Professional Documents
Culture Documents
Naturwissenschaftliche Fakultät
Fachbereich Psychologie
DIPLOMARBEIT
Fist of all I wish to sincerely thank Prof. Wilhelm Kempf, not only for supervising me in his
very expert and reliable manner in pursuing my own interests, but also for his confidence
and for providing over many years the best work environment I could have wished for.
Furthermore, I am very grateful to Prof. Sabine Sonnentag for additionally supervising this
work and providing a profound basis with her very proficient colloquium.
Prof. Ting-Toomey’s enthusiasm and mindful way of teaching us about intercultural conflict
are at the core of this work for which I am very thankful to her.
Dr. Willi Nagl gave me the best statistical support, I could have hoped for. I usually went to
his office with doubts, but I always left laughing. I would really like to thank him for that.
Moreover, I would like to thank Mr. Claus Hamdorf, Mr. Michael Diel, Mr. Rui Paulo Vigário,
Mr. Miguel Vigário, Mr. Jorge Matos, Mrs. João Matos, and all the other persons who have
assisted this work’s first steps into an international business environment. I was very
impressed with the positive experience, I was able to make when calling companies and I
would like to express my gratitude to all the persons I contacted for showing so much
interest and forwarding my e-mail to your colleagues and friends.
That this thesis is here today is to a big extent due to my family. I wish to sincerely thank
them for their patience and continuous support.
Furthermore, I am very thankful to my extended homes in Finland and Portugal for keeping
me in close contact with my own cultural ‘heritage’ and with their great ways of living an
international culture.
To my friends and colleagues at the University of Konstanz, especially Anne Rohn, André
Jenczmionka, Claus Rüegg, Andreas Kupka, Ralf Tichatzky, Ute-Vera Bayer, Judith
Glaesser, Verena Friedrich, Sybille Kruttschnitt and Heike Deyl I would like to tell how much I
appreciated our discussions, your helping hands, and all our joy.
Special thanks go to André for his professional proofreading, to Bettina for her thoughtful
comments on first drafts and to Georg for his ‘statistic hotline’. Furthermore I would like to
thank Ute and Norbert Annabring and everybody joining Thursday evenings for their
continuous encouragement and our good times.
To Ricardo I wish to express my deepest gratitude for all his help and patience, for
broadening my worldview, creating the very basis for this work and for making intercultural
conflicts so worthwhile!
Last but not least, I would like to sincerely thank everybody who took part in this study for
‘helping it onto its feet’!
th
Konstanz, 25 June 2002
ABSTRACT
Results revealed that neither did independent persons indicate to perceive conflicts in a
rather issue- or task-oriented manner, nor did interdependent persons point out to
perceive them in a rather relationship-oriented manner, as was originally expected.
Furthermore, these perception tendencies were not visible either in the definition of
integrating and compromising conflict management styles. Regarding conflict behavior,
independent persons pointed out direct styles, as well as integrating. Interdependent
persons, on the other hand, indicated indirect styles in addition to integrating and
compromising. Furthermore, a concern for self-face maintenance in conflict situations
was related to direct conflict behavior, a concern for other-face maintenance to indirect
and a concern for mutual-face maintenance to integrating and compromising conflict
management styles.
However, independent persons do not seem to be particularly concerned with self-face maintenance in a
conflict situation. Interdependent persons, on the other hand, appear to show other- and mutual-face
concerns. It was then concluded that face concerns do play a crucial role, but mainly in explaining an
interdependent person's conflict behavior. This was supported by the fact that other-face concern mediated
interdependent persons' tendency for conflict avoidance. These results are discussed and implications for
further research and practice are presented.
meinen Eltern
TABLE OF CONTENTS
1 INTRODUCTION 1
1.1 Introduction 1
1.2 Outline of the Study 2
2 THEORETICAL BACKGROUND 5
2.1 Culture 5
2.1.1 Culture and Organization 7
2.1.2 Culture and Self-concept: Independent and Interdependent Self 7
2.2 Conflict 10
2.2.1 Conflict and Organization 13
2.2.2 Conflict Perception: Task- / and Relationship Orientation 14
2.2.3 Conflict Behavior: Conflict Management Styles 14
2.3 Culture and Conflict 16
2.3.1 Culture & Task- and / Relationship-Orientation: 16
an Overview of their Connections and Recent Research Findings
2.3.2 Culture & Conflict Management Styles: 18
an Overview of their Connections and Recent Research Findings
2.3.3 Cultural Aspects of Conflict Management: So Far 19
2.4 The Face-Negotiation Theory 21
2.4.1 Face 21
2.4.2 The Face-Negotiation Theory 22
2.4.3 Culture & Face: 22
an Overview of their Connections and Recent Research Findings in the
Light of the Face-Negotiation Theory
2.5 Hypotheses 24
3 METHODS 27
3.1 Participants 27
3.2 Procedure 29
3.3 Measures 30
3.3.1 Independent and Interdependent Self-construal 30
3.3.2 Face-concerns 31
3.3.3 Task- and / Relationship-Orientation 31
3.3.4 Conflict Management Styles 32
3.4 Statistical Analyses 33
3.5 Specification of the Hypotheses 36
4 RESULTS 38
4.1 Descriptive Statistics and Correlations 38
4.2 Hypotheses Testing 41
4.2.1 Self predicting Conflict Perception/Orientation 41
4.2.2 Self predicting Style Definition 43
4.2.3 Self predicting Conflict Behavior 47
4.2.4 Self predicting Face-concerns 54
4.2.5 Face-concerns predicting Conflict Behavior 56
4.2.6 The Mediation Role of Face-concerns 64
4.3 Further Analyses of the Data 67
5 DISCUSSION 70
5.1 Discussion of the General Findings 70
5.1.1 Self predicting Conflict Perception/Orientation 70
5.1.2 Self predicting Style Definition 70
5.1.3 Self predicting Conflict Behavior 72
5.1.4 Self predicting Face-concerns 75
5.1.5 Face-concerns predicting Conflict Behavior 76
5.1.6 The Mediation Role of Face-concerns 78
5.1.7 Further Methodological Considerations 79
5.1.8 Brief Summary So Far 81
5.2 Implications for Further Research 83
5.3 Implications for Practice
86
6 REFERENCES
90
7 APPENDIX 97
Introduction 1
1 INTRODUCTION
1.1 Introduction
On the tide of globalization, companies increasingly expand, merge and form strategic
alliances across national boarders. The success and failure of those international
collaborations does not only depend on managing economic and legal obstacles.
Negotiations with partners from other cultures, temporary dispatch of workforce to foreign
countries and the setup of multinational work teams are on the day’s agenda. This
creates new challenges for companies’ employees and human resource departments.
Usual work demands are supplemented with cultural differences in partners’ values and
norms of behavior. Thus, management of diversity has increasingly become a topic for
company personnel and organizational researchers. Yet, little is known about what this
‘management’ really means (e.g. Erez & Earley, 1993).
A large number of mergers and acquisitions fail (e.g. Erez & Earley, 1993). Many
assignments of expatriates end before their official contract termination (e.g. Black et al.,
1991). Furthermore, there is increasing evidence that conflict is even more omnipresent
in multinational companies, where intercultural adaptation has to take place (e.g. Gladwin
& Walter, 1980).
These problems, mostly arising from intercultural conflicts go together with substantial
costs for the companies involved. In addition, intercultural conflicts and subsequent
failures in job performance have a strong impact on employees’ well-being and
productivity. Stress, loss of self-security and trust in one’s own competences, as well as
degrading of one’s carrier may be the results (e.g. Black et al. 1991). Moreover, the
promotion or fostering of prejudices and stereotypes are very probable. Thus, conflicts in
general, and intercultural conflicts in particular, go along with high psychological, social
and economical costs.
To date, however, it is commonly agreed that conflicts are neither negative nor positive
per se. It is the way they are framed and managed that leads to constructive or
destructive outcomes (e.g. Deutsch, 1976; Rahim, 1986; Tjosvold & Tjosvold, 1995;
Kempf et al., 1996).
Thus, to constructively manage conflicts, knowledge about the conflict conditions and
processes is needed. Furthermore, to constructively manage intercultural conflicts, i.e. to
manage diversity in organizations, additional knowledge about the cultural influence on
conflict conditions and processes is required, as “managing intercultural conflict
constructively means managing cultural-based conflict differences appropriately and
effectively” (Ting-Toomey, 1999, p.195).
Thus, the present study will focus on cultural aspects of conflict management in
organizations. Culture, conflict and organization are abstractions of highly complex
phenomena. In the Theory chapter, their understanding for the present research project
will be clarified.
Conflict behavior
- dominating
- integrating
- compromising
Face-concerns - obliging
- self-face - emotion
- other-face - neglect
- mutual-face - third party help
2 THEORETICAL BACKGROUND
2.1 Culture
Culture is one of the most often defined concepts within social sciences. In common
language, culture typically refers to ‘civilization’ or ‘refinement of the mind’. In particular, it
relates to the results of such refinement, like education, art, and literature (Hofstede,
1997). Formally, however, the social sciences have a much broader understanding of
culture. After reviewing more than one hundred definitions, the anthropologists Kroeber &
Kluckhohn (1952, in Adler, 1997) offered one of the most comprehensive and generally
accepted definitions:
“Culture consists of patterns, explicit and implicit, of and for behavior acquired
and transmitted by symbols, constituting the distinctive achievement of human
groups, including their embodiment in artifacts; the essential core of culture
consists of traditional (i.e. historically derived and selected) ideas and especially
their attached values; culture systems may, on the one hand be considered as
products of action, on the other, as conditioning elements for future actions” (p.
14).
Culture was thus thought to be to society what memory is to individuals (Triandis, 1995).
It provides an orientation system for one’s own behavior and for the interpretation of the
one of others. Thomas (1996) defines the characteristic features of such orientation
systems as culture standards. They refer to “all perceptive, cognitive, evaluative and
behavioral processes, which are considered normal, natural and binding for most
members of a society” (p.112). These standards are learned since early childhood
through socialization and are mostly unconscious, until one encounters with others.
A systematical recognition of the role of culture, in the fields of Psychology in general and
of Industrial and Organizational Psychology in particular, is fairly recent (e.g. Erez &
Earley, 1993). There, the influence culture exerts on individual experience and behavior
is commonly the focus of interest. More specifically, attention is nowadays attached to
the influence that certain explicitly measured aspects of culture exert, rather than simply
mentioning and comparing countries of origin (Triandis, 2000).
Hofstede’s (1980) culture dimensions are probably one of the best-known approaches to
provide a set of measurable aspects of cultural diversity. He conducted a questionnaire
study about work goals and values in IBM subsidiaries in 53 nations and regions and
found four dimensions to distinguish between these countries: individualism-collectivism
(IC), power-distance (PD), masculinity-femininity (MF) and uncertainty avoidance (UA)
(e.g. Hofstede, 1980, 2001 for a review).
Since, various research groups have tried to replicate those dimensions (e.g. Chinese
Culture Connection, 1987; Smith, Dugan & Trompenaars, 1996). IC was described as
perhaps the most important dimension and was used thereafter to explain a variety of
differences between individuals (e.g. Triandis, 1995 for a review).
The terms individualism and collectivism are widely used and were given various
meanings. Moreover, several researchers have defined IC, somewhat differing in their
emphasis (e.g. Hofstede, 1980, 1997, 2001; Triandis, 1994, 1995). This study will
therefore draw upon a ‘summary’ presented by Ting-Toomey et al. (2000). It states that
in essence, ”individualism refers to the broad value tendencies of a group in emphasizing
the importance of individual identity over group identity, individual rights over group
rights, and individual interests over group or relational interests.
In comparison, collectivism refers to the broad value tendencies of a group in
emphasizing the importance of the ‘we’ over the ‘I’ identity, group obligations over
individual rights, and in-group oriented needs over individual wants and desires” (p. 51).
While both sets of value tendencies exist in all cultures and persons in various
combinations, their relative importance is emphasized differently in predominantly
individualistic and collectivistic cultures. Depending on these emphases, people view
their world through different lenses, thus attaching different meanings to life events and
behaving differently in accordance to them (Triandis, 1995).
2.1.1 Culture and Organization
Organizational culture seems to be close to the subject of the present project and will
therefore be discussed briefly.
According to Erez and Earley (1993), organizations do not have cultures of their own, but
are rather formed as a function of societal culture. To them, organizational culture is to
organizations, what norms are to culture: culture tells people the inherent meaning of a
context, whereas norms tell people how to behave. Various studies have shown that
matching a companies’ organizational culture with societal culture and values results in
higher job satisfaction, higher effectiveness and easier decision-making (Erez & Earley,
1993).
On the other hand, many managers believe that employees working in the same
company, though with possible diverse cultural backgrounds, are more alike than
different (Adler, 1997). Besides, in times of globalization and proliferation of multinational
companies, it is reasonable to question, whether or not organizations themselves
become more similar. On the macro level (e.g. the technology employed by companies),
they indeed appear to grow comparable worldwide. However, on the micro level (e.g. the
behavior of people within organizations), they seem to preserve their cultural uniqueness
(Adler, 1997). Hofstede (1980), for example, found differences in employees’ attitudes
and behavior across his sample of cultures, even though his study was conducted within
one single company. Furthermore, it has been shown that within a given country, even if
companies differ in their structures, i.e. organizational cultures, behaviors are still closer
than when comparing behaviors in companies across countries (Kappe, 1996). It was
therefore decided not to consider organizational culture separately in the present study.
A variety of scholars have pointed out that it is problematic to use culture-level variables,
like IC, to explain cultural differences in individual-level behavior (e.g. Kashima, 1989;
Kagicibasi, 1994; Triandis, 1994; Singelis & Brown, 1995; Gudykunst et al., 1996).
