, i.e. Group Loyalty
and the cross-culturally stable and comparable six item interde-
pendence scale all correlate similarly with low power distance,
feminine cultures and high income. Moreover, the multivariate
analysis shows that Singelis’ Group Loyalty was specifically pre-
dicted by a feminine and low power distance or egalitarian
cultural context. This multivariate analysis suggests that egalitar-
ian and cooperative values are more important than collectivist
values in order to explain the relative dominance of group loyalty
beliefs. Low power distance and hierarchy cultures de-emphasize
obedience, inequalities, respect, deference and dependence on
superiors. Results confirm Hofstede’s argument: In low power
distance cultures most subjects should be interdependent, as
reflected on the higher value placed on Group Loyalty. In high
power distance cultures a few should be independent (as in our
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
54
MEP 1/2005 18/04/05 17:33 Page 54
university student samples) and most should be dependent
(Hofstede, 2001). It is because developed countries are usually
low in power distance or relatively egalitarian that social devel-
opment is related to interdependence.
However, results suggest that cultural values are more important
than socio-structural factors in order to explain interdependence.
Independent of social development, feminine cultures stress rela-
tions with others and social support, and by this token reinforce
loyalty towards groups and in-group harmony, while masculine
cultures de-emphasize interdependence and reinforce indepen-
dence. Results confirm Hofstede’s assumptions on the
importance of social relationships, modesty and harmony in cul-
tural femininity (Hofstede, 2001).
The fact that Relational Interdependence factor show low relia-
bility and than scores were higher in developed countries and
related to individualism is congruent with Oyserman, Coon and
Kemmelmeier’s (2002) meta-analysis, which found that when
studies assessed collectivism using items related to sense of
belonging and social support, individualistic Euro-Americans
scored as relatively collective. To be in groups and seeking social
support is a way of stressing interdependence not incongruent
with individualism and developed countries. Moreover, devel-
oped and post-materialist nations emphasize self-expression and
quality of life, particularly satisfactory interpersonal relationships
(Inglehart & Baker, 2000).
The independence items yielded two factors (Uniqueness and
direct communication and Low Self-Context), which are posi-
tively correlated. The confirmatory factor analysis showed that it
is difficult to extract a cross-culturally stable and comparable
independence scale, but confirmed the higher stability of the
interdependent construct.
Singelis’ Uniqueness factor score was associated with high
English fluency and low national identification, confirming that
cosmopolitanism is related to individualistic self-construal.
Low Context Self was related to high national identification and
negatively related to social development. Low Context Self scores
correlated with low social development (low income and lower
national level of education) and with collectivism. In spite of the
fact that collectivist cultures stress indirect communication in
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
55
MEP 1/2005 18/04/05 17:33 Page 55
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
56
personal relationships, our results suggest that these cultures are
less flexible in communication and behavior - at least in self-
reports. Because collectivist people rely on norms and roles
when deciding how to behave and probably pay less attention to
other people’s internal states we can explain why they have low
scores in the self-monitoring style scale - as several studies have
previously found. On the other hand, individualistic people
should be more socially attuned and flexible in order to establish
and maintain fluid social relationships with strangers. Other stud-
ies also suggest that sociability and concern for interpersonal
relationships are higher in individualistic societies, where inter-
dependence needs to be created and maintained voluntarily, than
in collectivist societies, in which subjects have defined and long-
term normative in-groups (Wheeler, Reiss & Bond, 1997).
Triandis et al. (1988) when discussing the cross-cultural results
for collectivism-individualism concluded that: “…people
are…more sociable in individualistic countries than they are in
collectivist because they have to work hard to get into and to
remain in their in-groups”. It is important to notice that these
beliefs were shared by students of poor and hierarchical coun-
tries, where they are part of the elite – by comparison to more
developed nations, in which a large proportion of the population
enters higher education. On the other hand, cultures which
share competitive and more egalitarian values were also specifi-
cally related to Low Context Self, controlling for other factors. In
masculine cultures “toughness” is valued, while in low power dis-
tance cultures superiors and inferiors are perceived as
interdependent, and subjects probably communicate in a more
direct manner (Hofstede, 2001).
