ARTIGOS

 

A study of values as cross-cultural competencies: a potential source of self-esteem?

 

Um estudo de valores como competências interculturais: uma fonte potencial de auto estima?

 

Un estudio de valores como competencias interculturales: ¿una fuente potencial de autoestima?

 

 

Chitra Raghavan

City University of New York

Endereço para correspondência

 

 

---

ABSTRACT

The goal of this study was to examine how individuals evaluate themselves and their ingroup on a series of values that vary in cultural importance across Asian American and North American cultural groups. Specifically, we examined cross-cultural differences in mean levels of culturally-based competencies, and explored whether there are also cross-cultural differences in which facets of these competencies are associated with positive affect. Participants were 85 Asian American (AA) and 83 European American (EA) college students. Group competency evaluation was more pertinent to AA than EA, whereas independent competency evaluation was more pertinent to EA than AA, both in terms of mean competency levels and the prediction of positive affect. Implications for counselors are discussed.The goal of this study was to examine how individuals evaluate themselves and their ingroup on a series of values that vary in cultural importance across Asian American and North American cultural groups. Specifically, we examined cross-cultural differences in mean levels of culturally-based competencies, and explored whether there are also cross-cultural differences in which facets of these competencies are associated with positive affect. Participants were 85 Asian American (AA) and 83 European American (EA) college students. Group competency evaluation was more pertinent to AA than EA, whereas independent competency evaluation was more pertinent to EA than AA, both in terms of mean competency levels and the prediction of positive affect. Implications for counselors are discussed.

Keywords: Self-esteem, Cultural competencies, Positive affect.

---

RESUMO

O objetivo deste estudo foi examinar como indivíduos avaliam a si mesmos e as ingroup em uma série de valores que variavam em importância cultural entre grupos de asiáticas-americanas e européias-americanas. Especificamente foram examinadas diferenças interculturais nos níveis médios de competências baseadas culturalmente, e explorou-se também se haviam diferenças interculturais nas quais facetas destas competências estavam associadas a afetos positivos. Participaram do estudo 85 estudantes universitárias asiáticas– americanas (AA) e 83 européias-americanas (EA). A avaliação de competência grupal foi mais pertinente ao grupo de AA do que ao grupo EA, ao passo que a avaliação de competência independente foi mais pertinente ao grupo de EA do que ao AA. Isto ocorreu tanto em nível de competência média como na predição de afetos positivos. Implicações para conselheiros são discutidas.

Palavras-chave: Auto-estima, Competência cultural, Afeto positivo.

---

RESUMEN

El objetivo de este estudio fue examinar como indivíduos evaluan a sí mismos y sus ingroup en una serie de valores que variaban en importancia cultural entre grupos de asiáticas-americanas y europeas-americanas. Especificamente nosotros examinamos diferencias interculturales en los niveles medios de competencias basados culturalmente y se exploraron también si habían diferencias interculturales en las cuales facetas de estas competencias estaban asociadas a afectos positivos. Participaron del estudio 85 estudiantes universitarias asiáticas–americanas (AA) y 83 europeas-americanas (EA). La evaluación de competencia grupal fue más pertinente al grupo de AA que al grupo EA, al mismo tiempo que la evaluación de competencia independiente fue más pertinente al grupo de EA que al grupo de AA. Esto ocurrió tanto al nivel de competencia media como en la predicción de afectos positivos. Son discutidas implicaciones para consejeros.

Palabras clave: Autoestima, Competencia cultural, Afecto positivo

---

 

 

Introduction

The concept of self-esteem, self-worth, self-regard and related concepts has received considerable attention in the last few decades. Nevertheless, theorists are divided on not only the definition of self-esteem, but also over processes can constitute self-esteem and its benefits or lack thereof. Certain theorists take the position that self-esteem refers to an individual’s global evaluation of their worth (HEINE; LEHMAN; MARKUS; KITAYAMA, 1999; ROSENBERG, 1965). Others argue that self-esteem is domain specific and is acquired by evaluations of different areas of the self (e.g., FLEMING; COURTNEY, 1984; PELHAM, 1995). Yet others address the controversial issue of whether self-esteem is Western- centric or if it can be redefined to be culturally relevant (e.g., CROCKER; LUHTANEN; BLAINE; BROADNAX, 1994; HEINE et al., 1999).

