The role of self-concept in cross-cultural communication

Abstract

Empirical evidence supports the notion that communication behaviors in intercultural encounters are effectively extensions of cultural values as well as epistemologies. Study 1 established communication behaviors of Asians and New Zealanders (NZs) as consistent with vertical collectivism and horizontal individualism, respectively. In particular, argumentativeness is positively related to independent self-construal (SC) and negatively related to interdependent SC. This supports Markus and Kitayama’s SC theory. Study 2 showed that NZs exhibited more idiocentric and argumentative behavior, while Asians displayed more sociocentric and less argumentative behavior during two actual interactions; specifically, participants diverged in their communication styles to be more consistent with their cultural values during intercultural interactions. Analyses of decision outcomes provide support that culture moderates cognitive consistency behaviors such that NZs exhibited more inconsistency-reduction behaviors, which is rooted in adherence to noncontradiction. In contrast, Asians exhibited more inconsistency-support behaviors, suggesting that naive dialecticism rooted in acceptance of contradiction is customary in Asian social interaction.

Keywords

Argumentativeness dialecticism idiocentric style intercultural communication self-concept sociocentric style values

Communication style as extensions of internalized cultural values and epistemology

The topic of cross-cultural communication has received some well-deserved attention in the literature in the past (Adair et al., 2001; Limaye and Victor, 1991; Lindsley, 1999) and recently due to the changes in the world’s workforce; now large percentages of a nation’s workforce come from a variety of cultural backgrounds (Okoro and Washington, 2012; MacKenzie and Forde, 2009). Prescriptions for effective communication in intercultural encounters often suggest adapting one’s behavior to that of the other culture; however, empirical evidence for actual behaviors and effectiveness of adaptive strategies is still equivocal (Gallois et al., 2004; Francis, 1991; Thomas and Ravlin, 1995; Tse et al., 1994). In fact, Adair et al.’s (2001) work suggests that adaptation by negotiating parties does not necessarily ensure the most effective outcomes.

This study aims to investigate communication tendencies of people from different cultures and the extent these styles are shown during initial intercultural interactions. To this end, two studies were conducted. Study 1 examined the relationship between individuals’ cultural values, self-concept, and argumentative behavior. Study 2 tested actual communication behaviors in intracultural and intercultural situations. As such, it extends previous works by examining the role of self-concept and communication behaviors during intercultural encounters that illustrate internalized cultural values as well as inconsistency-reduction and inconsistency-support behaviors (Kim et al., 2001; Pekerti and Thomas, 2003). The study also contributes to extant debate concerning Markus and Kitayama’s (1991) notion of self-concept and how it influences behaviors, showing that communication behaviors are extensions of independent self-construal (independent-SC) or interdependent self-construal (interdependent-SC).

Cultural values of individualism and collectivism

Consistent with previous research (Gudykunst et al., 1996; Kim et al., 2001; Pekerti and Thomas, 2003; Sanchez-Burks et al., 2003; Singelis and Brown, 1995; Suzuki and Rancer, 1994), we predict that culture influences communication behaviors in different situations. We adopted the vertical and horizontal dimensions of individualism and collectivism to differentiate our cross-cultural sample and base our predictions due to the broader variation of possible cultural profiles, that is, vertical collectivism (V-C) and horizontal collectivism (H-C) and vertical individualism (V-I) and horizontal individualism (H-I). In addition, research indicates that the polarity that emerges from the most sophisticated mapping of cultures to date (Schwartz, 1994) falls quite neatly along the broad cultural profiles of V-C and H-I (Smith and Bond, 1999). In short, verticality is conceptually equivalent to the power distance dimension, whereby the positive correlation between power distance and collectivism (r = 0.67, Hofstede, 1980) suggests that V-C and H-I may be the dominant cultural profiles around the world (Triandis, 1995).

The attributes of the V-C and H-I cultural profiles suggest the extent to which situational factors will influence behavior. For example, vertical collectivists see themselves as members of in-groups, acknowledging that members are different in terms of status. V-C cultures typically have social systems that do not reflect values of individual freedom and/or equity (Rokeach, 1973). In essence, verticalness causes one to be sensitive to cues coming from authorities, while collectivism predisposes them to be willing to sacrifice their individual goals (Triandis, 1995). In contrast, horizontal individualists are susceptible to self-referent factors in their exchange relationship because of their independent nature and high freedom of choice. H-I cultures are characterized by social systems that emphasize both the values of equality and individual freedom (Rokeach, 1973).

Communication style and internalized cultural values

A number of theoretical perspectives suggest communication styles also vary across cultures since they are logical extensions of internalized cultural values, which strongly influences style, conventions, and practices of communication (Gudykunst et al., 1996). One way in which communication varies as a function of culture is the extent to which messages are varied to communicate according to the culture’s norms and social beliefs (axioms) (Bond et al, 2004). The low-context (LC) and high-context (HC) communication typologies are one that posit communication is culturally driven (Hall, 1976). In LC cultures, messages are conveyed largely by verbalization, thus the use of both written and spoken modes, whereby meaning is contained in the verbalized message. However, in HC cultures, a good deal of the meaning is implicit, thus the verbal mode convey only a small part of the message. The rest of the message and meaning must be derived based on past knowledge of the speaker, setting, or other contextual cues.

Scholars of culture and communication have documented a link between communication style and cultural values, where LC and HC communications are linked to individualism and collectivism, respectively (Gudykunst et al., 1988; Hofstede, 1980). The logic of this relationship is based on the two primary functions of communication, affiliation (relational) and reducing uncertainty (functional) (Honeycutt, 1993). Hall (1976) argued that people in all cultures use both LC and HC communication just as individualism and collectivism exist in all cultures, but a culture will have predominate mode depending on that culture’s norms and social beliefs. Although the emphasis of communication on affiliation or reducing uncertainty varies according to situations, they also tend to vary according to culturally based motives for communication (Thomas et al., 2008).

Influence of interdependent-SC and independent-SC on communication

Literature suggests that socialization is the mechanism responsible for differing patterns of SCs, communication, and how people interpret information in individualistic and collectivistic cultures (Hofstede, 1980; Markus and Kitayama, 1991; Pekerti and Kwantes, 2011).

Markus and Kitayama (1991) posited that everyone in every culture has both independent-SC and interdependent-SC; however, similar to Hall’s (1976) view of communication modes one typology tends to predominate depending on the cultural milieu. They described people socialized in collectivistic societies as placing more value on social order and harmony (Markus and Kitayama, 1991; Oishi et al., 2004; Singelis, 1994; Singelis and Brown, 1995; Triandis et al., 1993). Actions stemming from a relational motive are labeled sociocentric, indicated by behaviors such as affiliative, flexible, avoiding arguments, fitting in with the environment and relevant others, and emphasizing group identity (Park and Levine, 1999), thus akin to HC communication (Gudykunst et al., 1996; Pekerti and Thomas, 2003).

In contrast, a functional approach to communication associated with the motive of achieving (Burgoon and Hale, 1987) is often associated with members of individualistic cultures (Kluckhohn and Strodbeck, 1961; Sanchez-Burks et al., 2003; Triandis et al., 1984). While most communication has both relational and functional purposes, when one is emphasized over the other, such as uncertainty reduction, it relegates the affiliative function to a secondary role. This, in turn, emphasizes personal identity and task orientation (Burgoon and Hale, 1987; Gallois et al., 2004; Park and Levine 1999). Behaviors associated with this task accomplishment motive are labeled idiocentric, indicated by direct communication and other actions such as propensity to argue (Gudykunst et al., 1996; Infante and Rancer, 1996; Pekerti and Thomas, 2003; Suzuki and Rancer 1994). We contend that these behaviors are akin to LC communication modes; please see Figure 1.

Figure 1.

Influence of culture on communication behavior (Source: Adapted from Gudykunst et al., 1996).

Extant literature suggests that individuals socialized in individualistic or collectivistic cultures are more likely to have an independent-SC and idiocentric behavior or interdependent-SC and sociocentric behavior, respectively (Markus and Kitayama, 1991; Triandis, 1989). Further, other scholars have documented that individualism and collectivism are associated with LC and HC communication (Gudykunst et al., 1996; Pekerti and Thomas, 2003). Based on the above discussions, we present the first of four hypotheses.

Hypothesis 1 (H1): Individuals from individualistic (Western) cultures will score higher on H-I and have a more salient independent-SC compared to individuals from collectivistic cultures (non-Western) who will score higher on V-C and have a more salient interdependent-SC.

Culture, self-concept, and argumentativeness

One communication construct that appears to support Hall’s (1976) and Markus and Kitayama’s (1991) theories is argumentativeness. Argumentativeness refers to a disposition that leads to the propensity to argue (Infante and Rancer, 1996). It is considered constructive since it “involves attacking the positions that others take on given issues as opposed to the negative disposition of verbal aggressiveness which involves attacking the self-concept(s) of others, rather than their positions” (Infante and Rancer, 1996: 320). Argumentativeness is related to workplace success, especially when used with an affirming communication style (Infante and Rancer, 1996). Argumentativeness has been validated cross-culturally, for example, Infante and Rancer (1996) as well as Suzuki and Rancer (1994) found members of collectivistic Japanese and Korean cultures less argumentative than members of individualistic North American culture. Kim et al.’s (2001) work indicate that individualism is positively linked to independent-SC and argumentativeness, including the idea that as one’s level of individualism decreases so does one’s argumentativeness.

We contend that argumentativeness is an extension of one’s cultural values and self-concept. Logically, collectivists with an interdependent-SC have lower tendencies to attack another’s position concerning a given issue or ideas because they do not make distinctions between ideas and the person who expresses them (Cross et al., 2002; Cross et al., 2000; Kim et al., 2001; Singelis and Brown, 1995). Since attacks on ideas would be perceived as an attack on the person, collectivists are less likely to be argumentative in communication interactions to ensure harmony with relevant others, especially if the other participants are of higher status. In contrast, individualists with an independent-SC do distinguish between ideas, and the person who expresses them thus do not mind attacking an idea and opposing it compared to collectivist, that is, being argumentative. Therefore, we present the following hypothesis regarding argumentativeness.

H2: Individuals with a more salient independent-SC are higher in argumentativeness compared to individuals with a more salient interdependent-SC.