In particular, Kashima (1989) argues that it is impossible to test causal explanations for
individual behavior based on culture-level factors, as for example, culture cannot be
controlled in an experiment. Furthermore, he points out that specific samples may not
correspond to the global culture-level scores of IC.
Kagicibasi (1994) therefore suggests that for testing causal explanations, researchers
should identify the underlying psychological processes related to IC. Following the same
line, Singelis and Brown (1995) state that “the popular use of IC requires an explanation
of the mechanisms and intermediate steps through which the various pressures inherent
in this broad-based construct shape individual behavior” (p. 355). One such mechanism
seems to be the self.
As mentioned earlier, a person’s cultural background strongly influences his or her
development. This was shown to result in different ways of construing one of the most
general aspects of a person’s self-concept: the way he or she relates to other persons
and the social environment. These differences in self-concept were referred to as
independent and interdependent self-construals (Markus & Kitayama, 1991). They
influence the way a person perceives and interprets his or her environment, which in turn
shapes his or her behavior. A person’s self-concept is thus supposed to be a mediator for
the cultural influences on behavior in general, and organizational behavior in particular
(Erez & Earley, 1993).
The two key aspects, independent and interdependent self-construals differ in, are their
structure of self and the role of others in its definition.
The independent self-construal involves the view of oneself as a unique and independent
person. This understanding of oneself is rather detached from the social context. A ‘core
self’ (e.g. “I am a kind person”) is carried along a variety of social situations, which
implies that the structure of an independent self is relatively stable.
Thus, in essence, having an independent self-concept means “construing oneself as an
individual, whose behavior is organized and made meaningful primarily by reference to
one’s own internal repertoire of thoughts, feelings, and actions, rather than by reference
to thoughts, feelings, and actions of others” (Markus & Kitayama, 1991; p. 226).
In cultures that promote the independent self-construal, the normative imperative is to
become autonomous and to discover and express one’s own unique attributions; a
concept that was referred to as ‘standing out’ (Weisz, Rothbaum & Blackburn, 1984).
How well a person fulfills these developmental tasks is the basis for his or her self-
esteem.
The interdependent self-construal, on the other hand, involves the view of oneself as
interrelated with other persons. This understanding of oneself is rather dependent on the
social context. It varies with the characteristics of the social situation (e.g. “I am a kind
person with my family”), which implies that the structure of an interdependent self is
relatively variable.
Thus, in essence, having an interdependent self-concept means “seeing oneself as part
of an encompassing social relationship and recognizing that one’s behavior is
determined, contingent on, and, to a large extent, organized by what the actor perceives
to be the thoughts, feelings, and actions of others in the relationship” (Markus &
Kitayama, 1991; p. 227).
In cultures that promote the interdependent self-construal, the normative imperative is to
integrate oneself, engage in appropriate actions, and maintain the interdependence
among individuals (Markus & Kitayama,1991; Matsumoto, 1996); a concept that was
referred to as ‘standing in’ (Weisz, Rothbaum & Blackburn, 1984). How well a person
fulfills these developmental tasks, again provides the basis for his or her self-esteem.
Various research outcomes support the fact that these self-concepts are, to a large
extent, products of culture.
2.2 Conflict
Conflicts are a pervasive aspect of existence. They may occur at all levels of social life,
may involve a variety of issues, and are complex because of many overlapping
processes (e.g. Deutsch, 1991; Isenhart & Spangle, 2000; Kempf, 2002). Furthermore,
the different disciplinary backgrounds of the various scholars investigating conflict,
together with their focuses on different types of conflict, have given its study a
fragmented appearance.
According to Deutsch (1991), however, several common themes seem to exist within the
different approaches, which he summarizes in the following propositions:
most conflicts involve both cooperative and competitive motives;
conflicts can be constructive as well as destructive;
the relative strength of the motives will strongly determine, whether more
cooperative or competitive processes arise;
the relative strength of the motives and their variation over the duration of the
conflict process will strongly determine, whether conflict outcomes will be more
constructive or destructive.
Moreover, within the behavioral sciences, two broad usages of the word conflict exist.
The first one refers to incompatibilities of goals and/or behavioral tendencies within a
single person, and thus to intrapersonal conflicts. The other describes incompatibilities of
goals and/or behaviors between two or more parties, which may be individuals, groups,
organizations or other social units, and thus refers to interpersonal or social conflicts
(Thomas, 1992; Kempf, 2001).
A further distinction regards the fact that social conflicts may be either latent or manifest.
Latent conflicts refer to incompatibilities between conflict parties, which they are not
aware of. As soon as incompatibilities become conscious, manifest conflicts will result
(e.g. Kempf, 2001; Kempf, 2002).
Manifest social conflicts are the focus in most literature on organizational disagreements
and will also be the issue in the present study.
Building on Ponsy’s (1967) argumentation for a broad working definition, most of the
recent theorists agree to understand manifest social conflict as a process. Along this line,
Thomas (1992) proposes the following model:
Third-party interventions
Primary effects
Feedback effects
Conflict occurs as a process or a sequence of events. These events take place in conflict
episodes between the conflict parties and include the parties’ internal experiences, like
cognitions and emotions, as well as their externally visible behavior. Such episodes have
an inherent logic, in which previous events cause later ones (Thomas, 1992; Kempf et
al., 1996). More precisely, the way one party experiences the conflict situation and
subsequently acts, will influence further experiences and succeeding behaviors of him- or
herself, as well as of the other party. This means that in interpreting the conflict situation,
the conflict parties are mutually dependent, which results in a complex co-construction of
meaning (Kempf, 2002). Furthermore, this process is influenced by structural conditions,
which include properties of the conflict parties as well as of the context in which they
interact. Shaped by structural conditions, the conflict process produces outcomes, which
comprise consequences for both issue/task accomplishment and social system
maintenance. Moreover, third party interventions may alter the sequence of events. In an
organizational environment, managers or mediators, for example, may intervene during
the conflict process or initiate changes on the social/organizational level.
The interrelations among the components of the model are assumed to be highly
complex. For simplicity, Thomas (1992) presents here the causal relationships and their
feedback effects. The feedback effects serve as reminders of the fundamental
interconnectedness of all variables, in that, structural variables, for example, are often
also affected by the conflict process.
To summarize, conflicts are processes of complex co-construction of meaning (Kempf,
2002), may involve incompatibilities over a variety of aspects concerning issues or
relationships (Ting-Toomey, 1994), are influenced by objective and subjective factors,
and produce objective and subjective outcomes (Deutsch, 1976; Deutsch, 1991;
Thomas, 1992; Kempf et al. 1996; Kempf, 2002).
Because of the quantitative nature of the present study, a precise definition of conflict
was required, in order to allow for comparisons between participants’ responses (Berkel,
1984). Furthermore, as this study focuses on cultural aspects of conflict management, a
‘culture sensitive’ description was chosen. Conflict was therefore defined as “perceived
and/or actual incompatibilities of values, expectations, processes, or outcomes between
two or more parties (…) over substantive and/or relational issues” (Ting-Toomey, 1994,
p. 360).
Organization is classically defined as an open system, which exists over a longer period
of time, pursues specific goals, consists of groups of individuals and has a certain
structure, which is mostly characterized by division of labor and hierarchy of
responsibilities (v. Rosenstiel, 1992 in Spieß, 1996, p.121).
From an organization-psychological viewpoint, the understanding of organization is
generally open to a blending with or separation from other areas of life (Berkel, 1984).
This means that it depends on a person’s comprehension of organization, to which extent
he or she behaves similarly at work and at home.
Still, an organization represents a specific context for behavior. It holds particular
characteristics, which make some types of behavior more expectable than others
(Berkel, 1984). Furthermore, interactions in organizations typically occur between
acquaintances, often characterized by a certain distance. Hence, when dealing with
problematic issues, individuals are expected to rely more on habituated modes of conflict
handling, than when in for example intimate relationships or situations with less time
constraints (Ting-Toomey et al., 2000).
Summarizing this section, it should be noted that all outcomes of this study regard
conflict perception and behavior in an organizational setting and may therefore not
necessarily apply to other contexts. Henceforth, when talking about conflict management,
the organizational frame is assumed.
2.2.2 Conflict Perception: Task- / and Relationship Orientation
In this study, task- / and relationship orientation refers to the fact that a person may be
primarily concerned with the conflict issue or his or her relationship with the other conflict
party. Thus, what is presented here is a difference in the way the conflict situation and
subsequent events are perceived and interpreted. It is a general focus of attention, which
will influence a person’s cognitions and emotions and in turn shape his or her behavior.
To separate task and relationship orientation may appear to be an academic distinction,
as each conflict usually involves both task, i.e. issue, and relationship aspects. However,
as ‘cultures’ seem to differ in their perception-orientation (which will be presented in
section 2.3.1) the distinction was considered important for this project.
Rahim (1983) developed one of the best-known scales to measure conflict behavior,
using these two dimensions (see Figure 3). The first dimension explains the degree to
which a person attempts to satisfy his or her own concerns. The second dimension
explains the degree to which a person wants to satisfy the concerns of others. The two
dimensions lead to five styles of conflict management:
dominating: (high in self-concern, low in other-concern),
which involves enforcing one’s own interests, even at the
expense of those of the other party;
dominating integrating
(win-lose) (win-win)
high
concern for self
compromising
avoiding obliging
(lose-lose) (lose-win)
low
low high
Figure 3. Styles of interpersonal conflict management. Model and terminology adapted from M. A. Rahim,
1983. Modes of presentation adapted from K. W. Thomas, 1992.
2.3 Culture and Conflict
To briefly summarize the previous sections: (1) cultures reinforce different self-concepts,
which lead persons to view their world through different lenses and behave accordingly;
(2) conflicts are complex co-constructions of meaning, which are influenced by structural
aspects and individual experiences.
It thus seems expectable that members of different cultures vary in their conflict
perception and conflict behavior within their workplace. The question, this study is now
concerned with, is how.
The interest in cultural aspects of conflict management grew rapidly over the past years.
Yet, the existing literature on this topic is not as comprehensive as in other sub fields of
organizational research (Kozan, 1997) and still generates many inconsistent outcomes.
In many non-Western cultures people are brought up with the philosophy of being part of
nature and to live in harmony with it (e.g. Rothbaum & Tsang, 1998). As a consequence,
the relationship between the self and other, or between subject and object is assumed to
be much closer. Therefore, most of these cultures insist on the inseparability of basic
elements, such as self and other, or person and situation (Galtung, 1981).
Kim et al. (1994) studied the relationship between self-construal and the perceived
importance of interactive constraints. They found that an independent self would
correlate positively with a concern for clarity, which was considered to be outcome- or
task-oriented. An interdependent self was positively correlated with concerns about the
other’s feelings and the avoiding of negative evaluations, which were considered to be
relationship-oriented.
Within an organizational setting, Oetzel & Bolton-Oetzel (1997) studied the relationship
between self-construal and dimensions of work-group effectiveness. Their results
indicate that task effectiveness was better predicted from an independent self-construal,
while relationship effectiveness was better predicted from an interdependent self-
construal.
It is important to note that these research outcomes, together with the suggestions made
by Ting-Toomey (1988) and Ting-Toomey & Kurogi (1998), certainly do not imply that in
individualistic cultures conflicts can’t be personal or that in collectivistic cultures solutions
for the task issues can’t be found! However, because of their different cultural
backgrounds, it is expected that persons from individualistic and collectivistic cultures
have different ways to perceive and interpret a conflict situation, i.e. different conflict
orientations and thus different approaches of conflict management.
2.3.2 Culture & Conflict Management Styles:
an Overview of their Connections and Recent Research Findings
third party help which involves the preference for a third party in helping to
settle the conflict.
To investigate the relationship between culture and conflict management styles, Ting-
Toomey et. al. (1991) compared self-reports from Japanese, Chinese, South Korean,
Taiwanese, and American students. Their findings show that American students used
dominating behavior to a higher degree when compared to their predominantly
collectivistic-oriented colleagues who employed more avoiding and obliging styles. The
results concerning integrating and compromising were mixed.
Another research group studied the influence of IC on self-reported conflict management
styles for Anglo-American and Taiwanese students. Taiwanese subjects were found to
use avoiding, obliging, integrating, and compromising styles more often than their
American counterparts (Trubinsky et al., 1991).
To examine the relationship between culture and conflict management styles in an
organizational setting, Elsayed-Ekhouly & Buda (1996) questioned employees in
companies located in the Middle Eastern countries as well as in the United States. Their
inventory data showed Middle Eastern executives to display more integrating and
avoiding, while U.S. executives used more obliging, dominating and compromising
styles.
Jordanian, Turkish and U.S. managers were interviewed by Kozan (1989) about their
conflict behavior towards superiors, peers and subordinates. Regarding conflict
management with peers, he found that managers of all three countries rather seem to
avoid conflicts.
A study comparing U.S. managers and managers from different Asian backgrounds, all
working in Singapore, has also shown that in general, IC tends to differentiate Americans
from their Asian counterparts in their use of styles. The picture, however, was not as
clear-cut, as expected (Wu et al., 2001).
1. In most studies up to date, IC was not measured, but assumed, drawing upon
Hofstede’s (1980) IC scores. This approach is already problematic in itself, as
culture is not assessed directly. Furthermore, as seen, IC is a culture-level
variable. To determine the influence of cultural IC on individual behavior, the
cultural aspects of a person’s self-concept are supposed, and were shown, to be
better predictors (e.g Gudykunst et al., 1996; Oetzel, 1998).
The present study will investigate cultural aspects of conflict management via
independent and
interdependent self-construals.
2. ‘Individualists’, or independent persons seem to be generally more task-oriented.