Solomon, Greenberg, and Pyszczynski (1991) were among the first modern theorists to reiterate that self-esteem is derived from fulfilling culturally prescribed tasks and norms that reflect desirable cultural competencies (MARKUS; KITAYAMA, 1991). This view suggests that self-esteem can be universally meaningful, but only if the specific goals and tasks are appropriately defined in cultural reality. In this context, several studies have highlighted the relevance of social group membership when examining self-esteem and related concepts (CROCKER et al., 1994; LUHTANEN; CROCKER, 1992; SINGELIS, 1994; TRIANDIS, 1989). Specifically, whereas the independent or individual self is organized with a primary focus on the individual, the interdependent or collective self stresses the self-in-relationship-to-others (MARKUS; KITAYAMA, 1991, p. 224) which may consist of relationships with important ingroups defined by kinship, race, and religion (TRIANDIS, 1989; 1995). This view of self-esteem suggests that for certain groups, what we evaluate and ultimately how we derive our positive self-regard is intimately linked with the content, structure, and psychological function of the shared view of the self (MARKUS; KITAYAMA, 1991; TRIANDIS, 1989). In this vein, Higgins (1996) echoing James (1890) advances the view that the same individual can have high, low, and moderate levels of different kinds of esteem depending on which selves and which reference points are important (p. 1.074).

Issues pertinent to multiple selves and esteem have been widely explored in East Asian and Asian American populations with differing interpretations. One body of research has found that Asian Americans (AA) consistently report lower levels of global self-esteem than do European Americans (EA) (CROCKER et al., 1994; FEATHER; MCKEE, 1996). Earlier explanations for these levels of esteem focused on Asian Americans reporting actual low esteem as a result of stigmatization and immigrant stress. More recent explanations (e.g., LUHTANEN; CROCKER, 1992) have highlighted the importance of measuring esteem that is derived from the collective (TRIANDIS, 1989) or interdependent self (MARKUS; KITAYAMA, 1991) that is suggested to be equally if not more salient for certain minority groups. For example, Crocker and Luhtanen (1992) argued that for certain groups of people, individuals may strive to not only maintain a positive self-regard but also a positive group identity and that such groups would report high levels of collective identity. Accordingly, they measured four types of collective esteem pertinent to group membership across African-American, Asian, and White students. However, contrary to their prediction, their results indicated that Whites reported higher total collective self-esteem scores than either of the two minority groups. The authors suggested that these results may be related to the fact that participants were not asked to specifically identify ethnic groups and may have used other reference groups such as gender or religious affiliation when completing the measure. Recently, Ahlering (2003) using the same measure as Crocker and colleagues reported that although Asian and Latino groups showed higher scores on a particular collective subscale (the Identity subscale) than did the White and African-American groups, they did not report higher overall collective esteem. Thus, the results pertinent to global collective self-esteem are inconclusive.

Heine et al. (1999) took a different approach altogether to explain the lower self-esteem scores in Asians and Asian Americans compared to North Americans. The authors argue that self-esteem is most likely applicable only to North American populations and is not universal. To support their argument, they suggest that self-esteem is best defined as a global positive self-view of the individual self. Furthermore, in cultures where the interdependent self is more relevant than the independent self, there is limited need for the self-affirming positive self. Instead, the affirmation of the interdependent self is maintained by a critical view of the self which is used to constantly monitor and adjust the self to maintain relational harmony. Although the idea proposed by Heine et al. (1999) is intriguing, and a radical departure from previous explanations, research indicates that both self-esteem and collective esteem predict Asian and Asian American well-being suggesting that both types of esteem are implicated in these cultural groups’ well-being. For example, Kwan, Bond, and Singelis (1997) reported that both relationship harmony and traditional self-esteem were important in predicting life satisfaction in a Hong Kong sample (KWAN; BOND; SINGELIS, 1997), a finding similarly reported by Benet-Martinez and Karakitapoglu-Aygun (2003) in an Asian American sample.