Communication in intercultural interactions

Understanding dominant communication styles used in different cultures is an important step in understanding behaviors of individuals from different cultures. However, this understanding is insufficient in intercultural interactions since it fails to address the extent to which individuals might alter their preferred behavior in these situations. Although there are sound theories that predict probable strategies people may adopt during intercultural encounters (Gallois et al., 2004), to date, we know only few studies that document actual communication behaviors during intercultural interactions (e.g.Adair et al, 2001; Pekerti and Thomas, 2003). As such, extant empirical evidence is still equivocal regarding accommodation strategy in intercultural interactions. Accommodating one’s communication style so that it converges with another person’s style to bridge cultural distance is largely based on the similarity attraction paradigm (Byrne, 1971; Giles and Noels, 1997). Logically, convergence leads to perceptions of similarity, which leads to positive attitudes toward members of the other culture (Gallois et al., 2004). However, the extent to which stylistic accommodation is viewed positively seems to depend on the motive to which it is attributed (Thomas and Ravlin, 1995). Research indicates that there might be an optimal level of adaptation concerning another culture’s pattern of behaviors, whereby the effects are less positive if a person goes beyond this threshold (Gallois et al., 2004; Francis, 1991). In short, when people interact with different cultures, they have a choice of using either an adaptive strategy or not (Rao and Hashimoto, 1996; Tse et al., 1994).

Other research suggests that when situations fail to yield enough contextual clues to provide participants with cues to behave in a particular manner, people tend to rely on internal cues for guidance (Catrambone and Markus, 1987;Fiske and Taylor, 1991). For example, Laurent (1983) found individuals in multicultural environments exhibiting behaviors that were more characteristic of their own culture (i.e. divergence behavior) than in monocultural situations.

In sum, extant research suggests that people may alter their communication style during intercultural interactions; however, it is still unclear which strategy will bring about a more productive outcome (Adair et al., 2001). We suggest that in situations where cues for behavior are ambiguous, such as in initial encounters or novel situations, a strong case can be made for individuals to revert to preferred communication behaviors that are consistent with their cultural norms. To explore whether people diverged in their communication behavior in ambiguous situation, such as initial encounters or novel situations, we propose the following regarding communication behaviors in initial intercultural encounters.

H3: In initial intercultural interactions, individuals will behave in way that reflect their cultural values, such that individualist will communicate in an idiocentric manner (task oriented, direct, and propensity to argue) and collectivist in a sociocentric manner (affiliative, flexible, and avoid argument).

Cross-cultural comparison of cognitive consistency behaviors

An alternative concept that further explains and predicts differences in intercultural encounters is cognitive consistency theory. Cognitive consistency theory posits that all things being equal, individuals will change their attitudes and/or behaviors when there is discrepancy between their views and behavior to maintain consistency (Heider, 1946), especially, when the consequences are of functional importance to them. The theory posits that one might also attempt to influence others to change their attitudes, standards, or behaviors to maintain consistency, thus called inconsistency-reduction behavior (Iwao, 1997; Kelman and Baron, 1968a,1968b). In contrast, there are people who are comfortable with differences between their views, behaviors, and not attempt to change their own views and/or behavior or in others; these behaviors are called inconsistency-support behaviors.

Studies have shown cultural effects in cognitive consistency behaviors. Iwao (1997) and Iwao and Triandis (1993) found Japanese participants exhibited inconsistency-support behaviors by not expressing discrepancy between one’s private and publically expressed views regarding an issue. In contrast, Iwao and Triandis (1993) found Americans attempted to reduce their cognitive inconsistency by attempting to change another person’s opinion to maintain consistency between their privately held and their publically expressed views. Similarly, Khokhlov and Gonzalez (1973) found Americans changed their views to maintain consistency with a significant other (friend, father, or country), whereas Greeks were comfortable with perceived cognitive inconsistencies and did not change their views.

The importance of individuals’ SCs lies in the notion that it is the framework that activates the processing of information from the environment (Triandis, 1989). Therefore, we argue that SC is linked to cognitive consistency orientation and communication behavior. For example, collectivists with interdependent-SC place primary importance on maintaining relationships thus will behave in ways that avoid conflict, loss of face, and protect relationships (Cross et al., 2002; e.g. amae in Japan—Doi, 1974). We argue that Japanese and Greek participants in the previous experiments (Iwao, 1997; Iwao and Triandis, 1993; Khokhlov and Gonzalez, 1973) exhibited inconsistency-support behaviors because their sociocentric communication norms had relegated expressing an opinion that one believes as true secondary to maintaining harmony. Therefore, Greek and Japanese participants did not attempt to maintain consistency between their private and expressed opinions by attempting to change others’ opinions.

In short, people who have an interdependent-SC and values harmony are more likely to exhibit inconsistency-support behaviors, such as being less argumentative compared to people who have an independent-SC who are more likely to be more argumentative to maintain consistency; please see Figure 1. In addition, due to the orthogonality of SCs (Cross et al., 2000; Pekerti and Kwantes, 2011), one’s self-concept is more likely to be a better predictor of cognitive consistency behaviors than one’s cultural background.

Dialecticism as cultural foundations of cognitive consistency

A metaphysical and epistemological view, which explains cognitive consistency and inconsistency in Asians versus Westerners in the workplace, is how people deal with paradoxes and/or contradictions (Cameron, 1986; Chen, 2002; Lewis, 2000; Nisbett et al., 2001; Paletz and Peng, 2009; Peng and Nisbett, 1999; Peng et al., 2006). In her exposition, Lewis (2000: 760) claimed that paradoxes are socially constructed concepts that incorporate contradictory yet interrelated elements. These works suggest that Westerners approach paradoxes via dialectical thinking that is rooted in the concept of noncontradiction. Westerners recognize contradiction then move on toward resolving it via verbal debate and argumentation, which fits in with Western preference for consistency (Boucher et al., 2009; Nisbett et al., 2001; Peng et al., 2006; Peng and Nisbett, 1999; Spencer-Rodgers et al., 2010).

In contrast, East and South East Asian societies (Chinese, Korean, Japanese, Singaporean, and Thailand) approach life and paradoxes via the middle way that prescribes a holistic approach (Chen, 2002; Peng and Nisbett, 1999; Spencer-Rodgers et al., 2010). The middle way has been attributed to the teachings of Lao Tzu and Tao philosophy (Chen, 2001, 2002), including Confucian teachings (Chen, 2002; Fletcher and Fang, 2006). Symbolized by the yin–yang, and coined naive dialecticism, this approach accepts contradictions in life, since life involves as a constant state of flux (Peng and Nisbett, 1999; Nisbett et al., 2001; Spencer-Rodgers et al., 2010), to the extent that individuals can change depending on the context (Boucher et al., 2009; Peng et al., 2006).

A naive dialectism approaches apparent paradox by believing that both views of a contradiction might be right, thus truth lies between the two perspectives (Peng and Nisbett, 1999). Since both views can be correct, people can have differing views without having to persuade another to take another perspective. For example, Peng and Nisbett (1999) found Chinese participants more likely to accept two contradictory propositions compared to American participants, suggesting a holistic approach by the Chinese as opposed to a polarizing approach in dealing with contradictions.

Recent works suggest that tolerance for contradiction has been documented in India and in cultural milieus that accept multitheism (Spencer-Rodgers et al., 2010). Although cognitive consistency theory posits that the motives for maintaining consistency are universal, studies indicate that occurrence of inconsistency-reduction behaviors may be lower in some cultures (Boucher et al., 2009; Peng and Nisbett, 1999; Spencer-Rodgers et al., 2010). Thus, the need to reconcile the inner-self with external behaviors may be lower for collectivists with an interdependent-SC than for individualists with and independent-SC (Gudykunst et al., 1996; Kashima et al., 1992); that is, collectivists accept contradictions as part of life (Boucher et al., 2009; Peng and Nisbett, 1999; Spencer-Rodgers et al., 2010). Based on these findings and previous discussions, we propose the following hypothesis:

H4: In initial intercultural interactions, individualist with and independent-SC will exhibit inconsistency-reduction behavior compared to collectivist with an interdependent-SC who will exhibit inconsistency-support behavior.

The present research

Two studies were designed to examine the relationship between cultural value orientations and communication style. Study 1 examines the relationship between cultural values, SCs, and the propensity to argue. Study 2 investigated actual communication behaviors in intracultural versus intercultural interactions with the aim of explaining behavioral adjustment strategies in intercultural interactions.

Study 1

Study 1 tested predictions regarding individuals’ cultural orientation, self-concept, and propensity to argue.

Method

Participants

Participants were undergraduate students from a large university in New Zealand (NZ). Surveys were administered in English, which was the language of instruction at the university. A total of 193 participants took part in the study, consisting of 96 Asian international students from East and South East Asia, 97 NZ Caucasian participants with NZ, English, European and North American descent. The average age for NZ participants was 21.1 years and that for Asians was 22.2 years. Of the 97 New Zealanders (NZs), 61percent were women, while 70 percent of the 96 Asian participants were women.

To account for acculturation effects, Asian participants were required to meet two criteria to control for possible acculturation effects. First, they had to be either first-generation Asian in NZ or international students.

Measures

Argumentativeness

We measured argumentativeness using Infante and Rancer’s (1982) 20-item scale, measuring the tendency to argue (ARGap) and avoid arguments (ARGav). The scale had been validated in individualist and collectivist cultures (Kim et al., 2001; Infante and Rancer, 1996; Suzuki and Rancer, 1994). To establish construct equivalence for this study, the scale was subjected to a confirmatory factor analysis (CFA) using structural equation modeling in each culture simultaneously (Bollen, 1989; van de Vijver and Leung, 1997). The model fit for factors ARGav and ARGap had goodness of fit indexes of 0.97, indicating equivalence of the measure across cultures. The reliabilities of the resulting factors were α = 70 for ARGap and α = 0.63 for ARGav.

Horizontal-, vertical-, individualism and collectivism

Horizontal-, vertical-, individualism and collectivism scales were measured with Triandis and Gelfand’s (1998) scale. Construct equivalence was established using structural equation modeling. The resulting four scales had goodness of fit indexes ranging from 0.98 to 0.99. The internal reliabilities of the resulting scales were α = 0.70 for H-I, α = 0.70 for V-I, α = 0.70 for H-C, and α = 0.70 for V-C.