‘Collectivists’, or interdependent persons appear to be generally more
relationship-oriented. However, this orientation tendency was never directly
studied concerning conflict perception.
The present study will investigate task- / and relationship-orientation in conflict
perception.
3. Summarizing previous research results about conflict behavior, two general lines
seem to exist. One, which connects individualism and independent self with a
dominating style, and another, which connects collectivism and interdependent
self with avoiding and obliging styles.
Integrating and compromising are sources of various inconsistencies. Ting-
Toomey & Kurogi (1998) suggest that a reconceptualization of these styles may
be needed. Depending on culture, the different meanings of either involving a
substantive-level or relational-level collaboration may be attached.
The present study will attempt to measure if independent and interdependent
persons differ in their definition of integrating and compromising styles.
4. Additionally, most studies in this field draw upon student populations, not
employees in organizational settings. When studying conflict behavior in
organizations, results seemed less clear-cut.
The present study will investigate cultural aspects of conflict management directly
in an organizational setting.
2.4.1 Face
Besides one’s self-concept, the public presentation of oneself plays a crucial role in
understanding social behavior (Wolfradt, 1996). This self-presentation is what several
scholars call ‘face’.
“Face, in essence, is a projected image of one’s self in a relational situation. It is an
identity that is conjointly defined by the participants in a setting” (Ting-Toomey, 1988,
p.215).
Face is associated with such concepts as respect, status, competence, and loyalty (e.g.
Ting-Toomey & Kurogi, 1998). It is a vulnerable resource, because it can be lost or
enhanced in any uncertain social situation and is thus something that must be constantly
attended to during interaction (Brown & Levinson, 1978; Ting-Toomey & Kurogi, 1998;
Oetzel et al., 2000). At the same time, it includes affective – like feelings of pride or
shame -, cognitive – like how much to give and receive face –, and behavioral layers
(Ting-Toomey & Kurogi, 1998; Oetzel et al., 2000).
To worry about one’s own image is referred to as self-face concern, whereas caring for
another person’s image is referred to as other-face concern. Mutual-face concern, first
introduced by Ting-Toomey (1988), is understood as the concern for both parties’ images
and/or the image of the relationship.
2.4.2 The Face-Negotiation Theory
Oetzel at al. (2000, p.11) summarize the core of the face-negotiation theory as follows:
(1) people in all cultures try to maintain and negotiate face in all communication
situations;
(2) the concept of ‘face’ is especially problematic in uncertain situations (such as
embarrassment or conflict), when the situated identities of the conflict parties are
called into question;
(3) cultural variability (e.g. IC), individual-level variables (i.e. independent,
interdependent self), and situational variables (e.g. characteristics of the conflict
situation, the relationship between the parties, etc) influence cultural members’
selection of face concerns (i.e. self-face versus other- and/or mutual-face
concern);
(4) cultural variability, individual-level variables, and situational variables influence
conflict management behavior in interpersonal and intergroup encounters.
Members of all cultures have and wish to maintain face (Goffman, 1967; Brown &
Levinson, 1978; Ting-Toomey, 1988; Ting-Toomey & Kurogi, 1998). Face, as has been
shown, is based on the construction of a ‘public self-image’. However, which ‘self’ is
projected into this self-image depends on one’s cultural self-concept. Furthermore, the
meaning of face in a certain situation, as well as facework - how to give or threaten face -
vary from one culture to another (Ting-Toomey, 1988; Ting-Toomey & Kurogi, 1998).
As mentioned, in individualistic cultures, the self is often defined as an intrapsychic
phenomenon. In social interactions, a person is relatively ‘free’ to decide how to behave.
The consistency between the ‘private self’ (‘who one is’) and the public presentation of
oneself, i.e. ‘the public self’ (‘how one wishes to appear in front of others’) is perceived to
be important. The ‘public self’ should therefore ideally correspond to a person’s ‘private
self’ and his or her internal states. In other words, one expresses who one is, what one
thinks and feels and behaves in accordance to one’s believes and attitudes.
In this context, face-giving and face-saving offers support to another person’s self
presentation, thus, ideally to this person’s ‘private self’ (Wolfradt, 1996).
Here, facework emphasizes the preservation of one’s own autonomy, own territory, and
own space, while simultaneously respecting the other person’s need for space and
privacy (Ting-Toomey, 1988).
Ting-Toomey et al. (1991) investigated the links between culture and face maintenance
and between face maintenance and conflict behavior. Results of a questionnaire study
involving five cultures have shown a higher degree of other-face concern for collectivistic
cultures. Other-face concern in turn was strongly associated with avoiding, integrating
and compromising styles. Self-face concern was strongly related to dominating conflict
behavior. Furthermore, face maintenance dimensions were shown to be better predictors
for conflict management styles than vice versa.
Concerning the connection between self-construal and face maintenance, Oetzel et al.
(2000) demonstrated that self has a stronger effect on face concerns than culture and
situational influences. Independent self is positively associated with self-face concern
and interdependent self with other- and mutual face-concerns.
No documentation concerning the relationships between self, face and conflict
management styles in organizational settings was found.
2.5 Hypotheses
Several scholars suggest that, at least, dominating and emotion are direct, whereas
avoiding, obliging, neglect and third party help are indirect conflict management styles
(Ting-Toomey, 1988; Ting-Toomey & Kurogi, 1998; Ting-Toomey et al., 2000).
Integrating and compromising conflict behaviors were considered to be rather direct.
However, persons from both individualistic and collectivistic backgrounds are expected to
use them, though possibly defining them differently.
Based on the above reasoning and the presented research findings, the following
hypotheses are proposed:
H 3.2: Interdependent persons tend to express indirect conflict management styles, such
as
(a) avoiding
(b) obliging
(c) neglect
(d) third party help.
In addition, they also tend to express
(e) integrating
(f) compromising.
4. Self predicting Face-concerns
H 4.1: In a conflict situation independent persons tend to be concerned with self-face
maintenance.
H 5.2: Persons who are concerned with other- and/or mutual-face maintenance tend to
express indirect conflict management styles, such as
(a) avoiding
(b) obliging
(c) neglect
(d) third party help.
3 METHOD
3.1 Participants
The sample in this study included 185 professionals from different countries. The majority
(156, 84 %) was European. A complete listening of the participant’s national
backgrounds is shown in Table 1.
(a) (b)
(c) (d)
(e) (f)
(g) (h)
Figure 4. Detailed description of the demographics. Sample distributions of (a) ‘age’, (b) ‘gender’; (c)
‘education’; (d) ‘company size’; (e) ‘tenure’ (in years within the company); (f) ‘position’ (according
to management duties); (g) ‘having lived abroad’ and (h) ‘understanding of English’.
3.2 Procedure
3.3 Measures
The questionnaire with all scales employed in this study is presented in Appendix B.
Items were assessed using a 5-point Likert scale ranging from “fully disagree” (1) to “fully
agree” (5).
The Cronbach’s alpha for the entire scale was .70, with .75 for the sub-scale of
independent self and .68 for the sub-scale of interdependent self.
3.3.2 Face-concerns
Face concerns were assessed by twenty-two items written by Ting-Toomey and Oetzel
(in Oetzel et al., 2000).
Self-face concern was measured by seven items, e.g., “I was concerned with
protecting my self-image”.
Other-face concern was assessed by eleven items, e.g., “I was concerned with
helping the other person to maintain his/her credibility”.
Mutual-face concern was measured by four items, e.g., “I was concerned with
respectful treatment for both of us”.
Items were assessed using a 5-point Likert scale ranging from “fully disagree” (1) to “fully
agree” (5).
The Cronbach’s alpha for the entire scale was .92, with .89 for the sub-scale of self-face
concern, .92 for the sub-scale of other-face concern and .80 for the sub-scale of mutual-
face concern.
Dominating was assessed by six items, e.g., “I would argue my case with the
other person to show the merits of my position.”
Integrating was measured by twelve items, e.g., “I would work with the other
person to reach a joint resolution to our conflict.”
Compromising was assessed by five items, e.g., “I would usually propose a
middle ground for breaking deadlocks.”
Avoiding was measured by seventeen items, e.g., “I would try to stay away form
disagreement with the other person.”
Obliging was assessed by six items, e.g., “I would generally try to satisfy the
needs of the other person”.
Emotion was measured by six items, e.g., “I would be emotionally expressive in
the conflict situation’ and ‘I would use my feelings to determine what I should do
in the conflict situation”.
Neglect was assessed by seven items, e.g., “While in the presence of the other
person, I would act as though he/she did not exist”.
Third party help was measured by seven items, e.g., “I would ask a third person
for advice in settling the dispute.”
Items were assessed using a 5-point Likert scale, ranging from “fully disagree” (1) to
“fully agree” (5).
Cronbach’ s alpha of the entire scale was .82. For the eight sub-scales the Cronbach’s
alphas were as follows: .67 for the sub-scale dominating, .84 for the sub-scale
integrating, .73 for the sub-scale compromising, .87 for the sub-scale avoiding, .75 for the
sub-scale obliging, .80 for the sub-scale emotion, .70 for the sub-scale neglect and .88
for the sub-scale third party help.
By excluding one item form the sub-scale dominating (DO3, see Appendix B),
Cronbach’s alpha rose from .67 to .71. The removal of the item did not significantly
change the aforementioned rating of the statements as task or relationship oriented. It
was therefore excluded from further analyses. The ‘dominating’ sub-scale consisted now
of five instead of six items.
Cronbach’s alpha for the entire scale with 65 instead of 66 items did not change.
All hypotheses were tested on a .05 significance level, using linear Hierarchical Multiple
Regression Analysis (HMRA) in the SPSS 10.0 program. This method estimates the
linear relation between several independent or predictor variables and one dependent or
criteria variable (Bortz, 1993). Its hierarchical nature stems from the fact that one is able
to measure the additional information conveyed by a newly introduced set of variables,
through the comparison between the regression analyses of the original and the
augmented set (referred to as Steps 1 and 2 in the following text).
Face-concern were at times independent and dependent variables in the regression
equations.
Culture, gender, age and position were ‘dummy coded’ (see Results chapter) and
included in all analyses as control variables, because of their potential predicting power
on conflict perception, conflict behavior and face-concerns (e.g. Ting-Toomey, 1988;
Rahim, 1986; Shimanoff, 1994). They were entered at Step 1 of the regression equation
and the independent variables – self-construals or face-concerns - at Step 2.
To assess the mediation effect of face, predicted in Hypothesis 6, the three-equation approach, suggested
by Barron & Kenny (1986) was used. According to the authors, the preconditions to test for mediating
effects are: first, the independent variable should
influence the mediator. Second, the independent variable should influence the dependent
variable. Third, when regressing the dependent variable on both the mediator and the
independent variable, the former should influence the dependent variable, whereas the
effect of the latter should be less (partial mediation) or even disappear (full mediation) in
the third compared to the second equation.
In this test, control variables were entered at Step 1, and mediator as well as the
independent variables at Step 2 of the regression equation (Barron & Kenney, 1986).
As the sample included 185 persons for all analyses, the normality assumption according
to the central limit theorem was expected to apply. The central limit theorem states that
the sum of n independent, identically distributed random variables tends to the normal
distribution, as n tends to infinity (e.g. Papoulis, 1991). Thus, all analyses were
performed with parametric tests.
Nevertheless, normality distribution was checked for the residuals of the analyses, using
Kolmogorov-Smirnov Tests. Results revealed that the residuals of the regression, testing
the relationship between self-construals and mutual-face concern, were not normally
distributed. Hence, the results of this analysis should be interpreted with special caution.
The inter-correlation of other-face and mutual-face concerns was, with .6, considerably
high. Thus, an exploratory factor analysis was performed and its results were used for
further analyses, as factor values are uncorrelated.
Factor analysis is a generic term for a family of statistical techniques concerned with the
reduction of a set of observable variables in terms of a smaller number of latent factors.
A distinction can be made between exploratory and confirmatory analyses. The latter is a
method to test hypotheses about the structure of a certain model, whereas the former
aims at discovering common factors on the basis of the empirical data (Brachinger & Ost,
1996).
The ‘Kaiser-Guttman rule’, also known as ‘eigenvalues greater than one’, was chosen to
determine the amount of factors to retain for further analyses. It states that the number of
factors to be extracted should be equal to the number of factors having an eigenvalue
(variance) greater than one (Botz, 1993).
The results of the aforementioned rule suggested that only one factor should be retained.
Yet, there is theoretical ground to assume that face-concern is a three-dimensional
construct (e.g. Oetzel et al., 2000). This structure was therefore imposed. Varimax-
Rotation, an orthogonal rotation method, aiming at maximizing the quadrate loadings per
factor, was used (Bortz, 1993). Results revealed three factors, strongly weighted on self-
face concern, other-face concern and mutual-face concern (Table 2). For reasons of
simplicity and because the new factors closely matched the old variables, they were
labeled ‘self-face concern II’, ‘other-face concern II’ and ‘mutual-face concern II’.
As mentioned in section 2.5 in the Theory chapter, several scholars suggest that, at
least, dominating and emotion are direct, whereas avoiding, obliging, neglect and third
party help are indirect conflict management styles (Ting-Toomey, 1988; Ting-Toomey &
Kurogi, 1998; Ting-Toomey et al., 2000). The integrating and compromising conflict
behaviors were considered to be rather direct, however, no clear suggestions seem to
exist.
To see whether an equivalent structure would be visible in the present data, another
exploratory factor analysis on the bases of the ‘Kaiser-Guttman rule’ was performed.
The principle component analysis revealed three main factors. Factor 1 includes
integrating and compromising; Factor 2 the indirect styles of avoiding, neglect and third
party help; and Factor 3 the direct styles of dominating and emotion. The ‘obliging’ style
was not fully explained by a single factor, but displays a stronger tendency to Factor 2.