The most convincing evidence that Asians do self-enhance (view themselves positively) comes from a series of studies conducted by Sedikides, Gaertner, and Toguchi (2003). The authors demonstrated that both Japanese and American students self-enhanced when the focus of enhancement was congruent to their self-construals. Because selfenhancement is an intrinsic part of the esteeming process, these authors concluded that self-esteem remains universally important but that in collectivist cultures is acquired by excelling in group-centered rather than self-centered behaviors. Similar results on selfenhancement have been reported by researchers working with Asian groups (e.g., YIK; BOND; PAULHUS, 1998). Tsai, Ying, and Lee (2001) suggest that Asian versus Western social contexts may influence whether self-enhancement or self-effacement is triggered in Asian Americans. Finally, Bae and Brekke (2003) suggest that some of the reported differences between Asian Americans and North Americans may arise from item response bias rather than self-effacement; the authors found that Korean Americans compared to Caucasian, African American, and Latino schizophrenic patients reported significantly lower self-esteem scores on positively but not negatively worded items.

In sum, multiple viewpoints have been offered to explain the definition and utility of self-esteem across cultures, with some theorists favoring the idea that self-esteem is multifaceted and linked to multiple selves and domains and others defining self-esteem narrowly and rejecting its utility in collectivist cultures. Support for the unique importance of collective esteem remains equivocal; however evidence for the need for a positive self enhanced view of culturally valid competencies is compelling. These results suggest that self-regard, self-worth, or self-esteem among the many numerous terms used to describe related concepts should be applied to different competencies and values across different cultures in order to enrich our understanding. Accordingly, the goal of the present study was to explore ways in which individuals from two different cultural backgrounds differ across culturally specific values when evaluating themselves and an important ingroup (SEDIKIDES et al., 2003). We adopt the view that self-worth can be measured by the degree to which individuals believe that they or an important ingroup( s) possesses culturally valued traits and thus desirable attributes or competencies (e.g., SOLOMON et al., 1991; SEDIKIDES et al., 2003). We focus on interdependent and independent domains as current research supports the distinctions between these domains and suggest that they are valued differently across cultures. Interdependent values promote harmonious relationships by adjusting oneself within such relationships and by discouraging competition among group members; examples of such values are politeness and devotion (SCHWARTZ, 1994; TRIANDIS, 1995). In contrast, independent competencies are those that promote the independent self and are competitive in nature (SCHWARTZ, 1994; TRIANDIS, 1995). The affirmation of such competencies, whether independent or interdependent, contributes to positive self-enhancement, self-worth or self-esteem. For example, in some cultures, politeness is a great competency (GEGEO; GEGEO, 1990) whereas in North America self-directed achievement is valued over smooth social interactions. Thus, in the first culture, “I am a polite person” would affirm one is worthy whereas ‘I am a go-getter’ would better indicate self-esteem in the latter.

Our overarching goal was to begin exploring whether the positive endorsement of culturally-specific values may in fact be another form of positive self-regard, and thus function in ways similar to esteem. We were particularly interested in the evaluation of specific values tied to identified ingroups because much of the cross-cultural esteem research has focused on global evaluations of assumed ethnic ingroups. In addition, we were interested in how evaluation of values tied to both the self and ingroup would compare to more traditional global notions of self-esteem in the prediction of positive affect across different cultural groups. According to this formulation, the evaluation of the self and the ingroup on independent and interdependent competencies results in four types of evaluation: (a) self-independent (self rates self on independent competencies); (b) self-interdependent (self rates self on interdependent competencies); (c) self-group independent (self rates group on independent competencies); and (d) selfgroup interdependent (self rates group on interdependent competencies).