Self-construal

Independent-SC and interdependent-SC were measured with Singelis’ scale (Singelis, 1994; Yamada and Singelis, 1999). Construct equivalence was established using structural equation modeling. The independent and interdependence scales had goodness of fit indexes of 0.97. The internal reliability coefficients were α = 0.60 for independence and α = 0.72 for interdependence, which are similar to other works (Green et al., 2005).

Results

A multivariate analysis of variance (MANOVA) was performed to test H1 and H2 with culture as the independent variable (Asians and NZs) and H-I, V-I, H-C, V-C, independence, interdependence, and argumentativeness as the dependent variables. There was an effect for culture, F (Wilks criterion; 7, 192) = 10.23, p < 0.001, partial (p) η² = 0.28; univariate effects are presented in Table 1. As shown in Table 1. NZs are higher on H-I compared to Asians, while Asians are higher on V-C and interdependence than NZs. Although not significant, the mean scores indicated that NZs are nominally more argumentative compared to Asians. These results particularly support H1, while results for argumentativeness are in the predicted direction since Asian participants with higher in interdependence are showing lower argumentativeness.

Table 1.

MANOVA, culture and gender on argumentativeness, horizontal–vertical individualism–collectivism, self-concept.

Dependent variable	Culture
Dependent variable	M Asians	SD	M NZ	SD	F(1, 192)	η²
H-I	6.7	1.33	7.1	0.96	8.33*	0.04
V-I	7.3	1.48	7.4	1.84	0.48	0.00
H-C	11.4	1.74	11.6	1.75	0.77	0.01
V-C	8.5	1.38	7.5	1.37	26.54**	0.11
Independent	2.0	0.29	2.06	0.28	2.03	0.01
Interdependent	2.29	0.34	2.0	0.37	31.9**	0.13
Argumentativeness	5.47	7.31	6.6	9.69	.83	0.00

Note: MANOVA: multivariate analysis of variance; H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism; NZ: New Zealand.

*p < 0.01; **p < 0.001.

Within-group correlations in Table 2 confirmed our predictions. Within NZs, argumentativeness was positively related to independence and V-I, while argumentativeness is negatively related to H-C. Interdependence was positively related to H-C and V-C, while independence is positively related to H-I and V-I. The Asian sample showed some interesting correlations, for example, argumentativeness is negatively correlated to interdependence but positively linked to independence, interdependence and independence are positively correlated, argumentativeness is positively related to H-I and V-I, interdependence is positively related to H-C and V-C, and independence is positively related to H-I and V-I as well as H-C and V-C. These findings indicate that individuals can be both independent-SC and interdependent-SC (Cross et al., 2000; Markus and Kitayama, 1991; Pekerti and Kwantes, 2011; Singelis, 1994).

Table 2.

Means, standard deviations, and intercorrelations of variables in study 1 for Asian and New Zealand data.

	Asian data (n = 96)
	M	SD	1	2	3	4	5	6	7
1 Argumentativeness	5.47	7.31	–
2 Interdependent	2.29	0.34	–0.12	–
3 Independent	2.00	0.29	0.13	0.27**	–
4 H-I	6.7	1.33	0.22*	–0.05	0.61**	–
5 V-I	7.3	1.48	0.27**	–0.06	0.21*	0.37**	–
6 H-C	11.4	1.74	–0.09	0.44**	0.22*	–0.00	–.09	–
7 V-C	8.5	1.38	0.03	0.55**	0.31*	0.17	–.03	0.63***	–
	New Zealand data (n = 97)
	M	SD	1	2	3	4	5	6	7
1 Argumentativeness	6.60	9.69	–
2 Interdependent	2.00	0.37	–0.13	–
3 Independent	2.06	0.28	0.33**	0.02	–
4 H-I	7.1	0.96	0.14	0.02	0.55**	–
5 V-I	7.4	1.84	0.35**	0.09	0.28**	0.18	–
6 H-C	11.6	1.75	–0.22*	0.46**	0.17	0.13	–0.03	–
7 V-C	7.5	1.37	–0.04	0.67**	0.15	0.11	0.22*	0.51**	–

Note: H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism.

*p < 0.05; **p < 0.01; ***p < 0.001.

Further analyses confirm that for Asians the concept of independence is driven by both H-I and V-C, but H-I is the predominant source of independent-SC, while V-C solely drives the concept of interdependence, as presented in Table 3. In contrast, argumentativeness is positively linked to the NZs' independent-SC and negatively linked to interdependence, thus supporting the argument that argumentativeness is effectively driven by one’s independent-SC (Kim et al., 2001; Infante and Rancer 1996; Suzuki and Rancer, 1994), as given in Table 4. Taken together, these findings partially support H2, in particular, independent-SC is positively linked to argumentativeness.

Table 3a.

Backward regression analyses testing link between independence and individualism–collectivism variables in Asian sample.

Equation	Variables	Predictor	B	ΔF	R²	pΔR ²
1	Independence (overall, F(4, 91) = 16.95***)			16.95	0.424	424***
		H-I	0.584***
		V-I	0.009
		H-C	0.151
		V-C	0.115
2	Independence (overall, F(3, 92) = 22.57***)			0.012	0.424	0.000
		H-I	0.587***
		H-C	0.150
		V-C	0.115
3	Independence (overall, F(2, 93) = 32.41***)			0.211	0.411	–0.013
		H-I	0.570***
		V-C	0.213**

Note: H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism.

*p < 0.05; **p < 0.01; ***p < 0.001.

Table 3b.

Backward regression analyses testing link between interdependence and individualism–collectivism in Asian sample.

Equation	Variables	Predictor	B	ΔF	R²	pΔR ²
1	Interdependence (overall, F(4, 91) = 11.35***)			11.35	.333	333***
		H-I	−0.133
		V-I	0.018
		H-C	0.126
		V-C	0.495***
2	Interdependence (overall, F(3, 92) = 14.32 ***)			1.973	0.318	–0.014
		V-I	–0.031
		H-C	0.144
		V-C	0.460***
3	Interdependence (overall, F(2, 93) = 21.62***)			0.129	0.317	–0.001
		H-C	0.148
		V-C	0.458***
4	Interdependence (overall, F(1, 94) = 41.13***)			0.178	0.304	–0.013
4	Interdependence (overall, F(1, 94) = 41.13***)	V-C	0.552***

Note: H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism.

*p < 0.05; **p < 0.01; ***p < 0.001.

Table 3c.

Backward regression analyses testing link between argumentativeness and independence-interdependence in Asian sample.

Equation	Variables	Predictor	B	ΔF	R²	pΔR ²
1	Argumentativeness (overall, F(2, 93) = 2.15)			2.15	0.44	0.44
		Interdependence	–0.172
		Independence	0.176
2	Argumentativeness [(verall, F(1, 94) = 1.60)			2.67	0.17	0.17
2	Argumentativeness [(verall, F(1, 94) = 1.60)	Independence	0.130

Note: H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism.

Table 4a.

Backward regression analyses testing link between independence and individualism–collectivism variables in NZ sample.

Equation	Variables	Predictor	B	ΔF	R²	pΔR ²
1	Independence (overall, F(4, 92) = 12.55***)			12.55	0.353	353***
		H-I	0.503***
		V-I	0.197*
		H-C	0.108
		V-C	0.003
2	Independence (overall, F(3, 93) = 16.31***)			1.178	0.345	–0.008
		H-I	0.514***
		V-I	0.180*
		V-C	0.061
3	Independence (overall, F(2, 94) = 24.35***)			0.498	0.431	–0.04
		H-I	0.519***
		V-I	0.193*

Note: H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism; NZ: New Zealand.

*p < 0.05; **p < 0.01; ***p < 0.001.

Table 4b.

Backward regression analyses testing link between interdependence and individualism–collectivism variables in NZ sample.

Equation	Variables	Predictor	B	ΔF	R2	pΔR2
1	Interdependence (overall, F(4, 92) = 22.37***)			22.37	0.493	0.493***
		H-I	–0.061
		V-I	–0.035
		H-C	0.153
		V-C	0.622***
2	Interdependence (overall, F(3, 93) = 29.72***)			0.650	0.489	–0.004
		V-I	–0.046
		H-C	0.145
		V-C	0.622***
3	Interdependence (overall, F(2, 94) = 44.71***)			0.356	0.488	–0.002
		H-C	0.153
		V-C	0.608***

Note: H-I: horizontal individualism; V-I: vertical individualism; H-C: horizontal collectivism; V-C: vertical collectivism; NZ: New Zealand.

*p < 0.05; **p < 0.01; ***p < 0.001

Table 4c.

Backward regression analyses testing link between argumentativeness and independent-interdependent in NZ sample.

Equation	Variables	Predictor	B	ΔF	R²	pΔR ²
1	Argumentativeness (overall, F(2, 94) = 7.02***)			7.02	.13	0.13***
		Interdependence	–0.137
		Independence	0.336***
2	Argumentativeness (overall, F(1, 95) = 11.88***)			2.03	.02	0.02***
		Independence	0.333***

*p < 0.05; **p < 0.01; ***p < 0.001.

Mediating effect of self-concept

To examine the possible mediating effect of self-construal as depicted in Figure 1, we used the three-equation approach recommended by Baron and Kenny (1986) as well as Preacher and Hayes (2004). Since H-I and V-C were found to be the factors driving independent-SC and interdependent-SC, respectively, we limited the mediation analyses to these particular relationships. Mediated regression results for interdependence and V-C on argumentativeness as well as independence and H-I on argumentativeness are presented in Table 5. As shown, the pattern of results supports a mediated relationship between H-I, independence, and argumentativeness (Preacher and Hayes, 2004; Preacher et al., 2007). That is, the relationship between H-I on independence and propensity to argue is as predicted, then H-I on argumentativeness becomes nonsignificant when independence is in the equation, the Sobel test supports this conclusion (Z = 2.47, p < 0.01). However, this effect was not found for V-C, interdependence, and argumentativeness; the Sobel test concurs (Z = 0.43, p < 0.068). As such, it confirms that an individual’s independent-SC will mediate one’s propensity to argue to the extent the more independent one is the more likely he/she is to be argumentative.

Table 5.
Test for mediated regression of independence and horizontal individualism on argumentativeness as well as interdependence and vertical collectivism on argumentativeness.