After having presented the operationalizations of the concepts and the statistical
methods this study uses, the hypotheses will be reformulated to specify these aspects.
4 RESULTS
Descriptive statistics for, and correlations between the employed variables are reported
in the Tables 4 and 5.
For symmetry reasons, all means and standard deviations are reported, also in cases
where indicators carry little or no information: As the control variables were dummy
coded, the means determine unequivocally the sample distributions, and the standard
deviations do not add information. Furthermore, self-face concern II, other-face concern
II and mutual-face concern II are factor values with mean zero and variance one.
Table 4. Descriptive statistics & non-parametric correlations for
control variables, task- / and relationship-orientation & the definition of integrating and compromising to be task- / or relationship-oriented
Mean S.D. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 Culture .36 .48
2 Gender .41 .49 .03
3 Age2 .48 .50 .09 .17 *
10 Pos4 .08 .27 -.02 -.09 -.12 .05 .15 * -.04 -.03 -.17 * -.14
11 Task – o. .43 .26 -.02 -.07 -.09 .08 -.04 .06 .05 -.11 .07 .10
12 Rel. – o. .57 .26 .02 .07 .09 -.08 .04 -.06 -.05 .11 -.07 -.10 -1.00 **
13 In. - task .50 .29 .00 -.10 -.03 .04 -.03 .06 -.01 -.08 .09 .13 .77 ** -.77 **
14 Co. - task .60 .37 -.07 -.10 .00 .00 -.04 .06 .01 -.15 * .00 .08 .76 ** -.76 ** .67 **
15 In. – rel. .50 .29 .00 .10 .03 -.04 .03 -.06 .01 .08 -.09 -.13 -.77 ** .77 ** -1.00 ** -.67 **
16 Co. – rel. .40 .37 .07 .10 .00 .00 .04 -.06 -.01 .15 * .00 -.08 -.76 ** .76 ** -.67 ** -1.00 ** .67 **
17 Independent 3.88 .41 -.02 .10 .11 -.06 -.13 .10 .01 -.03 -.04 .02 -.13 .13 -.14 -.12 .14 .12
18 Interdep. 3.60 .36 .09 -.07 .08 -.07 -.08 .11 .01 -.08 .07 -.04 -.08 .08 -.15 * -.22 ** .15 * .22 ** .07
* **
significant at p < 0.05 significant at p < 0.05 (p < 0.01) n = 185
Culture: 0 = Individualism, 1 = Collectivism1
Gender: 0 = masculine, 1 = feminine
Age: categories 1-6 were dummy coded (e.g. Age2: 21-30 years = 010000)2
Position: categories 1-4 were dummy coded (e.g. Pos.2: local management = 0100)
1
countries were coded according to Hofstede (1980). East-European countries were coded ‘collectivistic’ due to the regime in place during, at least, the childhood period of the sample’s participants.
2
due to the rule of including n-1 dummy variables in statistical analyses and the fact that the information contained in this data lies mostly in the ‘higher’ categories of the dummy variables, category 1 was excluded from the calculations.
Table 5. Descriptive statistics & non-parametric correlations for
control variables, conflict behavior, face-concerns & self-construals
Mean S.D. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
1 Culture .36 .48
2 Gender .41 .49 .03
3 Age2 .48 .50 .09 .17 *
10 Pos4 .08 .27 -.02 -.09 -.12 .05 .15 * -.04 -.03 -.17 * -.14
11 DO 3.17 .63 -.05 -.06 .07 -.01 -.03 -.04 -.09 -.04 -.04 .07
12 EM 3.25 .70 -.06 .08 .03 .01 -.02 -.03 -.17 * -.14 .03 .03 .13
13 AV 3.91 .49 -.04 -.15 * .04 -.08 -.01 .17 * -.02 -.01 -.05 -.01 .03 -.03
14 OB 3.76 .53 .07 -.07 .01 -.01 .04 .00 -.13 -.05 .01 .00 .00 .03 .24 **
15 NE 2.64 .57 -.03 -.03 -.07 .06 .08 -.06 -.10 .10 -.02 .10 .22 ** .10 .25 ** -.06
16 TP 2.89 .59 .04 -.13 -.04 .08 -.08 .05 .01 .15 * .00 .05 -.05 -.08 .19 ** .17 * .19 *
17 IN 1.88 .51 -.01 .02 .07 .00 -.07 -.09 -.01 -.08 .01 .02 .03 .20 ** -.25 ** .22 ** -.37 ** .11
18 CO 2.52 .72 -.06 -.06 .15 * -.04 -.19 * -.02 .03 -.22 ** -.06 -.03 .03 .09 .14 .30 ** -.22 ** .02 .43 **
19 SF II .00 1.00 -.09 -.09 .04 .04 -.11 .01 -.04 -.01 -.12 .05 .16 *
.14 .06 -.20 **
.17 *
.01 .09 .12
20 OF II .00 1.00 .15 *
-.15 *
-.03 -.02 .02 .01 .13 -.04 .00 .02 -.09 -.05 .22 **
.26 **
-.08 -.03 .19 *
.09 -.03
21 MF II .00 1.00 -.18 *
.01 .10 -.04 -.06 -.01 -.09 -.10 .00 -.16 *
-.03 .23 **
.01 .10 -.21 **
.00 .34 **
.33 **
.02 -.05
22 IND 3.88 .41 -.02 .10 .11 -.06 -.13 .10 .01 -.03 -.04 .02 .27 **
.29 **
-.14 -.70 -.16 *
-.02 .30 **
.11 .13 -.07 -.02
23 INTER 3.60 .36 .09 -.07 .08 -.07 -.09 .11 .01 -.08 .07 -.05 -.12 .21 **
.20 **
.35 **
-.02 .31 **
.33 **
.23 **
.02 .37 **
.17 *
.07
* **
significant at p < 0.05 significant at p<0.05 (p < 0.01) n = 185
DO = dominating, EM = emotion, AV = avoiding, OB = obliging, NE = neglect, TP = third party help, IN = integrating, CO = compromising
SF II = self-face concern II, OF II = other-face concern II, MF II = mutual-face concern
II IND = independent self-construal, INTER = interdependent self-construal
Results 41
Separate tables for each analysis can be found in the Appendix A. In this chapter, only
summary tables will be presented.
Hypothesis 1.1
Hypothesis 1.1 predicted that independent persons tend to show issue- or task-
orientation in a conflict situation. Thus, independent self-construal was expected to be
positively associated with the amount of conflict behavior statements rated task-oriented.
Results of testing this hypothesis are shown in Table 6. The control variables culture,
gender, age and position were entered in Step 1 of the regression equation and
explained 4 % of the variance (F10, 174 = .80, p>0.05). The introduction of independent and
interdependent self in Step 2 explained an additional 3 % of the variance and contributed
significantly to the prediction of task-orientation with F2, 172 = 3.21, p<0.05. However, the
regression model as a whole, explaining 8 % of total variance, failed to reach significance
(F12, 172 = 1.22, p>0.05). Furthermore, the Beta coefficient for independent self was
significantly negative (ß = -.16, p<0.05).
(−) Thus, Hypothesis 1.1 is not supported by the data and must be rejected.
Independent self-construal is not positively but even, in a significant
manner, negatively associated with the amount of conflict behavior
statements rated task-oriented. Therefore, it can not be concluded that
independent persons tend to show issue- or task-orientation in a
conflict situation, as even the reverse seems to apply.
Hypothesis 1.2
Hypothesis 1.2 predicted that interdependent persons tend to show relationship-
orientation in a conflict situation. Thus, interdependent self-construal was expected to be
positively associated with the amount of conflict behavior statements rated relationship-
oriented.
Results of testing this hypothesis are shown in Table 6. The control variables culture,
gender, age and position were entered in Step 1 of the regression equation and
explained 4 % of the variance (F10, 174 = .80, p>0.05). The introduction of independent and
interdependent self in Step 2 explained an additional 3 % of the variance and contributed
significantly to the prediction of relationship-orientation, with a F2, 172 = 3.21, p<0.05.
However, neither the regression model as a whole, explaining 7 % of total variance (F12,
172 = 1.22, p>0.05), nor the Beta coefficient for interdependent self (ß = .10, p>0.05)
reached significance.
(−) Thus, Hypothesis 1.2 is not supported by the data and must be rejected.
Interdependent self-construal is positively associated with the amount
of conflict behavior statements rated relationship-oriented, though not
in a significant manner. Therefore, it can not be concluded that
interdependent persons tend to show relationship-orientation in a
conflict situation.
Furthermore, independent self-construal is in a significant manner
positively associated with the amount of conflict behavior statements
rated relationship-oriented (ß = .16, p<0.05), as shown while testing H
1.1. Thus, it even seems, contrary to the predictions made in this study,
that instead of interdependent, independent persons tend to show
relationship-orientation in a conflict situation.
Table 6. Results of the HMRA for self-construals predicting conflict perception/orientation
Task-orientation Relationship-orientation
ß t ß t
Step 1
Culture -.01 -.08 .01 .08
Gender -.09 -1.16 .09 1.16
Age2 -.14 -.26 .14 .26
Age3 -.05 -.10 .05 .10
Age4 -.13 -.39 .13 .39
Age5 .02 .16 -.02 -.16
Age6 .01 .05 -.01 -.05
Position2 -.08 -.99 .08 .99
Position3 .04 .51 -.04 -.51
Position4 .10 1.28 -.10 -1.28
R2 .03 .03
F 3.21 * 3.21 *
*
significant at p < 0.05 n = 185
Table 7. Results of the HMRA for self-construals predicting the definition of integrating and compromising conflict
behavior integrating-task compromising-task integrating-rel.
compromising-rel.
ß t ß t ß t ß t
Step 1
Culture .01 .12 -.04 -.56 -.01 -.12 .04 .56
Gender -.10 -1.24 -.11 -1.43 .10 1.24 .11 1.43
Age2 .05 .10 -.32 -.63 -.04 -.24 .06 .38
Age3 .08 .16 -.31 -.61 .02 .16 .05 .36
Age4 -.04 -.11 -.26 -.79 .05 .66 .16 1.96
Age5 .04 .24 -.06 -.38 -.10 -1.23 .05 .56
Age6 -.2 -.16 -.05 -.36 -.13 -1.62 -.06 -.77
Position2 -.05 -.66 -.16 -1.95 -.05 -.10 .32 .63
Position3 .10 1.23 -.05 -.56 -.08 -.16 .30 .61
Position4 .13 1.62 .06 .77 .04 .11 .26 .79
R2 (R2 adj.) .04 (-.01) .05 (-.01) .04 (-.01) .05 (-.01)
F .80 .09 .80 .09
Step 2
Independent -.17 -2.30 * -.16 -2.23 * .17 2.30 * .16 2.23 *
R2 (R2 adj.) .10 (.04) .12 (.06) .10 (.04) .12 (.06)
F 1.59 2.04 * 1.59 2.04 *
*
significant at p < 0.05 **
significant at p<0.05 (p < 0.01) n = 185
4.2.3 Self predicting Conflict Behavior
Position3 .02 .19 .04 .52 -.10 -1.25 .00 .03 -.22 -.43 .11 .22 .02 .20 -.14 -1.74
Position4 .05 .61 .02 .22 -.01 -.11 -.03 -.41 -.10 -.31 -.02 -.06 .02 .25 -.05 -.64
R2 (R2 adj.) .03 (-.03) .08 (.03) .07 (.01) .04 (-.02) .05 (-.01) .06 (.01) .03 (-.02) .13 (.08)
F .50 1.61 1.27 .72 .89 1.17 .60 2.59 **
Step 2
Indep. .24 3.19 ** .29 4.16 *** -.16 -2.21 * -.05 -.67 -.12 -1.61 -.03 -.47 .23 3.31 ** .07 .99
Interdep. -.14 -1.91 .22 3.12 ** .19 2.56 * .37 5.24 *** -.04 -.46 .29 3.93 *** .39 5.72 *** .24 3.33 **
R2 (R2 adj.) .10 (.04) .21 (.16) .12 (.06) .17 (.12) .06 (-.01) .14 (.08) .23 (.18) .19 (.13)
F 1.56 3.87 *** 2.03 * 2.99 ** .98 2.35 ** 4.37 *** 3.32 ***
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001) n = 185
Results 54
Hypothesis 4.1
Hypothesis 4.1 predicted that independent persons tend to be concerned with self-face
maintenance in a conflict situation. Thus, independent self-construal was expected to be
positively associated with self-face concern II.
Results of testing this hypothesis are shown in Table 9. The control variables culture,
gender, age and position were entered in Step 1 of the regression equation and
explained 5 % of the variance (F10, 174 = .96, p>0.05). The introduction of independent and
interdependent self in Step 2 explained an additional 2 % of the variance and did not
contribute significantly to the prediction of self-face concern, with a F2, 172 = 1.55,
p>0.05. Furthermore, neither the regression model as a whole, explaining 7 % of total
variance (F12, 172 = 1.06, p>0.05), nor the Beta coefficient for independent self (ß = .12,
p>0.05) reached significance.
(−) Thus, Hypothesis 4.1 is not supported by the data and must be rejected.
Independent self-construal is positively associated with self-face
concern II, though not in a significant manner. Therefore, it can not be
concluded that independent persons tend to be concerned with self-
face maintenance in conflict situations.