Accordingly, the first goal of this research was to replicate self-enhancement processes in AA and EA when evaluating culturally-relevant competencies. We predicted that cultural groups would have the highest levels of competency evaluation (i.e., would enhance) on those facets that most closely match their orientation. Thus, we predicted that compared to AA, EA would evaluate themselves higher on independent competencies. We next predicted that AA would self-enhance on not only interdependent competencies, but that this tendency would be most pronounced when they rated their ingroup, specifically, self-group interdependent. We also expected that EA would report higher scores than AA on traditional global measures of esteem and collective esteem as has been typically found. In contrast, similar to past research, AA would not self-enhance when asked to evaluate their ingroup globally. In sum, our first goal was to examine whether cross-cultural differences in competency evaluation are moderated by the target (i.e., whether the self or the ingroup is being considered) and/or by the type of competency (i.e., whether independent or interdependent attributes are being considered).

The second goal of the research was to examine whether cultural group membership would moderate the relation between competency evaluation and positive affect. Research indicates that higher levels of global esteem tend to be associated with higher positive affect (e.g., CHENG; FURNHAM, 2003; WOOD; HEIMPEL; MICHELA, 2003). Accordingly, we predicted that the association between positive affect and self-independent competency evaluations would be stronger among EA than among AA, whereas the association between positive affect and self-group interdependent competency evaluations would be stronger among AA than among EA. Finally, we explored the contribution of competency evaluation in predicting positive affect once global esteem had been accounted for.

 

Method

Participants

Participants were 85 AA (53% women) and 83 EA (59% women) who received participation credit in a psychology course. The four largest ethnic groups in the AA sample were Indian (24.7%), Korean (22.4%), Filipino (21.2%), and Chinese (17.6%). Seven Asian participants (8.2%) did not identify specific ethnicities. About half the EA sample identified mixed european ethnicity (49.4%). The four largest ethnic groups (ranging from 6 to 8.4%) were Irish, Greek, Polish, and German with individual participants endorsing a wide variety of other ethnicities. Seven EA participants (8.4%) did not identify specific ethnicities.

Comparisons of the four largest ethnic groups within the AA sample revealed nonsignificant differences in their mean levels of esteem, positive affect and years in the U.S. Consequently, the analyses described below were performed using the two main cultural group members (AA, EA). The two groups ranged in age from 16 to 28 years (M =18.9, SD = 1.5) and did not differ significantly in age or gender. AA participants were asked to complete several items pertaining to their cultural identity. Fifty percent of the AA group had moved to the U.S. at or before age 12. About one third (38.8%) reported English being their second language, 41.2% reported that it was their first language, and 20% did not answer this question. AA were asked to rate themselves on what they felt their primary values to be (1=Highly Asian to 5 = Highly Western). AA indicated that the average value fell between mostly Asian and equally Western and Asian (M = 2.9, SD = .9). In sum, the AA population indicated a moderate level of acculturation to US culture with the majority indicating comfort with both Asian and US cultures. Participants were informed that the researchers were interested in exploring general identity, esteem, and positive affect.

Procedure and Measures

Participants completed several questionnaires that obtained demographic information, acculturation (if participant was AA), mood, two established measures of esteem and a new measure of competency evaluation.

Participant Ingroup and Individualism-Collectivism

Participants were first requested to identify an ingroup that “You identify with the most” and that is “most important to you.” Asian American participants were significantly more likely to identify groups defined by ethnic and mixed-ethnic religious background as their primary ingroup, whereas European American participants identified groups related to interests and activities as their primary ingroup, ?2(2, nAA= 79, nEA = 79) = 16.7, p < .001. Sample EA groups include “my sports group” whereas sample AA groups included “my Korean American friends.”

Participants completed an abbreviated measure that assessed individualism and collectivism (TRIANDIS; GELFAND, 1998; TRIANDIS, personal communication, 1996). Participants were presented with 16 scenarios each with 4 responses of which 2 were individualist and 2 collectivist. Participants ranked the top two responses for each scenario and responses were summed to create an overall individualism and collectivism score. No differences were observed between the two groups on either individualistic (AA: M = 12.9, SD= 2.5, EA: M = 12.8, SD = 2 .5) or collectivistic responses (AA: M = 1.8, SD= 2.4, EA: M = 1.7, SD = 2.5).