Equation Predictor B ΔF R² pΔR ²

Independence and horizontal individualism on argumentativeness

1 Independence Horizontal individualism 0.584* 98.77 0.341 0.000*

2 Argumentativeness Horizontal individualism 0.18 6.38 0.032 0.012

3 Argumentativeness Horizontal individualism 0.055 6.34 0.063 0.002

Independence 0.214

Equation Predictor B ΔF R² pΔR ²

Interdependence and vertical-collectivism on argumentativeness

1 Interdependence Vertical vollectivism 0.670* 155.99 0.450 0.000*

2 Argumentativeness Vertical collectivism –0.029 0.163 0.001 0.687

3 Argumentativeness Vertical collectivism 0.121 2.785 0.028 0.064

Interdependence –0.224*

p < 0.05; p < 0.01; p < 0.001

For the most part, results of study 1 are consistent with our predictions; however, a few unexpected findings were observed. Positive relationships between collectivism and independence in the Asian data are examples of such findings. These findings, however, are not surprising since the sample were students. In general, university students are trained to think independently, thus Asian students living and studying in individualistic nations may be more independent compared to Asian students living in their home nations. This also explains why Asians were not significantly different in independence compared to NZs. These results illustrate the advantages and disadvantages of obtaining data from a student sample. Heine et al. (2001) suggest that gathering data from people of similar background in different countries such as university students can validate cultural differences when they exist, which we did discover. However, at the same time, the similarities in experiences and opportunities that students have may also minimize potential cultural differences that may otherwise exist. Therefore, Heine et al.’s (2001) work suggests that if we were to interview Asians and asked to compare themselves with NZs, it is likely that Asians may view themselves as significantly less argumentative than NZs.

Summary and discussion

Study 1 provides empirical evidence of the relationship between people’s cultural value orientation, SC, and propensity to argue. Individuals from traditionally collectivistic culture were found to be more interdependent than those from traditionally individualistic culture. Results lend support to Markus and Kitayama’s (1991) theory in that H-I was the predominant factor driving independent-SC, while V-C was the predominant factor driving interdependent-SC.

The result also meets the seven criteria set by Matsumoto (1999) for validating the influence of culture on SC and behaviors. In short, using samples from two cultural backgrounds, study 1 showed that there was an effect for H-I and V-C and for SC with both effects and correlations in the predicted direction as described by Markus and Kitayama’s (1991) theory. Culture influenced the propensity to argue through the intermediate variable of SC. In particular, independent-SC was positively linked to argumentativeness, and the mean scores for argumentativeness were in the predicted direction with NZs more argumentative compared to Asians.

The nominal effect of argumentativeness and significant relationship between independence and argumentativeness found in this study are consistent with Kim et al.’s (2001) work. They found Koreans less argumentative compared to US participants from the mainland and from Hawaii but did not find a significant main effect for argumentativeness (Kim et al., 2001). The results suggest that when used with a student sample, the argumentativeness scale may suffer the lack of reference effect, which serves to minimize cultural effect (Heine et al., 2001). As such, future studies using the argumentativeness measure may need to specify a reference group to compare with to test predictions.

In study 2, we extend our findings by testing the link between culture and communication and adjustment behavior in intercultural interactions.

Study 2

Study 2 tested predictions that collectivists and individualists will exhibit sociocentric and idiocentric communication behavior, respectively, whereby in initial intercultural interactions, the two groups will exhibit communication styles that are representative of their own cultural norms. Finally, we tested the prediction that individualist are more likely to exhibit inconsistency-reduction behavior than collectivist who are more likely to exhibit inconsistency-support behavior.

Method

Participants

Participants were Caucasian NZs as well as East and South East Asian students at a large NZ university. Asian participants met two criteria to control for possible acculturation effects. First, they had to be a first-generation Asian or international student.The sample consisted of 96 volunteer participants with an average age of 23 years; however, each participant was paid NZ$10.00. The NZ Caucasian sample consisted of 24 men and 24 women who were from NZ of NZ, European, and North American descent; one participant was born in NZ with parents from South Africa. The Asian sample consisted of 24 men and 24 women from East and South East Asia. They were predominantly of Chinese ancestry with the exception of two participants who were indigenous Indonesians.

Procedures and experimental design

Participants interacted in a dyadic consensus communication task performed at two different times. This design was included to test whether subsequent interactions might affect behavior patterns since familiarity can change behaviors between dyads (Honeycutt, 1993; Neff and Karney, 2005). The interactions were videotaped and the setting standardized. The study used a 2 × 2 (Culture × Condition) design, with participants in each cultural groups interacting in either intercultural or intracultural condition. Because communication research has indicated that social roles and gender can affect communication behaviors (Fletcher et al., 1986), participants were uniformly blocked for gender and age, as well as culture, before being randomly assigned to each condition (Cook and Campbell, 1979). This blocking controlled for possible effects of gender and/or status caused by age.

Consensus task

The experimental task was to rank the severity of 15 crimes (Taylor, 1987/88). Each participant was asked to rank the crimes as quickly and as accurately as possible, individually (individual ranking). Participants were informed that there was a model answer derived from expert opinions consisting of lawyers and judges. After participants completed the task individually, they were asked to interact and complete a new set of rankings via consensus (consensus ranks); a maximum of 15 min was given to complete the consensus ranks. The consensus ranks were then compared against the model answers and dyads were given the feedback, “you did not do so well compared to the expert rankings”. They were then asked to improve their consensus ranks in a second attempt. This particular consensus task was chosen because it allowed participants to form individual opinions and to think about the rationales of why one maintains those opinions and/or change them during the task at two different points in time. The task fits the study well since we test whether participants’ communication styles and opinions differ at two different points in time. The interactions were conducted in English, which was the language of instruction at the university.

Scoring of crime-ranking task

The scoring of the crime-ranking task was as follows:
Individual scores (I) were the sum of absolute value of the differences between their individual ranking and the expert ranking.

Difference scores between individuals (DI) were the sum of absolute value of differences between individual’s ranks on each crime.

Difference scores between individual’s ranks and consensus ranks (DIC) tell us how much an individual had changed his/her rankings during the consensus task, this was the sum of the absolute value difference between the individual’s ranks and consensus ranks.

Because of the within-person design, we were able to record scores from Consensus Task at time 1 (CT1) and consensus task at time 2 (CT2).

Coding procedure of communication task

The experiment resulted in 96 videotaped communication interactions. Three coders unfamiliar with the study’s hypotheses coded the interactions. The coders represented both the Anglo NZ and Asian cultures and both genders. All coders were trained using videotaped interactions that were not part of the data. All coders completed six practice rounds before they coded the actual interactions. Based on observations, pilot study, and interviews, including previous research of what constitutes as dyadic and business communication behaviors (Burgoon and Hale, 1987; Gudykunst et al., 1988; Hall, 1976; Hinkle et al., 1988), 12 behavior categories were used to code the participants communication patterns. Interactions were coded according to the occurrence and intensity of behaviors consistent with dyad communication interactions. Each dyad were labeled A or B on the screen and coders rated who displayed more behaviors than the other participant did, then rated the intensities of these observed behaviors on a 7-point rating scale. For example, all participants displayed expressions; however, one participant might have been more emotive and colorful in his/her expressions; this difference in degree of expressiveness was the basis for the intensity rating. Cognitive consistency behaviors were determined by the degree of change between one’s individual crime–rank scores at CT1 and CT2 compared to the consensus scores at CT1 and CT2, respectively.

The interrater criterion was established at two of the three coders concurring on a coded behavior for it to be considered valid for analysis; thus, all data put forward for analysis had at least 0.67 interrater agreement. The study’s interrater agreement for the data was 0.88, which is above the acceptable recommended value of 0.61 to 0.80 for interrater reliability (Landis and Koch, 1977).

Results

Communication style

Testing our predictions involved examinations of participants’ behaviors intraculturally and interculturally; results are presented in Table 6a and 6b. Since each behavior occurred with reference to the other participant, the cell counts in Table 6a and 6b depict judgments identifying the participants that exhibited the most of each behavior in the three interaction conditions during CT1 and CT2, respectively.

Table 6a.
Comparative judgments of observed behaviors in intercultural and intracultural interactions during the consensus task at time 1.

Experimental condition

Individual behavior Intercultural NZ intracultural Asian intracultural

NZ NZ Asian Asian NZ Asian

Expressiveness 8 8 6 10 13 3

Dominance 10 6 7 9 14 2

Initiating action 7 9 6 10 13 3

Aggressiveness 8 8 7 9 13 3

More logical argument 10 6 9 7 12 4

Regulating flow of interaction 11 5 4 12 11 5

Finishing task 11 5 7 9 11 5

Stronger opinion 11 5 8 8 13 3

Attempted eye contact 7 9 4 12 12 4

Accommodating 7 9 7 9 3 13

Avoidance of argument 7 9 7 9 5 11

More inclined to shift opinion 6 10 9 7 3 13

Note: χ² for boldface cell in the same row and same column are significant at p < 0.05. Italicized behaviors denote behaviors that are akin to sociocentrism.

Table 6b.
Comparative judgments of observed behaviors in intercultural and intracultural interactions during the consensus task at time 2.

Experimental condition

NZ-Intracultural Asian-Intracultural Intercultural

Individual behavior NZ NZ Asian Asian NZ Asian

Expressiveness 10 6 10 6 13 3

Dominance 9 7 10 6 15 1

Initiating action 8 8 9 7 15 1

Aggressiveness 9 7 9 7 15 1

More logical argument 7 9 11 5 14 2

Regulating flow of interaction 9 7 10 6 14 2

Finishing task 7 9 8 8 13 3

Stronger opinion 10 6 12 4 15 1

Attempted eye contact 8 8 7 9 10 6

Accommodating 6 10 5 11 3 13

Avoidance of argument 7 9 5 11 3 13

More inclined to shift opinion 8 8 6 10 2 14

Note: χ² for boldface cell in the same row and same column are significant at p < 0.05. Italicized behaviors denote behaviors that are akin to sociocentrism.

As shown in Table 6a and 6b, individuals in the intracultural interactions were equally likely to exhibit each behavior during the interactions; the exception was in the Asian intracultural condition where one of the dyad members had regulated the flow interaction and attempted eye contact in CT1 and stronger opinions in CT2. In the intercultural condition, NZs and Asians demonstrated contrasting behaviors. NZs exhibited expressiveness, dominance, initiating action, aggressiveness, logical argument, stronger opinion and attempted eye contact in CT1; the same pattern was found for NZs in CT2 with additional behaviors regulating flow, finishing task being significantly different. Taken together, these behaviors are akin to idiocentric behaviors. In contrast, Asians exhibited behaviors that can be considered sociocentric behaviors, namely, accommodating, avoidance of arguments, and shifting opinion in both CT1 and CT2. The results suggest that in intercultural situations, both groups exhibited communication behavior consistent with their cultural norms; NZs were observed to be task oriented and opinionated (idiocentric), while Asians were flexible and accommodating (sociocentric).