R2 (R2 adj.) .03 (-.03) .09 (.03) .07 (.01) .04 (-.02) .05 -.01) .06 (.01) .03 (-.02) .13 (.08)
**
F .50 1.61 1.27 .72 .89 1.17 .60 2.59
Step 2
* * * * *
Self-face II .16 2.10 .16 2.17 .06 .78 -.17 -2.36 .19 2.51 .04 .47 .12 1.77 .10 1.49
** ** **
Other-face II -.05 -.62 .01 .07 .20 2.71 .25 3.33 -.04 -.59 -.08 -1.06 .21 2.87 .02 .29
* * *** ***
Mutual-f. II -.12 -1.58 .17 2.34 .07 .94 .10 1.33 -.16 -2.14 .02 .27 .33 4.58 .27 3.75
R2 (R2 adj.) .07 (-.00) .14 (.07) .11 (.05) .14 (.07) .11 (.04) .07 (.00) .19 (.13) .21 (.15)
* * *** ***
F .98 2.06 1.70 2.06 1.60 1.00 3.07 3.39
R2 .04 .05 .04 .10 .06 .01 .16 .08
* * *** * *** **
F 2.50 3.33 2.99 6.33 3.85 .47 10.97 5.40
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001) n = 185
Results 64
Hypothesis 6.1
Hypothesis 6.1 predicted that independent persons tend to express direct conflict
management styles because of their predominant concern for self-face maintenance.
Thus, it was expected that self-face concern II mediates the relationship between
independent self-construal and direct conflict management styles.
Hypothesis 6.2
Hypothesis 6.2 predicted that interdependent persons tend to express indirect conflict
management styles because of their predominant concern for other- and/or mutual-face
maintenance. Thus, it was expected that at least other- or mutual-face concerns II
mediate the relationship between interdependent self-construal and indirect conflict
management styles.
As presented in the Methods chapter, the preconditions, which must be met for a
mediation effect to be present, are:
1. the independent variable (self) should have a significant relationship with the
mediator (face)
2. the independent variable (self) should have a significant relationship with the
dependent variable (conflict behavior)
3. when regressing the dependent variable (behavior) on both, the mediator (face)
and the independent variable (self), the former (face) should have a significant
relationship with the dependent variable (behavior), whereas the effect of the
latter (self) should be less (partial mediation) or even disappear (full mediation) in
the third compared to the second equation (Barron & Kenny, 1986)
Precondition 1 was assessed by testing Hypotheses 4.1 and 4.2 (a) – (b). The positive
relationship between independent self and self-face concern II did not reach significance.
Interdependent self, on the other hand, was positively associated with other- and mutual-
face concerns II in a significant manner (see Table 9).
Precondition 2 was ensured by testing Hypotheses 3.1 (a) – (d) and 3.2 (a) – (f).
Significant relationships for interdependent self existed with: avoiding, obliging, third
party help, integrating, compromising and emotion (see Table 8).
The potential role of other- and/ or mutual-face concerns II as mediators for the
relationship between interdependent self and the indirect styles of avoiding and obliging
were further tested (i.e. precondition 3). As the indirect style ‘third party help’ was not
associated with neither other- nor mutual-face concern, shown while testing Hypothesis
5.2 (d), it was precluded from the mediation-testing.
Table 11 shows the results of the HMRA testing these hypotheses, which will be
summarized in the following section.
1. Testing face as a mediator for the relationship between interdependent self and
avoiding
The control variables culture, gender, age and position were entered in Step 1 of the
regression equation and explained 7 % of the variance (F 10, 174 = 1.27, p>0.05). The
introduction of interdependent self, other-face II and mutual-face II in Step 2 explained an
additional 5 % of the variance and contributed significantly to the prediction of avoiding,
with a F3, 171 = 3.56, p<0.05. The regression model as a whole, explaining 12 % of total
variance, also reached significance (F13, 171 = 1.84, p<0.05). Furthermore, only the Beta
coefficient for other-face concern II (ß = .16, p<0.05) was significant in this equation.
(√) Thus, interdependent self is no longer a significant predictor for
avoiding, when other-face concern II is also entered into the equation,
which makes other-face concern a full mediator for the relationship
between interdependent self and avoiding conflict behavior.
2. Testing face as a mediator for the relationship between interdependent self and
obliging
The control variables culture, gender, age and position were entered in Step 1 of the
regression equation and explained 4 % of the variance (F 10, 174 = .72, p>0.05). The
introduction of interdependent self, other-face II and mutual-face II in Step 2 explained an
additional 15 % of the variance and contributed significantly to the prediction of obliging,
with a F3, 171 = 10.49, p<0.001. The regression model as a whole, explaining 19 % of
total variance, also reached significance (F13, 171 = 3.06, p<0.001). However, only the
Beta coefficient for interdependent self (ß = .32, p<0.001) was significant in this equation.
(−) Thus, neither other- nor mutual-face concern seem to mediate the
relationship between interdependent self and obliging conflict behavior.
(√) On the other hand, Hypothesis 6.2 was partly supported by the data.
(−) Other-face concern II was mediating the relationship between
interdependent self and avoiding conflict behavior. However, neither
other- nor mutual-face concern mediate the relationship between
interdependent self and obliging conflict behavior. Therefore, it can be
concluded that interdependent persons tend to avoid a conflict, but do
not seem to accommodate to the other person’s wishes, because of
their predominant concern for other- face maintenance.
Table 11. Results of the HMRA predicting conflict behavior; predictors
interdependent self, other- and mutual-face concern II
avoiding t obliging t
ß ß
Step 1
Culture .00 -.03 .06 .73
Gender -.14 -1.80 -.08 -1.05
Age2 -.39 -.76 .01 .01
Age3 -.44 -.89 .00 .00
Age4 -.27 -.84 .05 .14
Age5 .08 .58 .00 .02
Age6 -.09 -.69 -.17 -1.31
Position2 .02 .27 -.06 -.76
Position3 -.10 -1.24 .00 .03
Position4 -.01 -.11 -.03 -.41
R2 .05 .15
*
F 3.56 10.49 ***
Only the demographic variables with theoretical implications on the study were included
in the regression analyses for testing the hypotheses. Yet, because other demographic
variables may still have prediction power on the dependent variables, further analyses
were conducted.
For all additional demographic variables, regressions were calculated to test their effects
on task-/ and relationship-orientation, conflict behaviors and face-concerns.
Only one other demographic variable, ‘having lived abroad,’ had a significant effect on
the dependent variables ‘relationship-orientation’ and ‘definition of integrating and
compromising to be relationship-oriented’.
Therefore, the HMRA, testing the Hypotheses 1.2 & 2.2 (a) – (b), were recalculated.
‘Having lived abroad’ was added as a fifth control variable in Step 1 of the regression
equation.
Testing the additional influence of ‘having lived abroad’ on relationship-orientation & the
definition of integrating and com promising to be relationship-oriented
2 2
‘Having lived abroad’ had a significant effect on relationship-orientation (R = .03 (R adj.
= .02), F = 4.60, p< 0.05). However, when entered as a fifth control variable into the
HMRA predicting relationship-orientation, its Beta coefficient did not reach significance (ß
= .13, p>0.05).
Therefore, it was not needed to re-assess Hypothesis 1.2.
‘Having lived abroad’ also had a significant effect on the rating of compromising to be
2 2
relationship-oriented (R = .02 (R adj. = .02), F = 4.42, p< 0.05). However, when entered
as a fifth control variable into the HMRA predicting compromising to be relationship-
oriented, its Beta coefficient did not reach significance (ß = .15, p>0.05).
Therefore, it was not needed to re-assess Hypothesis 2.2 (b).
Furthermore, ‘having lived abroad’ had a significant effect on the rating of integrating to
2 2
be relationship-oriented (R = .03 (R adj. = .03), F = 6.50, p< 0.05). Here, its addition as
a fifth control variable significantly improved the prediction of defining integrating to be
relationship-oriented. Therefore, the testing of hypothesis 2.2 (a), which predicted that
interdependent persons would define integrating to be relationship-oriented, was fully
repeated.
Results of testing this hypothesis are shown in Table 12. The control variables culture,
gender, age, position and ‘lived abroad’ were entered in Step 1 of the regression
equation and explained 7 % of the variance (F11, 173 = 1.13, p>0.05). The introduction of
independent and interdependent self in Step 2 explained an additional 6 % of the
variance and contributed significantly to the prediction of integrating to be relationship-
oriented, with a F2, 171 = 5.64, p<0.01. The regression model as a whole, explaining 13
% of total variance, also reached significance (F13, 171 = 1.87, p<0.05). Furthermore, the
Beta coefficient for interdependent self (ß = .16, p<0.05) remained significant.
(√) Thus, after additionally controlling for ‘having lived abroad’, Hypothesis
2.2 (a) is still supported by the data.
Step 2
Culture .00 .00
Gender .07 .92
Age2 -.21 -.41
Age3 -.19 -.40
Age4 .00 .01
Age5 -.08 -.54
Age6 -.01 -.05
Position2 .08 1.03
Position3 -09 -1.14
Position4 -.10 -1.27
Lived abroad .17 2.21 *
*
significant at p < 0.05 **
significant at p<0.05 (p < 0.01) n = 185
Discussion 70
5 DISCUSSION
The objective of this study was to investigate cultural aspects of conflict management in
organizations within the perspective of the face-negotiation theory (Ting-Toomey, 1988;
Ting-Toomey & Kurogi, 1998). It was hypothesized that the cultural variables of
independent and interdependent self would have specific relationships with certain
aspects of conflict perception and conflict behavior, as well as with certain face-concerns.
Furthermore, face-concerns were tested for their predicting power and potential
mediation role in the explanation of conflict behavior.
The results support most, but not all of the hypotheses, and some findings require further
investigation.
Contrary to what was expected, the independent self-construal did not show a significant
prediction power for task-orientation. It seems that independent persons tend not to show
task-, but rather relationship-orientation in a conflict situation and in fact more so than
interdependent persons.
These outcomes will be discussed together with the next item.
In line with the predictions, this study’s results show that interdependent persons seem to
define integrating and compromising to be relationship-oriented. On the other hand,
opposite to what was expected, also independent persons appear to do so.
One possible explanation for the fact that some results contradict the theoretical
positions presented in sections 2.3.1 and 2.3.3 could stem from the indirect assessment
of task- and relationship-orientation. No scales were found to measure these perception
tendencies. Therefore, persons were asked to indicate, for each of the questionnaire’s
conflict behavior statements, if they perceived it to
be either task- or relationship-oriented. Several experts regarded the phrasing of the
questionnaire, as a whole, to be ambiguous. Hence, the answers were expected to
reflect the participant’s general perception tendency in a conflict situation (see the
Methods chapter) as well as their general way to define integrating and compromising
styles. However, it could be that the questionnaire’s phrasing was not ambiguous
enough, maybe even indicating a bias towards relationship-orientation in general and for
the styles of integrating and compromising in particular. Thus, it might be possible that
not a participant’s general perception tendency was measured, but rather his or her
rating of whether the questionnaire and the statements indicating integrating and
compromising styles were written in a more task- or relationship-oriented fashion. The
unsettling results of this study could then stem from the fact that the writing of the
questionnaire was quite in line with an interdependent person’s way of thinking and
talking about conflict. As independent persons are expected to be generally more task-
oriented in a conflict situation, the questionnaire could then have been perceived as ‘too
much relationship-oriented’. Thus, for them, the contrast to their general perception
tendency would be bigger, leading them to state that the questionnaire in general, and
the statements indicating integrating and compromising in particular, seem to be
relationship-oriented.
The fact that most of the participants in this study were Europeans could pose a further
explanation. IC is expected, and was shown to exist, within Europe (e.g. Markus &
Kitayama, 1991, Hofstede, 1980, 2001). However, European cultures are relatively
similar, when compared to the more often studied exponents for individualism and
collectivism, the U.S. and Japan, for which the majority of the scales have been
designed.
Thus, a separation between task- and relational-aspects of a conflict situation,
particularly when regarding the styles of integrating and compromising, may not make
the same sense for European persons than for persons with a stronger degree of
individualism or collectivism.
Regarding integrating and compromising conflict styles, however, it could as well be that
those behaviors involve an inherently stronger relationship-orientation. To integrate one’s
viewpoint with the one of another person, or to find a middle ground between both,
demands not only to recognize the other one’s positions, but also to understand his or
her motives and to accept their right to exist. In conclusion, it may well be that, in
general, ‘individualists’ are more task- and ‘collectivists’ more relationship-oriented.
However, when it comes to integrating and compromising, it could be that at least for
‘individualists’, relational aspects start to play a more important role than in their other
modes of conflict resolution.
As expected, this study’s results reveal that independent persons seem to mainly use
dominating, emotion oriented and integrating styles. Interdependent persons, on the
other hand, appear to use predominantly avoiding, obliging, third party help, integrating,
and compromising styles. Thus, in general, independent persons seem to choose direct,
whereas interdependent persons seem to prefer indirect conflict behavior. Integrating
and compromising were assumed to be used by both. For integrating, this seems to be
the case. Compromising, on the other hand, appears to be employed mainly by
interdependent persons.
That independent persons do not indicate compromising could come from the fact that
this style, in contrast to integrating, involves giving up a part of one’s own goals and
positions. This may be more difficult for independent than interdependent persons, as the
latter seem to have a relatively stronger tendency to include the other one’s viewpoints
and wishes into one’s own doing.
Based on the fact that persons who subscribe to individualistic values tend to use direct
or upfront conflict behavior, and on Ting-Toomey et al.’s (2000) finding that those
persons seem to employ emotion oriented styles, this study considered ‘emotion’ to be a
direct conflict approach. It was therefore expected to be associated with an independent
self. Results, however, revealed that not only independent, but also interdependent
persons seem to indicate ‘emotion’.
The definition of this style includes two aspects, which may deserve separate attention.
One refers to the expression of emotions in a conflict situation, and the other to their use
in guiding conflict behavior (Ting-Toomey et al., 2000).