Uniqueness was related to cultural masculinity and high power
distance – to competitive and hierarchical cultures. However
uniqueness was also predicted by individualism.
Results also show that males are not more individualistic, with a
more independent self-construal, and that females are not simply
collectivist with a more interdependent self-concept (Kashima,
Yamaguchi, Kim, Choi, Gelfand & Yuki, 1995). Males report
higher Group Loyalty and at the same time more independence
(i.e. Uniqueness). Females report slightly higher scores on
Relational Interdependence. Results are congruent with previous
MEP 1/2005 18/04/05 17:33 Page 56
studies which show higher pro-social values and relationalism in
females. A meta-analysis of gender differences in Schwartz values
in 47 nations found that men endorse more often values related
to power (social status, dominance over people and resources)
while women agree more with values related to benevolence
(preserving and enhancing the welfare of those to whom one is
close) (Smith & Schwartz, 1997). Studies also support the idea
that women focus more on the relational aspects of interdepen-
dence and men emphasize more the collective aspects (i.e.
Group Loyalty of interdependence (Gabriel & Gardner, 1999).
Conclusions
Results partially support the idea that a sense of obligation and
duty to the in-group is a core attribute of collectivist self-con-
strual. As Oyserman et al have argued, relatedness to others,
seeking other people’s advice and sense of belonging to a group
were also related to individualism and social development. Being
voluntarily interdependent is congruent with individualistic val-
ues. This suggests that future research should take into account
the important difference between vertical and horizontal individ-
ualism – the latter focusing more on social relationships. A large
majority of individualistic cultures (western European countries)
are simultaneously high in autonomy and in egalitarian commit-
ment values (Schwartz & Ros, 1996).
Less interdependent and more independent self-construal is
more typical of hierarchical cultures – and not of individualism.
Interdependent and less independent self-construal is more
highly stressed in egalitarian cultures – and not in collectivist cul-
tures. Group Loyalty and low Uniqueness were related to low
power distance
Even more important, interdependence was also related to coop-
erative cultural values and independence to competitive cultures.
Group Loyalty, Relational interdependence and high Contextual
Self were related to cultural femininity.
Our study in part confirms that Uniqueness is a core attribute of
an individualist self-construal. Social development and higher
parents’ status were positively associated with Uniqueness.
Independence, personal achievement, direct communication
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
57
MEP 1/2005 18/04/05 17:33 Page 57
and competition are unrelated in this and other studies to indi-
vidualism (Hofstede, 2001). Independence, Uniqueness and high
scores in low contextual self were specifically associated with
competitive cultures.
Nevertheless, the bivariate and multivariate effect sizes explained
no more than 5% of the variance, suggesting that between nation
differences are modest and do not oppose radically different self-
concepts. Moreover, with the exception of the Interdependent
and Group Loyalty factors, the others show low reliabilities.
With the results obtained in our studies, and taking into account
mentioned limitations, we can conclude that those people (at
least university students) living in developed countries, and
cooperative cultures show a (slightly more) relational-interde-
pendent self-construal, and that people living in individualist,
hierarchical and competitive cultures share a (slightly less)
group-dependent and a (slightly more) independent self-con-
strual.
Future studies should “deconstruct” the interdependence and
independence dimensions, differentiating in-group loyalty, which
is more related to collectivism, from relatedness, linked to egali-
tarian and cooperative values, more than to collectivist values.
Independence measures should also differentiate sense of
uniqueness, related to individualism, from low contextual self,
more related to hierarchical and competitive values than to indi-
vidualism. Our present study, coupled with other research
conducted in this same area (and summarized in Oyserman et
al’s review), shows how important it is to overcome a dichoto-
mous analysis which “collapses” attitudes and beliefs into
stereotypical images. This study also reinforces the importance of
going beyond individualism-collectivism (Schwartz, 1994) and of
taking into account hierarchical values, or the vertical versus hor-
izontal cultural dimension, and competitive values in order to
explain national differences in attitudes and beliefs.