Competency evaluation

Participants were first asked to indicate how important various competencies were to them (1 = not all to 5 = extremely important). Subsequently, they completed the Competency Evaluation Questionnaire (CEQ) keeping in mind the ingroup they had identified. The CEQ is a self-report measure designed for this study that comprises 4 subscales. The four competency evaluation subscales were: (a) Self Interdependent Competency (Self-INT), which measured how descriptive the participant thought the interdependent competencies were of himself/herself (e.g., “I think that I am polite”); (b) Self Independent Competency (Self-IND), which measured how descriptive the participant thought the independent competencies were of himself/herself (e.g., “I think that I am ambitious”); (c) Self-Group Interdependent Competency (SelfGrp-INT), which measured how descriptive the participant thought the interdependent competencies were of the ingroup (e.g., “I think that my group is polite”); and (d) Self-Group Independent Competency (SelfGrp-IND), which measured how descriptive the participant thought the independent competencies were of the ingroup (e.g., “I think that my group is ambitious”).

The items that composed the four competency evaluation subscales were selected from a list of words generated by Schwartz (SCHWARTZ, 1994; SCHWARTZ; BILSKY, 1990) with reference to other literatures (MARKUS; KITAYAMA, 1991; TRIANDIS, 1989). We relied on Schwartz initially because broad support for the universal content and structure of values across cultures has been reported (SCHWARTZ; BILSKY, 1990). We selected items that demonstrated good convergent and discriminant validity across cultures and that fell consistently in either individual-related or collective/prosocial related categories. From this list, six graduate students and one faculty member rated attributes as a either individualistic or collectivistic competencies. Competencies that were rated by the majority of the raters (no more than 2 dissenting) as falling into one of the two categories were retained. Independent items chosen reflected the value types of achievement, stimulation, and self-direction which are associated with individualism (OISHI; SCHIMMACK; DIENER, 1998) and are as follows: “ambitious”, “independent” “creative”, “daring”, and “influential.” One other item “assertive” was drawn from other literatures (MARKUS; KITAYMA, 1991; TRIANDIS, 1989). Interdependent items reflect values of tradition, benevolence, and conformity which are all associated with collectivism (OISHI et al., 1998) and are as follows: “forgiving”, “helpful”, “obedient”, “polite”, “humble”, and “devout.” A sixth item “interdependent” was added from other literatures (MARKUS; KITAYMA, 1991; TRIANDIS, 1989). All items were rated as being descriptive of the respondent or the ingroup on a scale ranging from 1 (not at all) to 5 (extremely); thus, higher scores indicated higher competency evaluation. The internal consistencies for all four subscales were good for the combined cultural groups (r = .80 to .94).

Participants also completed the Collective Self-Esteem Scale (CSES) (LUHTANEN; CROCKER, 1992) which is typically used to measure global collective esteem. The CSES consists of 4 subscales with a total of 16 items and measures esteem that is derived from membership with a social group. The four subscales are Private Esteem (private cognitions about the ingroup), Public Esteem (other’s evaluation of the group), Identity (the importance of the group to one’s identity), and Membership (feelings about one’s membership in the group). Participants were asked to rate the extent to which they agreed or disagreed with each statement using a 7-point scale. Internal consistencies of the CSES subscales were acceptable (r = .77 to .83). Finally, the Rosenberg Self-Esteem Scale (RSE) (ROSENBERG, 1965), a widely used measure of self-esteem with established validity and reliability, was used to assess global self-focused esteem. The internal consistency of the RSE in this study was high (r = .89). Both The RSE and the CSES were included for comparison purposes with the new competency scale.

Positive Affect

Participants completed a subset of items from the Positive and Negative Affect Schedule- Expanded Form (PANAS-X) (WATSON; CLARK, 1991; WATSON; CLARK; TELLEGEN, 1988) to measure mood. We measured Positive Affect (PA) using a 10-item scale (WATSON et al., 1988). PA refers to one’s level of engagement with the environment (WATSON et al., 1988). Items on the PA scale include “attentive” and “inspired.” In addition, we included three additional items from the JOY subscale taken from the PANAS-X. These items were “happy”, “joyful”, and “delighted.” The PA and JOY scales were strongly correlated (r = .87, p < .001) and were combined to form a single scale. Internal consistency for the combined scale was high (r = .90).