Employing emotionally expressive conflict modes, the aspect of ‘emotion’ that has
received most attention in current research (Ting-Toomey et al., 2000), confronts the
other person with one’s ‘inner states’. In collectivistic cultures this seems to represent a
face-threat, which does not leave much room for the other person to react. As in this
study also interdependent persons indicated ‘emotion’, the following reasons could be
suggested:
It seems highly probable that a qualitative difference exists between Asian and the Latin
American, Southern European collectivism. Emotional expression in a conflict situation
would then be expected to be differently valued in Latin American and Southern
European cultures, and thus be more probable. Compared to ‘individualists’ emotional
expression, however, a further qualitative difference seems likely. The Latin American
and Southern European collectivists are expected to engage in true emotional
expression, whereas the ‘individualists’ would be thought to express emotions primarily
via communicating these.
Within the field of cross-cultural research, no documentation was found concerning ‘the
use’ of emotions to guide conflict behavior. Yet, it seems very likely that all of us are, to a
large extent, influenced by our emotions in our conflict actions. Recalling Thomas’
conflict model (1992), presented in section 2.2 of the Theory chapter, conflict behavior is
a function of cognitions and emotions (which are influenced by structural conditions).
However, how much one allows oneself to openly rely more on one ore the other, may
depend on the relative importance these aspects are given in conflict situations within
different cultures.
From our knowledge about members of individualistic and collectivistic cultures, and
taking into account the potential differences between Asian, Latin American and
Southern European collectivism, we may carefully speculate the following:
Ideally, ‘individualists’, in general strongly valuing rationalism, would be expected to try to
prevent their emotions from guiding their conflict behavior. However, they are seen to
express emotions, at least in the manner of communicating them. This could mean that
an individualist would let the other person know that he or she is hurt, while trying to
negotiate the issue detached from his or her feelings, in order to reach a reasonable
solution for the conflict.
Asian collectivists would be expected to attach great importance to the maintenance of
both conflict parties’ faces, which may lead face to strongly guide their conflict behavior.
This could mean that the person internally deals with the naturally evoked emotions in a
conflict situation, without letting them openly guide his or her behavior.
For Latin American and Southern European collectivists, as for the ‘individualists’,
emotional expression would be predicted. However, less disapproval would be assumed
from the former compared to the latter for the fact of openly letting one’s emotions guide
one’s behavior. Thus, instead of dealing with issue and emotions separately, it may be
perceived that a fair solution to the conflict can be better reached through active
negotiation of the involved feelings.
In conclusion, emotions are expected to be always a strong influencing factor, as they
interact with cognitions to guide conflict behavior. However, it is expected that ‘cultures’
differ in the standards of emotional expression as well as in the relative importance they
give emotions in conflict situations.
This study was conducted within an organizational setting, which by itself may be
perceived to demand an ‘appropriate behavior’ to a higher degree than, for example, any
intimate relationship. Furthermore, the assessment of neglect includes statements like: “ I
would say nasty things about the other person to other people.” Hence, a social
desirability bias seems a more plausible explanation for this study’s results concerning
neglect.
In general, however, the social desirability distortion is expected to be low in this data, as
participants filled the study’s questionnaire anonymously and on-line. In fact, an
experimental study by Joinson (1999) comparing Internet and paper and pencil
questionnaires in combination with anonymous and non-anonymous submission has
shown that social desirability answers were the least frequent in anonymous Internet
questionnaires. Furthermore, the participation was entirely voluntary and in no way
connected to the participant’s employer.
In any case, for the present study it was more important that no systematical cultural
differences in potential social desirability responses existed. The two styles mostly
susceptible of being influenced by this response bias, when compared to their actual use
in every day’s life, seem to be neglect, which participants may ‘under-indicate’, and
integrating, which participants may ‘over-indicate’. If ‘under- or over-indication’ still
existed within this study, neglect seems to have been ‘under-indicated’ and integrating
possibly ‘over-indicated’ by both independent and interdependent persons to the same
extent.
Moreover, it could be that cultures differ in what is considered socially desirable.
However, considerable variability in response patterns also existed within cultures, which
makes a systematic data distortion of this type little expectable.
Oetzel et al.’s (2000) study involved ‘the exponents’ of individualism (U.S.) and
collectivism (Japan).
The fact that the sample of the present study included ‘mild forms’ of each may account
for independent persons not showing the expected self-face maintenance. However,
‘mildly collectivistic’ as they may be, the answers of the interdependent persons were in
line with the expected response pattern. It therefore seems more plausible that these
outcomes are due to qualitative differences in the definition of face as well as in the
relative importance attached to it in individualistic compared to collectivistic cultures.
As presented in section 2.4.3 of the Theory chapter, in collectivistic cultures, especially
with Asian background, one’s self is maintained and codified through active facework,
which follows the principle of reciprocity (Ting-Toomey, 1988). In individualistic cultures,
on the other hand, facework, which does not follow explicit social norms, and is therefore
expected to be rather self-directed, seems less crucial for a person’s self-concept.
Additionally, the loss of personal face in collectivistic cultures does not only affect the
individual, as in individualistic cultures, but also his or her position (e.g. professorship)
(Kim, 2000). Thus, face seems to be differently defined and embedded in individualistic
and collectivistic cultures.
From the above reasoning it could be assumed that face maintenance is of lesser
importance for independent than for interdependent persons. Furthermore, for
independent persons self-face maintenance would be of a stronger concern, whereas for
interdependent persons, other- and mutual-face concerns would be in the foreground.
These assumptions seem to be reflected in the results of the present study. Independent
self is positively associated with self-face concern, though weakly and negatively with
other- and mutual-face concerns. Interdependent self, on the other hand, is positively
associated with all three, though significantly only with other- and mutual-face concerns.
In line with its expectations, this study’s results indicated that persons who are concerned
with is self-face maintenance seem to use the direct styles of dominating and emotion.
Persons who are concerned with is other-face maintenance, on the other hand, appear to
express the indirect styles of avoiding and obliging.
Results of the research questions revealed that persons who are mainly concerned with
mutual-face maintenance seem to show integrating and compromising styles. This is in
line with the suggestion by Ting-Toomey et al. (1991) that at least to account for
integrating, a mutual-face dimension would be required.
Contrary to this study’s predictions, ‘neglect’ was not indicated by persons who are
concerned with other- and/or mutual-face maintenance. However, this was the case for
persons who are concerned with self-face maintenance.
Additionally, persons who are concerned with other- and/or mutual face maintenance
seem not to ask significantly more for a third party’s help in solving a conflict than
person’s who are concerned with self-face maintenance. This means that participants in
this study appear not to consider the request for help as a face-maintenance move.
Although the interdependent persons in this sample often showed the predicted
collectivistic behavior patterns, a possible explanation for the above mentioned outcome
could stem once more from potential qualitative differences between Asian and Latin
American, or Southern European collectivism. In cultures with a strong normative
orientation regarding face, calling in a third party appears to take away the responsibility
of face-maintenance from the conflict parties, therefore saving both sides’ faces. This
seems to be the case in Asian collectivistic cultures, where third party intervention is
more institutionalized. Without such an ‘institutionalization’, however, calling in a third
party could also be perceived as exposing one’s problems to yet another party, towards
whom one may also lose face.
Furthermore, in this study not only persons who are concerned with self-face
maintenance, but also persons who are concerned with mutual-face maintenance
indicate ‘emotion’.
As discussed in part 5.1.3, this style seems to include emotional expression in a conflict
situation, as well as their use to define conflict behavior. Looking at emotion from its
expressive aspect, it could be considered as a direct style and expected to be indicated
mainly by independent persons. For ‘individualists’, expressing emotions, at least in the
way of openly talking about one’s feelings, could mean showing that one stays true to
oneself, thus maintaining one’s self-face. For Latin American and Southern European
‘collectivists’, as mentioned earlier, expressing one’s emotions in a conflict may be
perceived as part of a constructive way to argue, which would correspond to a concern
for the preservation of both parties’ faces.
It was expected that independent persons tend to express direct conflict management
styles, because of their predominant concern for self-face maintenance. Conversely, it
was thought that interdependent persons tend to express indirect conflict management
styles, because of their predominant concern for other- and/or mutual-face maintenance.
The precondition for mediation were not met in the case of independent persons and
self-face concerns (see part IV). Hence, it seems that their direct conflict behavior is not
well explained through a concern for self-face maintenance. The mediation role of other-
and/or mutual-face concerns was tested for the connection between interdependent self-
concept and the indirect styles of avoiding and obliging. The results indicated that
interdependent persons seem to accept avoiding a conflict, because of their concern for
other-face maintenance. However, they do not appear to be obliging, as a function of
their other- and/or mutual-face concerns.
It could be proposed that all conflict management styles involve both a concern for the
issue in stake, as well as some kind of face negotiation. In particular, it could be
suggested that face is more closely connected with the relational dimension, whereas
concerns are more closely related to the issue dimension of conflict negotiation.
Focusing on interaction, as social conflicts do, avoiding is per definition disconnected
form a negotiation of the conflict issue, as no outward attempts are initiated to come to
terms with a problem. If the precondition is met that the problematic issue is worked out
inwardly, the relational dimension, however, is fully preserved. Therefore, if actively used,
avoiding can only be explained by a weak concern for the issue and a strong concern for
the relation of the parties involved, hence by a concern for face.
As the conflict management styles move from one extreme, i.e. avoiding, to the other, i.e.
dominating, the weight balance between face and concern shifts accordingly. Obliging
implies that the problem was already put on the table, which brings a concomitant
increase in the concern for the issue. In proportion, this would lead to a decrease in the
explanatory power of face and therefore in it’s potential mediation role. This could explain
why face-concerns only mediate avoiding, even though they are present in all conflict
interactions.
Another possible explanation may be again rooted in the fact that this study’s European
sample involves a mild form of collectivism. It is possible that face concerns are generally
needed to explain ‘collectivists’ conflict behavior, but that this sample did not have the
required characteristics to fully reveal their important role. The one exception of other-
face concern mediating an interdependent persons’ use of the avoiding style is expected
to be so fundamental that it would even be visible within weaker system parameters.
Further reasoning is needed for the fact that only other-face, and not mutual-face or both
face-concerns were seen to mediate the avoiding conflict behavior. In this sample the
indirect styles are associated with other-face concern, whereas integrating and
compromising are stronger related to mutual-face concern. Thus, it seems that indirect
conflict behavior is perceived to involve generally more concern for the other than for
both-sides.
Another explanation may stem once more from the possibility that different types of
collectivism exist. Asian collectivistic cultures, from which our expectations were drawn,
assume a strong normative component for face. To avoid a conflict or to be obliging to
another person’s wishes, would save both parties’ faces and would hence be not only
other-, but also or even more mutual-face concerned. In Latin American and Southern
European collectivistic cultures, face seems to weight stronger on it’s relational than on
its normative aspect (e.g. Garcia, 1996 in Ting-Toomey et al., 2000). Thus, it could be
that here, indirect conflict behavior is more other-face directed.
To provide an adequate interpretation frame for the present study’s results, several
methodological issues should still be addressed in addition to the ones mentioned
throughout the discussion of the hypotheses. Implications will be mostly taken up in the
section about suggestions for further research.
One issue concerns the fact that independent and dependent variables were assessed
using the same method. A problem with such approaches is that potential subjective
distortions would affect all measures included. Yet, as mentioned, at least in general,
there seem to be grounds to assume little systematic response biases. Another problem
is this study’s cross-sectional design. It means that all data was collected at once and
does therefore neither allow to draw conclusions about the temporal evolution of the
measured aspects nor about causality. As mentioned in the Theory chapter, a person’s
cultural background influences his or her development of self-concept from early
childhood on. Furthermore, face is the projection of this cultural self-concept within
interaction. This seems to make a causal influence from independent and interdependent
self-construal as well as from self-, other- and mutual-face concerns on conflict
perception and behavior, more plausible than the other way around.
Additionally, it should be noted that the majority of participants in this project had a high
level of education, which makes them a fairly select group. However, the present study
did not aim at generalizations from these respondents to other members of the same
culture. Instead it attempted at deriving statements about today’s professional workforce
with a potential, or actual carrier in globally operating companies, for which this sample
could be considered quite ideal.
Furthermore, this research was solely conducted in English. This may produce a
distortion of the data in that not everyone attaches the same meaning to the
questionnaire statements. Nevertheless, sixty-three percent of the participants indicated
that their understanding of English was very good, which allows to expect a potential
distortion to be small.
Finally, it should still be pointed out that this study was drawing upon self-reports for data
collection. This method does not measure actual conflict behavior, but a person’s general
attitude towards such conduct. Thus, even though systematic response biases are
assumed minimal, statements about conflict behavior can only have the form of
expectations. As mentioned in the Introduction, if the person really engages in the conflict
behavior, he or she indicated is nonetheless subject to further influencing factors.
Henceforth, when talking about ‘independent or interdependent persons seem to use …’,
the presented frame is assumed.
5.1.8. Brief Summary So Far
Despite the indirect assessment of task- and relationship-orientation, the use of a self-
reporting strategy, together with a cross-sectional design and the inclusion of the vaguely
defined ‘emotion’, ‘neglect’ and ‘mutual-face’ concern, this study produced some very
interesting results:
An exploratory factor analysis of the five ‘classical’ (e.g. Rahim, 1983) plus three
additional conflict management styles (Ting-Toomey et al., 2000) revealed three factors,
which seem to describe direct, indirect, and integrating plus compromising styles,
respectively.
Direct styles were associated with self-face concern, indirect styles with other-face
concern and the integrating and compromising styles mainly with mutual-face concern.