Further analysis focused on the intensity of behaviors exhibited by participants, which was analyzed by conducting a 2 × 2 MANOVA controlling for gender. In CT1, a significant effect was found for culture F (Wilks criterion; 12, 95) = 2.16, p < 0.02; p η² = 0.25, but not for condition. The effect for gender was also not significant. However, there was a significant interaction effect for Culture × Condition, F (12, 95) = 1.92, p < 0.04; p η² = 0.22. These multivariate results were replicated in CT2.

Univariate analysis indicated that participants’ culture had a significant effect on 11 of 12 behaviors in CT1 (finishing task was not significant) and 10 of 12 behaviors in CT2 (finishing task and attempted eye contact were not significant). Taken together, NZs exhibited more intensity for idiocentric behaviors while Asians for sociocentric behaviors. The impact of culture and situation was reflected by significant Culture × Condition interactions on 10 of 12 behaviors in CT1 (finishing task and avoidance of argument were not significant) and 11 of 12 behaviors in CT2 (finishing task was not significant).

Examination of the interaction effects indicated that communicating with a member of another culture enhanced the intensity of idiocentric behaviors for NZs and of sociocentric behaviors for Asians on both occasions. Table 7a and 7b depicts the intensity of observed behaviors in the three interaction conditions in CT1 and CT2, respectively. As depicted in Table 7a, individuals in the two intracultural interactions were equally likely to exhibit similar intensity of behaviors during their interactions; the exceptions were for regulating flow of interactions and attempted eye contact in the Asian intracultural condition. However, in the intercultural condition, all behavior categories were significant, NZs exhibited higher intensities expressiveness, dominance, initiating action, aggressiveness, logical argument, regulating flow, stronger opinion, attempted eye contact, while Asians exhibited higher intensities of behaviors accommodating, avoidance of arguments, and shifting opinion in CT1.

Table 7a.
Summary of intensity in communication behaviors in intercultural and intracultural interactions for consensus task at time 1.

Experimental condition

NZ intracultural Asian intracultural Intercultural

Individual behavior NZ NZ Asian Asian NZ Asian

Expressiveness 1.79 1.7 1.91 1.93 3.42* 0.58*

Dominance 1.78 1.34 1.65 2.34 3.41* 0.27*

Initiating action 1.24 1.36 1.43 1.68 2.86* 0.66*

Aggressiveness 1.61 1.67 1.41 1.64 2.56* 0.53*

More logical argument 2.11 0.99 1.59 1.46 2.88* 0.56*

Regulating flow of interaction 1.62 1.03 0.88* 2.01* 2.48* 0.31*

Finishing task 1.3 1.08 1.19 1.15 1.65* 0.57*

Stronger opinion 1.98 1.53 2.23 1.7 3.23* 0.55*

Attempted eye contact 1.62 1.8 0.66 2.58 2.62* 0.61*

Accommodating 1.06 1.76 2.04 2.44 0.45* 3.14*

Avoidance of arguments 1.15 1.42 2.28 1.66 0.64 2.29

More inclined to shift opinion 1.14 1.74 2.09 1.77 0.47* 2.59*

Note: NZ = New Zealand. Bold means denote higher scores within the columns significant at: p < 0.05; p < 0.01; p < 0.001. Italicized behaviors denote behaviors that are akin to sociocentrism.

Table 7b.
Summary of intensity in communication behaviors in intercultural and intracultural interactions for Consensus Task at Time-2.

Experimental condition

NZ-Intracultural Asian-Intracultural Intercultural

Individual behavior NZ NZ Asian Asian NZ Asian

Expressiveness 2.32 1.36 2.16 1.29 3.27* 0.55*

Dominance 2.03 1.61 2.84 1.49 4.13* 0.19*

Initiating action 1.4 1.11 2.19 1.2 3.38* 0.14*

Aggressiveness 1.76 1.3 2.12 1.23 2.89* 0.22*

More logical argument 1.46 1.49 2.44 0.63 3.38* 0.25*

Regulating flow of interaction 1.47 1.07 2.13* 1.02* 3.13* 0.47*

Finishing task 0.98 1.44 1.62 1.29 1.91 0.44

Stronger opinion 1.91 1.33 2.87 1.01 3.60* 0.23*

Attempted eye contact 1.54 1.3 1.53 1.54 1.71 1.27

Accommodating 1.32 2.03 1.41 3.51 0.50* 3.62*

Avoidance of arguments 1.33 1.49 1.38 2.64 0.36* 3.01*

More inclined to shift opinion 1.37 1.37 1.36* 2.71* 0.31* 3.12*

Note: NZ = New Zealand. Bold means denote higher scores within the columns significant at: p < 0.05; p < 0.01; p < 0.001. Italicized behaviors denote behaviors that are akin to sociocentrism.

In CT2, the results were replicated for the NZ intracultural condition and intercultural condition; however, there were slight variations in the Asian intercultural condition. As shown in Table 7b, Asians in the intracultural condition exhibited normative dyad behavior, where one person exhibited higher intensities of regulating behavior and presenting logical arguments and a stronger opinion, while his/her counterpart exhibited higher intensities of accommodating behavior and avoiding arguments and shifting his/her opinions.

Cognitive consistency

Examining our predictions with regard to inconsistency reduction or inconsistency support involved an examination of the crime-ranking scores from CT1 and CT2. First, ANOVA was conducted on the difference scores between individuals’ rankings (DI) across the three interaction conditions. The ANOVA showed a main effect, F (Wilks criterion; 2, 47) = 8.27, p < 0.001, p η² = 0.27, the largest difference was found in the intercultural condition (M = 52.06), then Asian intracultural condition (M = 49.88) and smallest in the NZ intracultural condition (M = 34.94). Because the consensus ranks were derived during dyad interactions, we also examined the difference scores between individual and consensus ranks for both participants A and B (DIC A and DIC B, respectively) in each interaction condition in CT1 and CT2.

Table 8 presents the mean difference scores between individual versus consensus ranks at CT1 and CT2. At CT1, the difference scores between participants in the NZ intracultural condition was the highest and close to significant F(1, 31) = 3.93, p < 0.057. The mean of difference scores between individual versus consensus ranks in CT2 were not significant in all three conditions. Despite NZs in the intracultural interactions having the least difference (M = 34.94) in their rankings, the actual difference between their individual and their consensus ranks indicate that one person in the dyad made larger changes compared to the other person in the interaction. Therefore, it is possible that the more dominant person persuaded the other to a greater degree in an attempt to maintain consistency between his/her individual ranks by changing the other person’s view.

Table 8.
Within condition ANOVA for difference between individual ranks versus consensus ranks (DIC) at CT1 and CT2.

Means CT1

DIC Participant A Participant B F(1, 30)

Asian intracultural 33.44 30.69 0.33

NZ intracultural 17 23.43 3.93*

Intercultural 30.38 36.19 1.29

DIC Means CT2

Participant A Participant B F (1, 30)

Asian intracultural 43.69 43.69 0

NZ intracultural 30.19 30.06 0.001

Intercultural 38.06 39.75 0.072

Note: CT1 = consensus task time 1; CT2 = consensus task time 2. DIC scores indicate difference between individual rankings and consensus ranks.

p = 0.057: indicates that participant B* made more changes nominally than participant A.

To explore which of the two participants in the dyad made the largest change between their individual ranks during the consensus task, we regressed the difference between individuals’ ranks (DI) versus the consensus ranks (DIC A and DIC B). As shown in Table 9, in CT1, in the NZ intracultural interactions, participant B made significant changes toward the consensus rankings, β = 0.82, p < 0.01, while participant A made relatively smaller changes. In the Asian intracultural condition, both participants made significant changes toward the consensus rankings, β = 0.72, p < 0.05; and β = 0.60, p < 0.05. In the intercultural condition, both participants made significant changes toward the consensus rankings, β = 0.42, p < 0.05 and β = 0.43, p < 0.05.

Table 9.
Regression analysis: difference scores between individual rankings and consensus ranks for CT1 and CT2.

CT1

Experimental condition DIC A DIC B

Asian intracultural β 0.72** 0.60*

SE β 0.22 0.26

DIC A DIC B

NZ intracultural β 0.36 0.82**

SE β 0.28 0.23

DIC NZ DIC Asian

Intercultural β 0.42* 0.43*

SE β 0.19 0.2

Experimental condition CT2

DIC A DIC B

Asian intracultural β 0.54* 0.64*

SE β 0.29 0.3

DIC A DIC B

NZ intracultural β 0.62* 0.2

SE β 0.34 0.27

DIC NZ DIC Asian

Intercultural β 0.3 0.39*

SE β 0.3 0.22

Note: CT1 = consensus task time 1; CT2 = consensus task time 2. Scores indicate the degree of change each participant (A and B) made during the consensus task; a higher score indicate larger changes made. DIC A and DIC B are the difference scores between participant A and participant B’s ranks and consensus ranks, respectively.

*p < 0.05; **p < 0.01.

In CT2, the pattern within the Asian intracultural condition was replicated; both participants made significant changes toward the consensus rankings, β = 0.54, p < 0.05 and β = 0.64, p < 0.05. However, the pattern in the NZ intracultural and intercultural condition changed. In CT2 of the NZ intracultural condition, participant A made significant changes toward the consensus rankings, β = 0.62, p < 0.05, whereas participant B made relatively smaller changes. This is different to the pattern found in CT1 since it was participant B who made the significant changes in CT1. In CT2 of the intercultural condition, it was only the Asian participants who made significant changes toward the consensus, β = 0.39, p < 0.05. Again this is different to the pattern found in CT1 since it was both participants who made significant changes to their rankings in CT1.

Taken together, it suggests that Asians are open to shift their opinions in intracultural andintercultural interactions, as opposed to their NZ counterparts; this effect was especially evident in CT2. This evidence suggests Asians were exhibiting inconsistency-support behavior, that is, they did not mind contradicting their initial position. In contrast, while some of NZs did not mind changing their minds, the results suggest that one was consistently persuading the other half of the dyad to change their minds. This finding corroborates our findings regarding communication styles in intercultural interactions, namely, that culturally based communication style became more pronounced during the intercultural interactions.