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
58
MEP 1/2005 18/04/05 17:33 Page 58
References
Brislin, R. W. (1986). The wording and translation of research
instruments. In W. J. Lonner & J. W. Berry (Eds.). Field methods in
cross – cultural research (pp. 137 – 164). Newbury Park, CA:
Sage.
Coon, H.M. & Kemmelmeier, M. (2001). Cultural orientations in
the United States: (Re) examining differences among ethnic
groups. Journal of Cross-Cultural Psychology, 32, 348-364.
Cordelier, S., & Didiot, B. (1997). The world: Annual world
review of economy and geopolitics, Paris: La Decouverte.
Cross, S.E., Bacon, P.L. & Morris, M.L. (2000). The relational-inter-
dependent self-construal and relationships. Journal of
Personality and Social Psychology, 78, 791-808.
Fernández, I., Carrera, P., Sánchez, F., Paez, D., & Candia, L.
(2000). Differences between cultures in emotional verbal and
non-verbal reactions. Psicothema, 12, 83-92.
Fiske, A.P. (2002). Using Individualism and Collectivism to com-
pare cultures - a critique of the validity and measurement of the
constructs: Comment on Oyserman et al. (2002). Psychological
Bulletin, 128, 78-88.
Furnham, A., Bond, M.H., Heaven, P., Hilton, D., Lubel, T.,
Masters, J., Payne, M., Rajamanikan, R., Stacey, B. & Van Daalen,
H. (1993). A comparison of Protestant work ethics in 13 nations.
Journal of Social Psychology, 133, 185-197.
Gabriel, S. & Gardner, W.L. (1999). Are there «his» and «hers»
types of interdependence? The implications of gender differ-
ences in collective versus relational interdependence for affect,
behavior and cognition. Journal of Personality and Social
Psychology, 77, 642-655.
Goodwin, R. (1999). Personal relationships across cultures.
London: Routledge.
Gudykunst, W.B., Yang, S.M. & Nishida, T. (1987). Cultural differ-
ences in self-consciousness and self-monitoring.
Communication Research, 14, 7-34.
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
59
MEP 1/2005 18/04/05 17:33 Page 59
Hofstede, G. (1998). Masculinity and Femininity. Thousand
Oaks: Sage.
Hofstede, G. (2001). Culture’s Consequences. Second Edition.
Thousand Oaks: Sage.
Inglehart, R. (1997). Modernization and Post-Modernization.
Cultural, Economic and Political Changes in 43 Nations.
Princeton: Princeton University Press.
Inglehart, R. & Baker. W.E. (2000). Modernization, cultural change
and the persistence of traditional values. American Sociological
Review, 65, 19-51.
Inglehart, R., Basañez, M. & Moreno, A. (1998). Human values
and beliefs: a cross-cultural sourcebook. Ann Arbor: The
University of Michigan Press.
Kashima, Y., Yamaguchi, S., Kim, U., Choi, S.C., Gelfand, M.J. &
Yuki, M. (1995). Culture, gender and self: a perspective from indi-
vidualism-collectivism research. Journal of Personality and
Social Psychology, 69, 925-937.
Kitayama, S. (2002). Culture and basic psychological theory -
Toward a system view of culture: Comment on Oyserman et al.
(2002). Psychological Bulletin, 128, 89-96.
Kurman, J. (2001). Self-enhancement: Is it restricted to individu-
alistic cultures?. Personality and Social Psychology Bulletin, 27,
1705-1716.
Lee, A.Y., Aaker, W.L. & Gardner, W. (2000). The pleasures and
pains of distinct self-construals: the role of Interdependence in
regulatory focus. Journal of Personality and Social Psychology,
78, 1122-1134.
Markus, H., Kitayama, S., Fiske, A.P. & Nisbett, R.E. (1998). The
cultural matrix of social psychology. In D. Gilbert, S.T. Fiske & G.
Lindzey (Eds.). The Handbook of Social Psychology. Fourth
Edition. Second Volume (pp.914-981). Boston: McGraw Hill.