It seems that independent persons tend to use direct or upfront conflict management
styles, as well as integrating. Interdependent persons, on the other hand, seem to use
indirect or smoothing conflict management styles, as well as integrating and
compromising.
In a conflict situation, independent persons seem not to be too concerned with face
maintenance in general, whereas interdependent persons seem to care mainly for other-
and mutual-face maintenance.
Finally, interdependent persons appear to show some degree of conflict avoidance, due
to their tendency for other-face maintenance.
In the following example, for the five classical conflict management styles, bold
fonts correspond to a stronger weighting factor for the particular concern:
Compromising: high in parts of both & high in mutual-face concern parties’ concerns
Although this study has revealed very interesting results regarding cultural aspects of
conflict management in organizational environments, it is clear that limitations in time and
resources preclude the very thorough assessment of all aspects involved. In this section
several points for further research shall be discussed, which may give better answers to
some of those limitations.
One of this study’s main goals was to measure cultural aspects of conflict perception.
This objective was not fully reached, due to the lack of existing suitable assessment
tools. Since there are strong theoretical grounds to assume cultural differences, and
since the perception of a social situation plays a crucial role in shaping a person’s
behavior, the investment in the development of a direct assessment tool seems fully
justified.
As aforementioned, the aspect of conflict perception studied here differentiated between
task- and relationship orientation. These rely on an underlying dimension, assumed
fundamental in the understanding of cultural aspects of conflict management, which
refers to the normative standards of how to read social situations and to decide which
behaviors to consider appropriate.
Cultures seem to significantly differ in the weight they put on normative rules to guide
their social interactions. Societies with relatively few normative rules follow E.T Hall’s
(1976, 1983, in Ting-Toomey, 1988) definition of ‘low-context culture’ (LCC). Most
information about how to understand a situation and which behavior to consider
appropriate is rather conveyed through direct communication than through the context
itself. Many individualistic cultures are assumed to be LCC (e.g. Ting-Toomey, 1988). On
the other end of the spectrum lay cultures with strong behavior ‘regulations’. There, the
context, rather than the communication provides the needed information for adjusting
one’s behavior. This concept is referred to as ‘high context culture’ (HCC; Hall, 1976,
1983, in Ting-Toomey, 1988), which many collectivistic cultures are assumed to belong
to (e.g. Ting-Toomey, 1988). As mentioned earlier, these characteristics are expected to
be represented in a person’s self-concept. However, no explicit clarifications exist
concerning this issue, which would be needed for a more comprehensive determination
of cultural influences on experiences and behavior via the self-concept.
As presented in section 2.1.2 of the Theory chapter, self is a better predictor for a
person’s behavior than culture itself. Accordingly, it was assumed to be a better
approach to the investigation of individual behavior tendencies within different cultures.
On the other hand, in line with several researchers (e.g. Gudykunst et al., 1996; Oetzel et
al., 2000), it must be noted that self is not the only determinant for an individual’s
behavior. A person may understand him or herself as relatively independent in a
collectivistic culture and still not behave in the same way as he or she would in an
individualistic culture. Cultures, through means of social sanctions, will, for example,
support some and hold back other behaviors, depending on their norms. To clarify this
aspect, further research is necessary, which directly assesses the two ways that culture
seems to exert its influence through: a person’s cultural self-concept and the cultural
value dimensions.
Another issue for further investigation concerns the definitions of integrating and
compromising. A qualitative approach to a person’s understanding of these styles could
be promising. Furthermore, it would be interesting to test whether these definitions differ
significantly, when applied to ‘stronger or weaker’ forms of individualism and collectivism.
Regarding the cultural influences on conflict behavior, the enlarged group of eight conflict
management styles may benefit from further attention.
In particular, an exploration of the role of emotions in conflict management seems to be
of fundamental importance. Intercultural encounters very often result in expectations not
being met. Naturally these disappointments go along with feelings like irritation,
frustration, and even anger. The way these emotions are internally and externally dealt
with is a major determinant of the process and outcome of conflict interactions.
Considering the time and resource limitations when writing a thesis, the method of
employing a questionnaire for the study of cross-cultural issues seems the most
reasonable and promising. However, a more thorough research should include direct
observations of actual conflict behaviors as well as qualitative approaches to the
meanings of conflict, conflict perception and conflict behavior in different cultures.
Additionally, longitudinal designs and priming experiments, for example, would be helpful
in determining true causal relationships.
This study took a culture comparative approach, in which general cultural tendencies
were the focus of attention. Several scholars suggested the possibility that the aspect of
one’s self-concept (independent or interdependent) that is activated, as well as one’s
conflict behavior, may be in addition to culture also situation dependent (e.g. Gudykunst
et al., 1996 and Oetzel, 1999, respectively). A possible next step could be to compare
the present outcomes with a person’s understanding of self and conflict behavior in other
contexts.
Furthermore, it was assumed that a rather dominating style towards subordinates and a
rather obliging style towards superiors would be culture dependent only in quantity. Thus,
the strongest cultural variability was assumed to exist in conflict behavior among peers.
However, this would deserve to be tested.
Ethnical differences within one culture would be as well an important issue to address in
further research, as considerable variations seem to exist (e.g. Ting-Toomey et al.,
2000).
One of the most important contributions of the present study is the suggestion to combine
Rahim’s ‘concerns’ approach with Ting-Toomey’s face-negotiation theory. The perceived
need for such a hybrid model is founded on the assumption that concerns as well as face
concerns are present in any conflict situation, and on the observations in this study that
they seem to be differently weighted depending on culture. It would now be needed to
verify if the theoretical considerations, which were raised as justification for
this model, stand an empirical testing.
One of the main issues to address within this line of thought should be the investigation
of how face is defined and weighted in individualistic compared to collectivistic cultures.
Moreover, it would be needed to investigate if the ‘concerns’ and the ‘face-concerns’ truly
represent more issue or relational aspects, respectively.
Furthermore, it has been discussed in the present study that face may have more than
one type of influence over a conflict situation. In particular, it seems clear that it
comprises a relational as well as a normative component. A better knowledge of the
weight given by different cultures to each of these components, as well as its variation
over consecutive stages of the conflict, may lead to a deeper understanding of the role of
face in conflict handling.
It is a common knowledge that systems work reasonably well, as long as they remain
isolated. Furthermore, we all know that though ‘black’ and ‘white’ are useful
simplifications of complex topics, the world in fact is a shade of grays.
Since the reality is ‘multi-systemized’ nowadays, new and more complex topics have to
be addressed. And since this is much easier through ‘black’ and ‘white’ this study did and
will assume a talking about ‘individualists’ and ‘collectivists’. However, …
Throughout this study, it became clear that members of different cultures view their world
through different lenses, attach different meanings to particular events, and have
different standards on how to behave in given situations. This makes intercultural
encounters very susceptible to misunderstandings.
Stress, time pressure, uncertainties, and specific expectations are characteristics of
every day’s business life. They make intercultural encounters very susceptible to
habituated behaviors, from which cultural misunderstandings are nearly always
guaranteed.
Thus, intercultural encounters often result in expectations not being met. This frequently
ends up in irritation and frustration, but also in time delays and decreases in productivity.
However, since intercultural encounters are nowadays more common than rare, we will
have to work with them, preferably in a constructive manner.
To do so, at least three fundamental issues are proposed: first of all, knowledge is
needed about ‘how’ culture influences conflict management, as well as ‘why’ members of
different cultures behave in the way they do. Second, training is important, to offer
experiences with culture in general and cultural differences in particular. And third,
tangible applications are required to provide an infrastructure for intercultural encounters
at the workplace.
Concerning the ‘how’, this study has shown that ‘individualists’ tend to use direct,
whereas ‘collectivists’ tend to use indirect conflict behavior.
Regarding the ‘why’, results have indicated that for ‘individualists’, ‘attacking’ the issue
appears to be a major concern. This could be the reason for their direct conflict
approach.
For ‘collectivists’, on the other hand, saving or at least not harming another person’s face
seems fundamental. This appears to explain their tendency for indirectness.
The presented knowledge is essential for working with the common prejudices that
‘individualists’ use a ‘brute force’ approach, whereas ‘collectivists’ do not ‘come to the
point’. It helps us to understand, why these tendencies seem to exist, a fact that is
elementary for conflict prevention and resolution.
Persons who deal with intercultural encounters in their working place should not only be
sensitized, but also trained to deal with them. Knowledge is essential. However, if
provided with no other support, knowledge holds the danger of producing or reinforcing
stereotypes. Thus, a general attitude towards dealing with cultural differences, as well as
the development of interaction skills, needs to be trained and embedded into an
international company’s structure.
A good approach for the required attitude is given by what was called ‘mindfulness’
(Thich, 1991 in Ting-Toomey & Kurogi, 1998). It means “attending to one’s own internal
assumptions, cognitions and emotions and simultaneously attuning attentively to the
other’s assumptions, cognitions and emotions while focusing the five senses” (p. 203).
The interaction skills to be trained include an appropriate, effective, and adaptive
communication in a given situation (Ting-Toomey & Kurogi, 1998).
All this sounds rather common knowledge. However, for many reasons this does not
happen by itself, as both changes in attitude and acquisition of skills require great mental
flexibility and considerable effort. And, to bring the subject ‘down to working earth’ again,
it also requires at least one additional aspect: time.
6 REFERENCES
th
Bortz, J. (1993). Statistik für Sozialwissenschaf tler 4 ed. Berlin, Heidelberg: Springer
- Verlag.
Chinese Culture Connection (1987). Chinese values and the search for culture-free
dimensions of culture. Journal of Cross-cultural Psychology, 18, 143-164.
Deutsch, M. (1976). Konfliktregelung. München: Reinhardt.
Erez, M. &. Earley P. C (1993). Culture, self-identity and work. New York: Oxford
University Press.
Fisher, R., Ury, W. & Patton, B. (1995). Das Harvard-Konzept. Sachgerecht verhandeln -
erfolgreich verhandeln. Frankfurt: Campus Verlag.
Galtung, J. (1981). Structure, culture, and intellectual style: an essay comparing Saxonic,
Teutonic, Gallic and Nipponic approaches. Social Science Inform ation, 20, 817-856.
Gladwin, T. N. & Walter, I. (1980). Multinationals under f ire. New York: John Wiley &
Sons.
Goffman, E. (1967). Interaction ritual. Essays on face-to-face behavior. Garden City, New
York: Doubleday.
Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K. & Heyman, S.
(1996). The influence of cultural individualism-collectivism, self-construals, and individual
values on communication styles across cultures. Hum an Communication Research, 22,
510-543.
Jehn, K. A. & Mannix E. A. (2001). The dynamic nature of conflict: a longitudinal study of
intragroup conflict and group performance. Academ y of Managem ent Journal, 44, 238-
251.
Kempf, W. (2001). Peace building and social cognition. Paper presented at the 7th.
international sym posium on the contributions of psychology to peace. Ateneo de Manila,
July 16-22, 2001.
Kempf, W. (2002). The social construction of international conflict. Colloquium on
constructive conflict coverage. A social psychological approach. Psychology Departm ent,
York University, Toronto (Ontario), February 7, 2002.
Kim, M. S., Sharkey, W. F. & Singelis, T. M. (1994). The relationship between individuals'
self-construals and perceived importance of interactive constraints. International Journal
of Intercultural Relations, 18, 117-140.
Markus, H. R. & Kitayama, S. (1991). Culture and the self: Implications for cognition,
emotion and motivation. Psychological Review, 98, 224-253.
Morris, M. W., Williams, K. Y., Leung, K., Larrick, R., Mendoza, T., Bhatnagar, D., Li, J.,
Kondo, M., Luo, J-L., Hu, J-C. (1998). Conflict management style: accounting for cross-
national differences. Journal of International Business Studies, 29, 729-748.
Oetzel, J. G. (1999). The influence of situational features on perceived conflict styles and
self-construals in work groups. International Journal of Intercultural Relations, 23, 679-
695.
Oetzel, J. G. &. Bolton-Oetzel, K. (1997). Exploring the relationship between self-
construal and dimensions of group effectiveness. Management Com m unication
Quarterly, 10, 289-315.
Oetzel, J., Ting-Toomey, S., Matsumoto, T., Yokochi, Y., Pan, X., Takai, J., Wilcox, R.
(2000). Face and facework in conflict: a cross-cultural comparison of China, Germany,
Japan and the United States. Paper presented to the International Com munication
Association, International and Intercultural Division. Acapulco, Mexico, May 2000.
Rothbaum, F. M. & Tsang B. Y. P. (1998). Love songs in the United States and China: on
the nature of romantic love. Journal of Cross-cultural Psychology, 29, 306-319.
Singelis, T. M. & Brown W. J. (1995). Culture, self and collectivist communication: linking
culture to individual behavior. Hum an Communication Research, 21, 354-389.
Smith, P. B., Dugan, S. & Trompenaars, F. (1996). National culture and the values of
organizational employees. A dimensional analysis across 43 nations. Journal of Cross-
cultural Psychology, 27, 231-264.
Ting-Toomey, S., Gao, G., Trubinsky, P., Yang, Z., Kim, H. S., Lin, S-L. & Nishida, T.
(1991). Culture, face maintenance, and styles of handling interpersonal conflict: a study
in five cultures. International Journal of Conf lict Managem ent, 2, 275-296.
Ting-Toomey, S., Yee-Jung, K., Shapiro, R. B., Garcia, W., Wright, T. J. & Oetzel, J. G.
(2000). Ethnic/cultural identity salience and conflict styles in four U.S. ethnic groups.
International Journal of Intercultural Relations, 24, 47-81.
Tjosvold, D. & Tjosvold M. M. (1995). Psychology for leaders. Using m otivation, conflict
and power to manage more effectively. New York: John Wiley & Sons.