Summary and discussion

Study 2 contributes to our understanding of cross-cultural interactions in several ways. First, it provides actual empirical behaviors as opposed to recalled or intended strategies, documenting distinctive communication styles exhibited by collectivists and individualists during intercultural interactions. Second, it showed that idiocentric style was dominant for NZs and sociocentric style dominant for Asians as well as evidence for cognitive inconsistency-reduction versus cognitive inconsistency-support behaviors, respectively. Third, we documented that communication styles were different in intercultural compared to intracultural interactions, with patterns suggesting that in intercultural interactions participants were more likely to diverge and rely on their culturally dominant communication style as rather than converging to the style of their counterpart.

Fourth, notwithstanding the Asian intracultural condition showing variations in intensities of communication behavior in CT2 compared with CT1, it still points to the fact that Asians were the group exhibiting inconsistency-support behaviors. Analyses on the crime-ranking scores indicate that all Asian participants in the intracultural and intercultural conditions made significant changes in ranking the crimes toward the consensus in both CT1 and CT2. In the Asian intracultural condition, both participants made changes equally in both CT1 and CT2, while in the intercultural condition CT2, it was only the Asian participants who made significant changes. This provides some evidence that collectivists have a propensity to be flexible during communication interactions. This pattern is different to NZs. For the most part, it was only one participant or the other who significantly changed their rankings in the NZ intracultural condition. It suggests that one-half of the dyad consistently exhibited inconsistency-reduction behaviors, possibly to reduce cognitive dissonance (Peng and Nisbett, 1999). This corroborates that Westerners have a propensity to take a strong view of initial preference or positions compared to Asians (Nisbett et al., 2001; Peng and Nisbett, 1999; Spencer-Rodgers et al., 2010) and will argue to maintain this position.

General discussion

The generalizability of this research has some limitations associated with survey research and student samples. However, we contend that appropriate use of a student sample can properly reflect a fundamental process such as communication, especially for cross-cultural communication (Basil, 1996; Locke, 1986). A number of factors point to the appropriateness of our sample and suggest that our findings are generalizable to cross-cultural interactions. First, we used a sample from cultures that were different to NZ who can speak English fluently. Second, since we are investigating cultural patterns, our sample with mean age of 21.5 years has internalized their respective cultural values. Third, the task performed required argumentation, a skill that university students have been trained to do. This leads to our final point in that cross-cultural studies with samples that have similar experiences and opportunities across cultures, such as a student population may actually have higher validity than if we are less certain about the sample’s conditions (Heine et al., 2001). With that in mind, the sample was appropriate to address Matsumoto’s (1999) concerns regarding empirically demonstrating the validity of core beliefs underlying our theories and research. In particular, it showed that the theoretical predictions posed by Markus and Kitayama’s (1991) independent and interdependent construal theory were valid, despite the fact that the sample had cross-cultural experiences, which might have minimized the effect.

Underlying any intercultural interaction is communication behaviors and the processes involved in the interactions. Results of this research provide evidence that culturally based rules govern the style, conventions, and practices of language usage. Additionally, it validates communication styles that have not been empirically well documented in previous research (e.g. Glenn 1981; Hall, 1976; Servaes, 1989). Overall, findings indicate that Asians who are collectivists interdependents are less argumentative. These findings suggest that collectivists might ascribe a higher priority to the affiliative function of communication, which are prescribed by harmony-maintenance scripts found in collective cultures. In contrast, NZs who are more individualists are predisposed to be argumentative. It suggests that individualists may ascribe a higher priority to the functional facet of communication, which are associated with a direct and task-oriented style during communication (Gudykunst et al., 1996; Pekerti and Thomas, 2003).

A key issue in understanding intercultural interactions is the extent to which the act of interacting with a member of another culture influences behavior. Results showed that interacting with a member of a different culture enhanced individuals’ culturally dominant communication style. Therefore, participants diverged by exhibiting more and higher intensities of their culturally dominant behavior. We offer several possible reasons for this divergent behavior. First, interacting with a member of a different culture may have created uncertainty concerning the appropriate behavioral norms. Therefore, in the absence of explicit situational cues, individuals may have relied on their culturally based norms for guidance. In addition, the presence of a culturally different other may have caused individuals to represent their culture and motivated them to display culturally consistent behavior (Laurent, 1983). In either case, results suggest that the dominant tendency in initial intercultural interaction is not adaptation toward the style of the counterpart but enhancement of their culturally dominant behavior. These findings are consistent with other previous research on intercultural and intergroup interactions (e.g. Aberson et al., 2002; Laurent, 1983; Pekerti and Thomas, 2003).

Our present study partially corroborates Adair et al.’s (2001) and Sanchez-Burks et al.’s (2003) findings, which suggest that communication behaviors across context may be consistent for members of different cultures; however, the behaviors within each context are different when cross-cultural comparisons are made. The implication is that when cross-cultural encounters do occur, members of these different cultures are likely to communicate using different styles and behaviors thus may lead to misunderstandings if people are not prepared.

This study underscores that it is the interdependent person socialized in a collective society and not the independent person socialized in an individualistic society that has changed his/her behavior more during a consensus task; that is, collectivists/interdependents exhibit inconsistency-support behaviors (Iwao, 1997; Khokhlov and Gonzalez, 1973; Suh, 2002). It supports cognitive consistency theory’s premise that consistency behaviors are determined by individuals’ perception of value concerning a particular issue (Kelman and Baron, 1968b). This was evident in our study since NZs in the intercultural condition exhibited higher intensities of finishing the task and more occurrences (in CT2), suggesting that completing the task was prioritized, which facilitated idiocentric behaviors compared to Asian participants who may have prioritized other things, such as harmony.

Results also implied that inconsistency support could also be viewed as behavior that is consistent with the valued norm of harmony in collective cultures (Oishi et al., 2004; Suh, 2002). For example, in Japan, the ability to behave in a face-saving manner (tatemae) as opposed to expressing brutal truths (honne) verbally or nonverbally is valued. “In Japan inconsistency-support is at least on certain occasion valued more than inconsistency-reducing behavior” (Iwao, 1997: 324). The fact that both Asians and NZs in the intercultural condition made significant changes toward a consensus during the first interaction clearly indicated that both participants were active. However, it was only Asians who made the significant changes toward a consensus during the second interaction; taken together, this point to Asians in the intercultural condition not simply acquiescing during the second interaction but consciously invoking g sociocentric communication behaviors. This finding corroborates Cross et al.’s (2002) and Doi’s (1974) views that exhibiting sociocentrism such as avoiding conflict is in part face-maintenance behaviors that is paramount in collective and interdependent societies thus serving to avoid loss of face and protect relationships.

These findings also lend support to theories that posit naive dialecticism is prevalent a social processing norm in Asian societies (Peng and Nisbett, 1999); in practice, Asians are sensitive to change and do not mind changing their own views to maintain harmony (Nisbett et al., 2001; Paletz and Peng, 2009; Spencer-Rodgers et al., 2010). Along the same theoretical vein, the fact that NZ participants took turns persuading each other in the intracultural situation provide evidence for cognitive consistency or the need to reduce dissonance (Peng and Nisbett, 1999). In short, Western dialectical approach of noncontradiction and preference for consistency can manifest as a propensity to take a strong view of initial preference or position, which in turn may require arguing and persuading others to change their positions (Nisbett et al., 2001; Peng and Nisbett, 1999; Spencer-Rodgers et al., 2010).

Conclusions and implications

In summary, our study showed that communication styles are extensions of one’s cultural values, metaphysics, and epistemology. There are several practical implications for cross-cultural management from these findings.

First, where people were socialized is consistent with their cultural value orientations including their SC. H-I was found as a significant contributor to one’s independent-SC and, in turn, one’s propensity to argue. The implication for individuals and managers in intercultural settings is that they have a basis for systematically predicting probable communication behaviors of their counterparts or colleagues in initial interactions. At the same time, our results serve as a caveat that we must not interpret these findings as a form of stereotypes but rather to go beyond it. In particular, we found that Asians in our sample may have salience of both independent-SC and interdependent-SC. It points to the importance of knowing people’s background and experiences beyond where they come from as well as their chosen self-identity. In other words, an Asian with a more salient independent-SC can be argumentative. An awareness of this possibility can help managers not to interpret argumentativeness (an unexpected and non-stereotypical behavior) from members of collective societies as undesirable behaviors (Burgoon and Hale, 1988; Rosenthal, 2002) but rather a positive individual idiosyncrasy.

Second, individualists and collectivists, respectively, use idiocentric and sociocentric communication styles, and these behaviors are more likely to be enacted and exaggerated during intercultural interactions. Further, given the opportunity to adjust, participants from two different cultures still communicated with styles that were representative of their own cultures. The implication for individuals and managers in intercultural settings is that despite calls for communication accommodation and convergence strategies, in practice, people tend to behave and communicate in ways that reflect their cultural background. In sum, one cannot assume that members of different cultures will accommodate to communication and behavior patterns of host cultures, especially, in initial interactions. In fact, our findings suggest that the behavior that one encounters in initial intercultural interactions may be more culturally stereotypical than an individual’s style once he/she is familiar with the counterpart. In addition, it may take more time than two encounters for hosts and members of different cultures to accommodate their behaviors and communication styles.

Third, culturally dominant communication styles may be compounded by the fact that the need to reconcile the inner-self with external behaviors may be lower for collectivists than for individualists. The inconsistency-support behaviors exhibited by Asians in our sample suggest that outcomes of initial negotiation and business discussions may not reflect the true views of collectivist participants. The implication for individuals and managers in intercultural settings is that views expressed in public meetings or initial encounters may not be consistent subsequent views or views expressed privately in an attempt to maintain harmony.

Finally, managers should acknowledge that adaptation of behaviors to or in response to another culture is not a normal reaction. Adapting one’s behavior may not seem natural and might be attributed to insincere motives (Thomas and Ravlin, 1995), which in turn may negate any potential positive effects from the adaptation. It is crucial for managers to implement training to overcome workers’ normal tendencies to behave in their culturally dominant styles in different settings. Further, training and/or mentoring where there are opportunities for feedback is important to ensure that adaptive behaviors are effectively appropriate and not interpreted as patronizing and/or offensive in different multicultural settings.