Matsumoto, D. (2002). Culture and Psychology. Second Edition.
Belmont, CA: Wadsworth/Thomson Learning.
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
60
MEP 1/2005 18/04/05 17:33 Page 60
Matsumoto, D., Grissom, R.J. & Dinnel, D.L. (2001). Do between-
culture differences really mean that people are different?.
Journal of Cross-Cultural Psychology, 32, 478-490.
Oyserman, D., Coon, H.M. & Kemmelmeier, M. (2002).
Rethinking individualism and collectivism: Evaluation of theoret-
ical assumptions and meta-analyses. Psychological Bulletin, 128,
3-72.
PNUD (1999). Informe sobre desarrollo humano [World devel-
opment report, 1999]. Madrid: PNUD-Munidprensa.
Rime, B., Finkenauer, K., Luminet, O., Zech, E. & Philippot, P.
(1998). Social sharing of emotions: New evidence and new ques-
tions. European Review of Social Psychology, 9, 145-189.
Sastry, J. & Ross, C.E. (1998). Asian ethnicity and the sense of per-
sonal control. Social Psychology Quarterly, 61, 101-120.
Sato, T. & Cameron, J.E. (1999). The relationship between collec-
tive self-esteem and self-construal in Japan and Canada. Journal
of Social Psychology, 139, 426-435.
Scherer, K. R. (1997). The role of culture in emotion – antecedent
appraisal. Journal of Personality and Social Psychology, 73, 902
– 922.
Schwartz, S.H. (1994). Beyond Individualism-Collectivism: new
dimensions of values, In U. Kim, H.C. Triandis, C. Kagitcibasi, S.C.
Choi & G. Yoon (Eds.). Individualism and Collectivism.
Thousand Oaks: Sage.
Schwartz, S.H. & Ros, M. (1996). Values in the West: a theoretical
and empirical challenge to the Individualism-Collectivism
Cultural dimension. World Psychology, 2, 91-122.
Singelis, T. M., Triandis, H. C., Bhawuk, D. S. & Gelfand, M.
(1995). Horizontal and vertical dimensions of individualism and
collectivism: a theoretical and measurement refinement. Cross –
cultural Research, 29, 240 – 275.
Singelis, T.M. & Sharkey, W.F. (1995). Culture, self-construal and
embarrassability. Journal of Cross-Cultural Psychology, 26, 622-
644.
Singelis, T.M. (1994). The measurement of independent and
interdependent self-construal. Add reference
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
61
MEP 1/2005 18/04/05 17:33 Page 61
Singelis,.M., Bond, M.H., Sharkey, W.F. & Lai, C.S.Y. (1999).
Unpacking culture’s influence on self-esteem and embarrassabil-
ity. Journal of Cross-Cultural Psychology, 30, 315-341.
Smith, P. B. & Bond, M. H. (1993). Social Psychology across cul-
tures. New York: Harvester.
Smith, P. B. & Bond, M. H. (1998). Social Psychology across cul-
tures. (Second edition) London: Prentice Hall Europe.
Smith, P.B. & Schwartz, S.H. (1997). Values. In J.W. Berry, M.H.
Segall & C. Kagitcibasi (Eds.) Handbook of Cross-Cultural
Psychology. Volume 3. Social Behavior and Applications (pp.77-
118). Boston: Allyn & Bacon.
Smith, P.B., Dugan, S. & Trompenaars, F. (1996). National culture
and managerial values: a dimensional analysis across 43 nations.
Journal of Cross-Cultural Psychology, 27, 231-264.
Smith, P.B., Dugan, S. & Trompenaars, F. (1998), Locus of Control
and affectivity by Gender and Occupational status: a 14 nation
study, Sex Roles, 36, 51-77.
Snyder, M. (1987). Public appearances, Privates realities: the
psychology of self-monitoring. New York: W.H.Freeman and Co.
Triandis, H.C., Bontempo, R, Villareal, M.J., Asai, M. & Lucca, N.