Triandis, H. C. (1989). The self and social behavior in differing cultural contexts.
Psychological Review, 96, 506-517.
Trubinsky, P., Ting-Toomey, S. & Lin, S-L. (1991). The influence of individualism-
collectivism and self-monitoring on conflict styles. International Journal of Intercultural
Relations, 15, 65-84.
Weisz, J. R., Rothbaum, F. M. & Blackburn, T. C. (1984). Standing out and standing in:
the psychology of control in America and Japan. Am erican Psychologist, 39, 955-969.
Wu, W., Yuen, E. & Zhu, J. J. (2001). Individualism-collectivism and conflict resolution
styles: a cross-cultural study of managers in Singapore.
http://www.fba.nus.edu.sg/fba/m scphd/0106.pdf.
Appendix 97
7 APPENDIX
Questionnaire
7 APPENDIX A (Tables for all Regression Analyses separately)
Hypothesis 1.1
Table 13. Regression analysis for self-construals predicting task-orientation
ß t R2 F R2 F
Step 1 .04 .80 .04 .80
Culture -.01 -.08
Gender -.09 -1.16
Age2 -.14 -.26
Age3 -.05 -.10
Age4 -.13 -.39
Age5 .02 .16
Age6 .01 .05
Position2 -.08 -.99
Position3 .04 .51
Position4 .10 1.28
Step 2 .08 1.22 .03 3.21 *
Hypothesis 1.2
Table 14. Regression analysis for self-construals predicting relationship-orientation.
ß t R2 F R2 F
Step 1 .04 .80 .04 .80
Culture .01 .08
Gender .09 1.16
Age2 .14 .26
Age3 .05 .10
Age4 .13 .39
Age5 -.02 -.16
Age6 -.01 -.05
Position2 .08 .99
Position3 -.04 -.51
Position4 -.10 -1.28
Appendix 98
Table 13 continued
Step 2 .08 1.22 .03 3.21 *
*
significant at p < 0.05 **
significant at p<0.05 (p < 0.01)
n = 185
Hypothesis 2.1 (b)
Table 16. Regression analysis for self-construals predicting compromising to be task-orientated
ß t R2 F R2 F
Step 1 .05 .91 .05 .91
Culture -.04 -.56
Gender -.11 -1.43
Age2 -.32 -.63
Age3 -.31 -.61
Age4 -.26 -.79
Age5 -.06 -.38
Age6 -.05 -.36
Position2 -.16 -1.95
Position3 -.05 -.56
Position4 .06 .77
Step 2 .12 2.04 .07 7.35 **
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001)
n = 185
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001)
n = 185
Hypotheses 3.1 (d) & 3.2 (f)
Table 22. Regression analysis for self-construals predicting compromising
ß t R2 F R2 F
Step 1 .13 2.59 ** .13 2.59 **
*
significant at p < 0.05 **
significant at p<0.05 (p < 0.01) ***
significant at p<0.05 (p < 0.001)
n = 185
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001)
n = 185
Hypothesis 3.2 (c)
Table 25. Regression analysis for self-construals predicting neglect
ß t R2 F R2 F
Step 1 .05 .89 .05 .89
Culture -.06 -.83
Gender -.01 -.09
Age2 -.13 -.85
Age3 -.15 -1.20
Age4 .14 1.71
Age5 .05 .54
Age6 .09 1.18
Position2 -.26 -.50
Position3 -.22 -.43
Position4 -.10 -.31
Step 2 .06 .98 .01 1.44
Culture -.06 -.83
Gender .00 .04
Age2 -.10 -.70
Age3 -.15 -1.13
Age4 .13 1.60
Age5 .04 .49
Age6 .10 1.24
Position2 -.23 -.44
Position3 -.20 -.40
Position4 -.10 -.31
Independent -.12 -1.61
Interdependent -.04 -.46
n = 185
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001)
n = 185
Hypothesis 4.1
Table 27. Regression for analysis self-construals predicting self-face concern II
ß t R2 F R2 F
Step 1 .05 .96 .05 .96
Culture -.07 .-92
Gender -.15 -1.91
Age2 .21 .40
Age3 .18 .37
Age4 -.01 -.03
Age5 .04 .25
Age6 .02 .14
Position2 -.02 -.14
Position3 -.09 -1.04
Position4 .04 .50
Step 2 .07 1.06 .02 1.55
Culture -.07 -.93
Gender -.16 -2.00 *
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001)
n = 185
*
significant at p < 0.05
n = 185
Research question 1
Table 36. Regression analysis for face-concerns II predicting integrating
ß t R2 F R2 F
Step 1 .03 .60 .03 .60
Culture -.03 -.36
Gender .01 .16
Age2 .24 .46
Age3 .22 .44
Age4 .11 .33
Age5 -.09 -.63
Age6 .01 .11
Position2 -.08 -1.01
Position3 .02 .20
Position4 .02 .25
Step 2 .19 3.07 *** .16 10.97 ***
*
significant at p < 0.05 **
significant at p<0.05 (p < 0.01) ***
significant at p<0.05 (p < 0.001)
n = 185
Research question 2
Table 37. Regression analysis for face-concerns II predicting compromising
ß t R2 F R2 F
Step 1 .13 2.59 ** .13 2.59 **
* significant at p < 0.05 ** significant at p<0.05 (p < 0.01) *** significant at p<0.05 (p < 0.001)
n = 185
Hypothesis 6 (1)
Table 38. Regression analysis predicting avoiding, predictors interdependent self & other- / mutual-face concerns II
ß t R2 F R2 F
Step 1 .07 1.27 .07 1.27
Culture .00 -.03
Gender -.14 -1.80
Age2 -.39 -.76
Age3 -.44 -.89
Age4 -.27 -.84
Age5 .08 .58
Age6 -.09 -.69
Position2 .02 .27
Position3 -.10 -1.24
Position4 -.01 -.11
Table 38 continued
Step 2 .12 1.84 * .05 3.56 *
Hypothesis 6 (2)
Table 39. Regression analysis predicting obliging, predictors interdependent self & other- / mutual-face concerns II
ß t R2 F R2 F
Step 1 .04 .72 .04 .72
Culture .06 .73
Gender -.08 -1.05
Age2 .01 .01
Age3 .00 .00
Age4 .05 .14
Age5 .00 .02
Age6 -.17 -1.31
Position2 -.06 -.76
Position3 .00 .03
Position4 -.03 -.41
Step 2 .19 3.06 *** .15 10.49 ***
*
significant at p < 0.05 **
significant at p<0.05 (p < 0.01) ***
significant at p<0.05 (p < 0.001)
n = 185
7 Appendix B
University of Konstanz
Department of Psychology
Dear Madam/Sir,
Thank you very much for the interest you have shown in this study on “cultural aspects of conflict management in
organizations”. Its aim is to investigate how people in different cultures deal with conflicts within a work
environment.
Conflicts are inevitable in companies' everyday life. Their impact on the well-being and productivity of the
employees, however, depends on the way they are managed. It is important to note that different cultures differ in
how they deal with the same conflict.
In these times of global strategical alliances, mergers & acquisitions and an increasing exchange of workforces across
borders, the awareness of cultural conflict specificity could play a crucial role in managing intercultural differences
more effectively. At the end of the project, you will be informed of the general outcome.
This investigation is an essential part of the dissertation I am presently writing. It addresses the issue of 'workplaces in
a global environment', and is developed in close collaboration with the groups of 'conflict research' and 'work and
organizational psychology' of the university of Konstanz, Germany.
In order to get a representative sample, I would be very grateful if you could get a sizeable number of co-workers
and staff to join in answering to the questionnaire.
This questionnaire is given to several companies in different contries. Because of language equivalence, it is written in
English. Therefore everybody, who has knowledge of English is welcome to take part.
The questionnaire can be found on the Internet at: http://www.cis.hut.fi/rvigario/Questionnaire (if the
automatic connection does not work, please ‘copy paste’ the address to the address field in your web
browser).
Completing it will take approximately 20 minutes and pressing the 'send' button will deliver it directly to me. I would
ask you to please send answers until the 11th of January 20023.
I assure you that all answers will be handled anonymously and confidential. For that purpose, no names (neither of
the company nor of the individual) will be included. Yet, in order to be able to perform my analysis, a code will be set
up in the beginning of the questionnaire.
Furthermore, you will be able to check how your answers have been stored, as well as all the programs behind
the questionnaire (all the relevant files will be located in the same page as the questionnaire's main page).
If you have any questions or comments, don't hesitate to contact me (see contact information at the bottom).
Dorothea Hamdorf
Mail address: Steinstr. 21 Email: Dorothea.Hamdorf@uni-konstanz.de
D - 78457 Konstanz // Germany Phone: +49 - (0)7531 917425
3 for convenience this date was moved to the end of February.
University of Konstanz
Department of Psychology
Methods and Conflict Research
Prof. Dr. Wilhelm Kempf
Dear Participant,
Thank you very much for taking part in this investigation about cultural influences on conflict behavior
in organizations. Please take a little time to read the instructions carefully.
On the following pages you will be asked to what degree you agree or disagree with a statement. No judgement
can and will be made on your answer. Answering from your personal experience will be a valuable contribution to
increase the knowledge we need for a better intercultural understanding. I would therefore ask you to answer
spontaneously and honestly. Sometimes no statement will fully match you, but please select what characterizes you
the most, as answering all questions is essential.
I assure you that all answers will be handled anonymously, and kept confidential. For that purpose, no names will
be included. Yet, in order to be able to perform my analysis, a code will be set in the beginning of the questionnaire.
Furthermore, you will be able to check how your answers have been stored, as well as all the programs behind the
questionnaire (all the relevant files will be located in the same page as the questionnaire's main page).
As completing the questionnaire should only take around 20 minutes, I would be very happy to get it as soon
as possible, but please latest until the end of January, 2002.
For further questions and comments don't hesitate to contact me (contact information can be found at the bottom
of this page).
The choice of putting up the questionnaire in this electronic format (to make your task easier) precludes the sending
of a motivational chocolat bar. Yet, in appreciation for your time, I leave you with something I really enjoy (thank
you Mr. Bill Watterson):
I thank you very much for your time and cooperation.
Dorothea Hamdorf
Contact Address:
Dorothea Hamdorf
Email: Dorothea.Hamdorf@uni-konstanz.de
Phone: +49 - (0)7531 917425
Questionnaire on Conflict Behavior
As stated in the instructions page, in order to be able to perform my analysis, an anonymous code will be set, using the
answers to the following items:
Demographic Data
In the following section you will be requested to answer some questions about yourself.
Nationality:
Region of birth:
Age:
Gender:
Completed years of formal
education:
Understanding of English:
Management duties:
Years in this company:
In the following section you will find some statements about how you define yourself (your self-concept). Please
indicate to what degree you agree or disagree with each statement (1-5).
In the following section you will find some statements about conflict behavior.
Conflict is defined as the perceived and/or actual incompatibility of values, expectations, processes or outcomes between
two or more parties over substantive and/or relational issues.
Please picture a recent conflict you had with a colleague and indicate to what degree you agree or disagree with each
statement (1-5). Additionally, please indicate, if you perceive the statement to be task or relationship oriented (T-R).
IN8 I would make sure the other person realized that resolving
our differences was important. 1 2 3 4 5 T R
NE7 While in the presence of the other person, I would act as
though he/she did not exist. 1 2 3 4 5 T R
IN7 I would try to get us to work together to settle our
differences. 1 2 3 4 5 T R
AV7 I would generally "grin and bear it" when the other person
did something I did not like. 1 2 3 4 5 T R
NE3 I would say nasty things about the other person to other
people. 1 2 3 4 5 T R
NE1 When we discuss the problem I would refuse to cooperate.
1 2 3 4 5 T R
disagree agree e
nor agre
disagree Task Relationship
disagree agree e
nor agre
disagree Task Relationship
NE4 I would let the other person know that I did not want him/her
to ever talk to me again. 1 2 3 4 5 T R
EE4 I would use my feelings to determine what I should do in the
conflict situation. 1 2 3 4 5 T R
NE6 I would say and do things out of anger to make the other
person feel bad. 1 2 3 4 5 T R
DO4 I would use my expertise to make a decision in my favor.
1 2 3 4 5 T R
OB4 I would usually allow concessions to the other person.
1 2 3 4 5 T R
IN10 I would tell the other person that there were problems and
suggest that we work them out. 1 2 3 4 5 T R
AV2 I would try to stay away from disagreement with the other
person 1 2 3 4 5 T R
OB5 I would often go along with the suggestions of the other
person. 1 2 3 4 5 T R
TP6 I would appeal to a person at a higher level to settle my
conflict with the other person. 1 2 3 4 5 T R
DO5 I would be generally firm in pursuing my side of the issue.
1 2 3 4 5 T R
AV4 I would try to keep my disagreement with the other person
to myself in order to avoid hard feelings. 1 2 3 4 5 T R
AV1 I would attempt to avoid being "put on the spot" and try to
keep my conflict with the other person to myself. 1 2 3 4 5 T R
In the following section you will find some statements about conflict interaction. Please picture a recent conflict you
had with a colleague and indicate to what degree you agree or disagree with each statement (1-5).
OF1 I was concerned with maintaining the poise of the other person.
1 2 3 4 5
MF1 I was concerned with respectful treatment for both of us.
1 2 3 4 5
SF3 I didn't want to embarrass myself in front of the other person.
1 2 3 4 5
OF11 I was concerned with helping the other person to preserve his/her self-
image. 1 2 3 4 5
Contact Address:
Dorothea Hamdorf
Email: Dorothea.Hamdorf@uni-konstanz.de
Phone: +49 - (0)7531 917425