Equation		Predictor	B	ΔF	R²	pΔR ²
Independence and horizontal individualism on argumentativeness
1	Independence	Horizontal individualism	0.584***	98.77	0.341	0.000***
2	Argumentativeness	Horizontal individualism	0.18**	6.38	0.032	0.012**
3	Argumentativeness	Horizontal individualism	0.055	6.34	0.063	0.002**
		Independence	0.214**
Equation		Predictor	B	ΔF	R²	pΔR ²
Interdependence and vertical-collectivism on argumentativeness
1	Interdependence	Vertical vollectivism	0.670***	155.99	0.450	0.000***
2	Argumentativeness	Vertical collectivism	–0.029	0.163	0.001	0.687
3	Argumentativeness	Vertical collectivism	0.121	2.785	0.028	0.064
		Interdependence	–0.224*

	Experimental condition
Expressiveness	8	8	6	10	13	3
Dominance	10	6	7	9	14	2
Initiating action	7	9	6	10	13	3
Aggressiveness	8	8	7	9	13	3
More logical argument	10	6	9	7	12	4
Regulating flow of interaction	11	5	4	12	11	5
Finishing task	11	5	7	9	11	5
Stronger opinion	11	5	8	8	13	3
Attempted eye contact	7	9	4	12	12	4
Accommodating	7	9	7	9	3	13
Avoidance of argument	7	9	7	9	5	11
More inclined to shift opinion	6	10	9	7	3	13

	Experimental condition
Expressiveness	10	6	10	6	13	3
Dominance	9	7	10	6	15	1
Initiating action	8	8	9	7	15	1
Aggressiveness	9	7	9	7	15	1
More logical argument	7	9	11	5	14	2
Regulating flow of interaction	9	7	10	6	14	2
Finishing task	7	9	8	8	13	3
Stronger opinion	10	6	12	4	15	1
Attempted eye contact	8	8	7	9	10	6
Accommodating	6	10	5	11	3	13
Avoidance of argument	7	9	5	11	3	13
More inclined to shift opinion	8	8	6	10	2	14

	Experimental condition
Expressiveness	1.79	1.7	1.91	1.93	3.42***	0.58***
Dominance	1.78	1.34	1.65	2.34	3.41***	0.27***
Initiating action	1.24	1.36	1.43	1.68	2.86***	0.66***
Aggressiveness	1.61	1.67	1.41	1.64	2.56***	0.53***
More logical argument	2.11	0.99	1.59	1.46	2.88***	0.56***
Regulating flow of interaction	1.62	1.03	0.88*	2.01*	2.48***	0.31***
Finishing task	1.3	1.08	1.19	1.15	1.65*	0.57*
Stronger opinion	1.98	1.53	2.23	1.7	3.23***	0.55***
Attempted eye contact	1.62	1.8	0.66**	2.58**	2.62***	0.61***
Accommodating	1.06	1.76	2.04	2.44	0.45***	3.14***
Avoidance of arguments	1.15	1.42	2.28	1.66	0.64**	2.29**
More inclined to shift opinion	1.14	1.74	2.09	1.77	0.47***	2.59***

	Experimental condition
Expressiveness	2.32	1.36	2.16	1.29	3.27***	0.55***
Dominance	2.03	1.61	2.84	1.49	4.13***	0.19***
Initiating action	1.4	1.11	2.19	1.2	3.38***	0.14***
Aggressiveness	1.76	1.3	2.12	1.23	2.89***	0.22***
More logical argument	1.46	1.49	2.44**	0.63**	3.38***	0.25***
Regulating flow of interaction	1.47	1.07	2.13*	1.02*	3.13***	0.47***
Finishing task	0.98	1.44	1.62	1.29	1.91**	0.44**
Stronger opinion	1.91	1.33	2.87**	1.01**	3.60***	0.23***
Attempted eye contact	1.54	1.3	1.53	1.54	1.71	1.27
Accommodating	1.32	2.03	1.41**	3.51**	0.50***	3.62***
Avoidance of arguments	1.33	1.49	1.38**	2.64**	0.36***	3.01***
More inclined to shift opinion	1.37	1.37	1.36*	2.71*	0.31***	3.12***

	Means CT1
Asian intracultural	33.44	30.69	0.33
NZ intracultural	17	23.43	3.93*
Intercultural	30.38	36.19	1.29
DIC	Means CT2
Participant A	Participant B	F (1, 30)
Asian intracultural	43.69	43.69	0
NZ intracultural	30.19	30.06	0.001
Intercultural	38.06	39.75	0.072

	CT1
Asian intracultural	β	0.72**	0.60*
SE β	0.22	0.26
		DIC A	DIC B
NZ intracultural	β	0.36	0.82**
SE β	0.28	0.23
		DIC NZ	DIC Asian
Intercultural	β	0.42*	0.43*
SE β	0.19	0.2
Experimental condition	CT2
		DIC A	DIC B
Asian intracultural	β	0.54*	0.64*
SE β	0.29	0.3
		DIC A	DIC B
NZ intracultural	β	0.62*	0.2
SE β	0.34	0.27
		DIC NZ	DIC Asian
Intercultural	β	0.3	0.39*
SE β	0.3	0.22

Footnotes

Acknowledgment

The authors are grateful to Dharm Bhawuk, Cindy Gallois, Gary Oddou, and Colleen Ward for comments on a previous version of this manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Aberson

Healy

Romero

(2000) Ingroup bias and self-esteem: a meta-analysis. Personality and Social Psychology Review 4:157–73.

Adair

Okumura

Brett

(2001) Negotiation behavior when cultures collide: the United States and Japan. Journal of Applied Psychology 86 (3):371–85.

Baron

Kenny

(1986) The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51 (6):1173–82.

Basil

(1996) Standpoint: the use of student samples in communication research. Journal of Broadcasting & Electronic Media 40 (3):431–40.

Bollen

(1989) Structural Equations with Latent Variables. New York: Wiley.

Bond

Leung

(2004) Culture-level dimensions of social axioms and their correlates across 41 cultures. Journal of Cross-Cultural Psychology 35 (5):548–70.

Boucher

Peng

Shi

(2009) Culture and implicit self-esteem Chinese are “good” and “bad” at the same time. Journal of Cross-Cultural Psychology 40 (1):24–45.

Burgoon

Hale

(1988) Nonverbal expectancy violations: model elaboration and application to immediacy behaviors. Communication Monographs 55:58–79.

Burgoon

Hale

(1987) Validation and measurement of the fundamental themes of relational communication. Communication Monographs 54: 19–41.

10.

Byrne

(1971) The Attraction Paradigm. New York: Academic Press.

11.

Cameron

(1986) Effectiveness as paradox: consensus and conflict in conceptions of organizational effectiveness. Management Science 32:539–53.

12.

Catrambone

Markus

(1987) The role of self-schemas in going beyond the information given. Social Cognition 5 (4):349–68.

13.

Chen

M-J

(2002) Transcending paradox: the ‘middle way’ perspective. Asia Pacific Journal of Management 19: 179–99.

14.

Chen

Y-C

(2001) Chinese values, health and nursing. Journal of Advanced Nursing 36(2):270–73.

15.

Cook

Campbell

(1979) Quasi-Experimentation: Design &Analysis Issues for Field Settings. Boston: Houghton.

16.

Cross

Morris

Gore

(2002) Thinking about oneself and others: the relational-interdependent self-construal and social cognition. Journal of Personality and Social Psychology 82 (3): 399–418.

17.

Cross

Bacon

Morris

(2000) The relational-interdependent self-construal and relationships. Journal of Personality and Social Psychology 78 (4):791–808.

18.

Doi

(1974) Some psychological themes in Japanese human relationships. In: Condon

Saito

(eds) Intercultural Encounters with Japan: Communication-Contact and Conflict. Tokyo: Simul Press, pp. 17–26.

19.

Fiske

Taylor

(1991) Social Cognition. New York: McGraw-Hill.

20.

Fletcher

GJO

Danilovics

Fernandez

(1986) Attributional complexity: an individual differences measure. Journal of Personality and Social Psychology 51 (4):875–84.

21.

Fletcher

Fang

(2006) Assessing the impact of culture on relationship creation and network formation in Asian emerging markets. European Journal of Marketing 40(3/4): 430–46.

22.

Francis

JNP

(1991) When in Rome? the effects of cultural adaptation on intercultural business negotiations. Journal of International Business Studies 22(3): 403–28.

23.

Gallois

Ogay

Giles

(2004) Communication accommodation theory: a look back and a look ahead. In: Gudykunst

(ed) Theorizing About Intercultural Communication. Thousand Oaks: Sage, pp. 121–48.

24.

Giles

Noels

(1997) Communication accommodation in intercultural encounters. In: Martin

Nakayama

Flores

(eds) Readings in Cultural Contexts. Mountain View: Mayfield, pp. 139–49.

25.

Glenn

(1981) Man and Mankind. Ablex: Norwood.

26.

Green

EGT

Deschamps

J-C

Paez

(2005) Variation of individualism and collectivism within and between 20 countries. Journal of Cross-Cultural Psychology 36 (3):321–39.

27.

Gudykunst

Ting-Toomey

Chua

(1988) Culture and Interpersonal Communication. Beverly Hills: Sage.

28.

Gudykunst

Matsumoto

Ting-Toomey

(1996) The influence of cultural individualism-collectivism, self-construals, and individual values on communication styles across cultures. Human Communication Research 22 (4):510–43.

29.

Hall

(1976) Beyond Culture. New York: Doubleday.

30.

Heider

(1946) Attitudes and cognitive organization. Journal of Psychology 21:107–12.

31.

Heine

Lehman

Peng

(2001) What’s wrong with cross-cultural comparisons of subjective Likert scales? The reference group effect. Journal ofPersonality and Social Psychology 82 (6): 903–18.

32.

Hinkle

Stiles

Taylor

(1988) Verbal processes in a labor/management negotiation. Journal of Language and Social Psychology 7 (2):123–36.

33.

Hofstede

(1980) Culture’s Consequences. Beverly Hills: Sage.

34.

Honeycutt

(1993) Components and functions of communication during initial interaction, with extrapolations to beyond. In: Deetz

(ed) Communication Yearbook. Newbury Park: Sage, pp. 461–90.

35.

Infante

Rancer

(1982) A conceptualization and measure of argumentativeness. Journal of Personality Assessment 46:72–80.

36.