(1988). Individualism and collectivism: Cross-cultural perspec-
tives of self-ingroup relations. Journal of Personality and Social
Psychology, 54, 323-338.
Triandis, H.C. (1995). Individualism and Collectivism. Boulder,
Co.; Westview.
Van de Vliert, E. (1998). Gender role gaps, competitiveness and
masculinity. In G. Hofstede (Ed.). Masculinity and Femininity
(pp.117-129). Thousand Oaks: Sage.
Waterman, A.S. (1981). Individualism and Interdependence.
American Psychologist, 36, 762-773.
Wheeler, L., Reiss, H.T. & Bond, M.H. (1997). Collectivism-
Individualism in everyday social life: the middle kingdom and the
melting pot. In L.A. Peplau & S. Taylor (Eds.). Sociocultural per-
spectives in social psychology: Current readings (pp.297-313).
London: Prentice Hall.
INDEPENDENT AND INTERDEPENDENT SELF-CONSTRUAL AND CULTURE
62
MEP 1/2005 18/04/05 17:33 Page 62
Annex
Argentine: H. de Pascuale, National University of Cordoba, and E.
Zubieta University of Buenos Aires.
Belgium: B. Rimé, P. Philippot and Y. Yabar. Catholic University of
Lovaina. Leuven.
Bolivia: J. Ortego. Catholic University of Cochabamba.
Brazil: L. Camino. Paraiba University, Joao Pessoa.
Chile: D. Asún. University of Diego Portales, Santiago of Chile.
China: M. Salas and Xu Lin Bao, Beijing University, Beijing.
El Salvador: W. A. Hernández. El Salvador University, San Salvador.
France: H. Paicheler and N. Aguilera, Paris VIII University.
Germany: H. C. Traue, Ulm University.
Ghana: A. Olowu, University of the Cape Coast, Cape Coast.
Greece: T. Apostolidis and N. Christakis, Univ. of Ioannina,
Athens.
Guatemala: E. Cajas, San Carlos University, Guatemala City.
Iran: B. Behrouz University of Lausanne and S. Mortzavi,
Shahid Beheshti University, Teheran.
Italy: G. Belleli, F. Stazolla. University of Bari, Bari.
Lebanon: J. Akiki, Université Saint Esprit de Kaslik, Beirut.
Mexico: J. Alvarez. Monterrey University, Monterrey, and A.
Blanco, Univ. Autonoma de Mexico, Mexico City.
Nigeria: A. Olowu, Obafemi Awolowo University, Lagos.
Panama: M. Lombardo. University of Panama, Panama City.
Portugal: R. Ramos and J. Márques (FPCE-UP), Oporto University,
Oporto.
Peru: A.M. Sono, Ricardo Palma University, Lima.
Russia: I. Bovino, Moscow University, Moscow, and A. Zlobina,
Yaroslavi State University, Tula.
Singapore: G. Bishop, The National University of Singapore.
Spain: P. Carrera and F. Sánchez (Autonomous University of
Madrid), M. López and M. Navas (U.N.E.D. Madrid and
Almeria University), J. Mª Canto (Malaga University), A.
Vergara and I. Fernández (Basque Country University), J.
M. Sabucedo (Santiago de Compostela University), F.
Martinez (Murcia University) and A. Rodríguez (La
Laguna University).
Switzerland: J-C. Deschamps. University of Lausanne, Laussane.
Taiwan: L. Cheng, Univ. of Salamanca.
Turkey: O. Paker, Ege University, Istanbul.
U.S.A.: L. Candia and L. Martínez, Kansas State University, Kansas
Venezuela: E. Casado, Central University of Caracas; Caracas, and M.
León, The Andes University, Trujillo.
The Taiwanese students were enrolled in psychology and social science
courses and were interviewed by Li Fen Chen in Mandarin while they
were attending a Spanish summer course in Salamanca (Spain).
REVUE INTERNATIONALE DE PSYCHOLOGIE SOCIALE 2005 N° 1
63
MEP 1/2005 18/04/05 17:33 Page 63
MEP 1/2005 18/04/05 17:33 Page 64