Infante

Rancer

(1996) Argumentativeness and verbal aggressiveness: a review of recent theory and research. Communication Yearbook 19:319–51.

37.

Iwao

(1997) Consistency orientation and models of social behavior: is it not time for west to meet east? Japanese Psychological Research 39 (4): 323–32.

38.

Iwao

Triandis

(1993) Validity of auto- and heterostereotypes among Japanese and American students. Journal of Cross-Cultural Psychology 24 (4): 428–44.

39.

Kashima

Siegal

Tanaka

(1992) Do people believe behaviors are consistent with attitudes? Towards a cultural psychology of attributions. British Journal of Social Psychology 31: 111–24.

40.

Kelman

Baron

(1968a) Inconsistency as a psychological signal. In: Abelson

Aronson

McGuire

Rosenberg

Tannenbaum

(eds) Theories of Cognitive Consistency: A Sourcebook. Chicago: Rand McNally, pp. 331–36.

41.

Kelman

Baron

(1968b) Determinants of modes of resolving inconsistency dilemmas: a functional analysis. In Abelson

Aronson

McGuire

Rosenberg

Tannenbaum

(eds) Theories of Cognitive Consistency: A Sourcebook. Chicago: Rand McNally, pp. 670–83.

42.

Khokhlov

Gonzalez

(1973) Cross-cultural comparison of cognitive consistency. International Journal of Psychology 8 (2):137–45.

43.

Kim

M-S

Aune

Kim

H-J

(2001) The effect of culture and self-construals on predispositions toward verbal communication. Human Communication Research 27 (3):382–408.

44.

Kluckhohn

Strodbeck

(1961) Variations in Value Orientations. Evanston, Illinois: Harper & Row.

45.

Landis

Koch

(1977) The measurement of observer agreement for categorical data. Biometrics 33: 159–74.

46.

Laurent

(1983) The cultural diversity of western conceptions of management. International Studies of Management and Organization 13 (1): 75–96.

47.

Limaye

Victor

(1991) Cross-cultural business communication research: state of the art and hypotheses for the 1990’s. Journal of Business Communication 28 (3): 277–99.

48.

Lindsley

(1999) A layered model of problematic intercultural communication in U.S.-owned maquiladoras in Mexico. Communication Monographs 66 (2): 145–67.

49.

Lewis

(2000) Exploring paradox: toward a more comprehensive guide. Academy of Management Review 25: 760–76.

50.

Locke

(1986) Generalizing from laboratory to field: ecological validity or abstraction of essential elements? In: Locke

(ed) Generalizing from Laboratory to Field Settings. Lexington: D.C. Heath and Company, pp. 3–9.

51.

MacKenzie

Forde

(2009) The rhetoric of the 'good worker' versus the realities of employers' use and the experiences of migrant workers. Work Employment Society 23:142–59.

52.

Markus

Kitayama

(1991) Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review 98 (2), pp. 224–53.

53.

Matsumoto

(1999) Culture and self: an empirical assessment of Markus and Kitayama’s theory of independent and interdependent self-construals. Asian Journal of Social Psychology 2: 289–310.

54.

Neff

Karney

(2005) To know you is to love you: the implications of global adoration and specific accuracy for marital relationships. Journal of Personality and Social Personality 88: 480–97.

55.

Nisbett

Peng

Choi

(2001) Culture and systems of thought: holistic versus analytic cognition. Psychological Review 108 (2): 291–310.

56.

Oishi

Diener

Scollon

(2004) Cross-situational consistency of affective experiences across cultures. Journal of Personality and Social Psychology 86 (3): 460–72.

57.

Okoro

Washington

(2012) Workforce diversity and organizational communication: analysis of human capital performance and productivity. Journal of Diversity Management 7 (1): 57–62.

58.

Paletz

SBF

Peng

(2009) Problem finding and contradiction: examining the relationship between naïve dialectical thinking ethnicity and creativity. Creativity Research Journal 21 (2–3): 139–51.

59.

Park

Levine

(1999) The theory of reasoned action and self-construal: evidence from three cultures. Communication Monographs 66 (3): 199–218.

60.

Pekerti

Kwantes

(2011) The effect of self-construals on perceptions of organizational events. International Journal of Cross-Cultural Management 19(1): 1–21.

61.

Pekerti

Thomas

(2003) Communication in intercultural interaction: an empirical investigation of idiocentric and sociocentric communication styles. Journal of Cross-Cultural Psychology 34 (2): 139–54.

62.

Peng

Nisbett

(1999) Culture, dialectics, and reasoning about contradiction. American Psychologist 54: 741–54.

63.

Peng

Spencer-Rodgers

Zhong

(2006) Naïve dialecticism and the Tao of Chinese thought. In: Kim

Yang

K-S

Hwang

K-K

(eds) Indigenous and Cultural Psychology:Understanding People in Context. New York: Springer, pp. 247–62.

64.

Preacher

Hayes

(2004) SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments & Computers 36 (4): 717–31.

65.

Preacher

Rucker

Hayes

(2007) Addressing moderated mediation hypotheses: theory, methods, and prescriptions. Multivariate Behavioral Research 42 (1): 185–227.

66.

Rao

Hashimoto

(1996) Intercultural influence: a study of Japanese expatriate managers in Canada. Journal of International Business Studies 27 (3): 443–66.

67.

Rokeach

(1973) The Nature of Human Values. New York: Free Press.

68.

Rosenthal

(2002) The pygmalion effect and its mediating mechanisms. In: Aronson

(ed) Improving Academic Achievement. San Diego: Academic Press, pp. 25–36.

69.

Sanchez-Burks

Lee

Choi

(2003) Conversing across cultures: east-west communication styles in work and non-work contexts. Journal of Personality and Social Psychology 85 (2): 363–72.

70.

Schwartz

(1994) Beyond individualism/collectivism: new dimensions of values. In Kim

Triandis

Kagitçibasi

Choi

Yoon

(eds) Individualism and Collectivism: Theory, Applications, and Methods. Thousand Oaks: Sage, pp. 85–119.

71.

Servaes

(1989) Cultural identity and modes of communication. In: Anderson

(ed) Communication Yearbook. Newbury Park: Sage, pp. 383–416.

72.

Singelis

(1994) The measurement of independent and interdependent self-construals. Personality and Social Psychology Bulletin 20: 580–91.

73.

Singelis

Brown

(1995) Culture, self, and collectivist communication. Human Communication Research 22 (3): 354–89.

74.

Smith

Bond

(1999) Social Psychology Across Cultures. Boston: Allyn & Bacon.

75.

Spencer-Rodgers

Williams

Peng

(2010) Cultural differences in expectations of change and tolerance for contradiction: a decade of empirical research. Personality and Social Psychology Review 14 (3): 296–312.

76.

Suh

(2002) Culture, identity, consistency, and subjective well-being. Journal of Personality and Social Psychology.83 (6):1378–91.

77.

Suzuki

Rancer

(1994) Argumentativeness and verbal aggressiveness: Testing for conceptual and measurement equivalence across cultures. Communication Monographs 61: 256–79.

78.

Taylor

(1987/88) Crimes, death and stress: three new consensus tasks. Organizational Teaching Review 12: 115–7.

79.

Thomas

Stahl

Ravlin

(2008) International Journal of Cross Cultural Management 8(2): 123–43.

80.

Thomas

Ravlin

(1995) Responses of employees to cultural adaptation by a foreign manager. Journal of Applied Psychology 80 (1): 133–46.

81.

Triandis

(1989) The self and social behavior in differing cultural context. Psychological Review 96(3):506–20.

82.

Triandis

(1995) Individualism and Collectivism. Boulder: Westview.

83.

Triandis

Gelfand

(1998) Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology 74 (1): 118–28.

84.

Triandis

Lisansky

Marín

(1984) Simpatiá as a cultural script of Hispanics. Journal of Personality and Social Psychology 47 (96): 1363–75.

85.

Triandis

McCusker

Betancourt

(1993) An etic-emic analysis of individualism and collectivism. Journal of Cross-Cultural Psychology 24 (3): 366–83.

86.

Tse

Francis

Walls

(1994) Cultural differences in conducting intra- and inter-cultural negotiations: a Sino-Canadian comparison. Journal of International Business Studies 25 (3): 537–55.

87.

van de Vijver

Leung

(1997) Methods and Data Analysis for Cross-Cultural Research. Thousand Oaks: Sage.

88.

Yamada

A-M

Singelis

(1999) Biculturalism and self-construal. International Journal of Intercultural Relations 23(5):697–709.

	Experimental condition
Individual behavior	Intercultural		NZ intracultural		Asian intracultural
Individual behavior	NZ	NZ	Asian	Asian	NZ	Asian
Expressiveness	8	8	6	10	13	3
Dominance	10	6	7	9	14	2
Initiating action	7	9	6	10	13	3
Aggressiveness	8	8	7	9	13	3
More logical argument	10	6	9	7	12	4
Regulating flow of interaction	11	5	4	12	11	5
Finishing task	11	5	7	9	11	5
Stronger opinion	11	5	8	8	13	3
Attempted eye contact	7	9	4	12	12	4
Accommodating	7	9	7	9	3	13
Avoidance of argument	7	9	7	9	5	11
More inclined to shift opinion	6	10	9	7	3	13

	Experimental condition
	NZ-Intracultural		Asian-Intracultural		Intercultural
Individual behavior	NZ	NZ	Asian	Asian	NZ	Asian
Expressiveness	10	6	10	6	13	3
Dominance	9	7	10	6	15	1
Initiating action	8	8	9	7	15	1
Aggressiveness	9	7	9	7	15	1
More logical argument	7	9	11	5	14	2
Regulating flow of interaction	9	7	10	6	14	2
Finishing task	7	9	8	8	13	3
Stronger opinion	10	6	12	4	15	1
Attempted eye contact	8	8	7	9	10	6
Accommodating	6	10	5	11	3	13
Avoidance of argument	7	9	5	11	3	13
More inclined to shift opinion	8	8	6	10	2	14

	Means CT1
DIC	Participant A	Participant B	F(1, 30)
Asian intracultural	33.44	30.69	0.33
NZ intracultural	17	23.43	3.93*
Intercultural	30.38	36.19	1.29
DIC	Means CT2
DIC	Participant A	Participant B	F (1, 30)
Asian intracultural	43.69	43.69	0
NZ intracultural	30.19	30.06	0.001
Intercultural	38.06	39.75	0.072