 

Results

Correlations between Competency Evaluation and Global Esteem Scales

We first examined whether the competency evaluation subscales would demonstrate preliminary convergent and discriminant validity with respect to individualism-collectivism and esteem measures. We thus examined the inter-correlations among the competency evaluation subscales, the associations of the CEQ with global esteem measures, the RSE, and the CSES, and the measure of individualism-collectivism.

Table 1. Correlations Among Esteem Subscales and with CSES, the RSE, and Individualism-Collectivism

 

 

Of the four competency subscales, (Table 1) the independent-related subscales tended to be more strongly associated with each other than with the interdependent-related subscales. In contrast, the interdependent-related subscales tended to be more strongly associated with each other than with the independent-related subscales. The pattern of correlations among the four CEQ scales, the Rosenberg, and the CSES also gave a preliminary indication that the CEQ scales were a good measure of the purported underlying constructs. For example, of the four competency scales, SelfGrp-INT, theoretically hypothesized to be the most interdependent in nature, was unrelated to the RSE, whereas the other three scales showed a large to modest relationship. Additionally, SelfGrpINT was unrelated to Member, which is the most individualistic of the CSES subscales, measuring the extent to which an individual feels she is a good group member (LUHTANEN; CROCKER, 1992). Self-IND, theoretically hypothesized to be the most independent in nature, had a large association with the RSE and relatively weaker associations with the other CSES subscales. Finally, the CEQ subscales showed a theoretically consistent pattern of associations with the individualism-collectivism scenario scales. Overall, the patterns of correlations are consistent with what we would expect from scales purporting to measure positive self-regard derived from evaluating independent and interdependent competencies.

Mean Differences

Mean scores for the non CEQ subscales were compared across the two cultural groups using an Analysis of Variance (ANOVA). First, we examined potential cultural group differences in competency importance scores. As expected, AA (M=3.7, SD=.7) rated interdependent competencies as being more important than did EA (M =3.5; SD =.7), F(1,166) = 5.55, p < .05. However, both groups rated independent competencies as being equally important (AA: M = 3.6, SD = .6; EA: M = 3.7, SD =.6). We next examined whether the typical findings of past research of EA having higher levels of global self-esteem, collective esteem and positive affect than AA would be replicated in our sample. As expected, EA had significantly higher scores (M= 3.2, SD=4) than did AA (M = 2.9, SD=.5), F(1,164)=16.25, p <.001 on the RSE. However, no significant differences were observed in any of the CSES subscales (AA: M = 5.4, SD = .8; EA: M = 5.6, SD = .8) or on the overall mean CSES scores. Finally, as expected, EA had significantly higher scores (M= 3.6, SD= .5) than did AA (M=3.4, SD= .7), F(1,164) = 16.25, p < .001 on PA/JOY.

We then tested our hypothesis that group differences in competency evaluation would be moderated by the target and type of attribute or competency (i.e., independent vs. interdependent) being examined. To accomplish this goal, we conducted a culture (AA vs. EA) x target (a within-subject variable with two levels: self vs.ingroup) x attribute=competency (a within-subject variable with two levels: independent vs. interdependent) repeated measures MANOVA with the following 4 dependent variables: Self-IND, Self-INT, SelfGrp-IND, and SelfGrp-INT. Each of the two groups’ standardized mean scores is presented in Figure 1. In addition to a significant main effect for target (with the self being rated more highly than the ingroup), F(1,165) =6.04, p < .05, there was a statistically significant culture x attribute interaction, F(1,165) = 6.21 p < .05. There were also trends for a target x attribute interaction, F(1,165)= 2.86, p < .1, and a threeway, target x attribute x cultural group interaction, F(1,165) = 2.81, p < .1.

The interactions indicated that there were significant effects for culture but that they were moderated by what was being rated and to a lesser extent who was being rated. AA rated interdependent competency evaluation higher than independent competency evaluation. The opposite effect was found for EA. This attribute x culture interaction can be seen quite clearly in the competency profiles depicted in Figure 1. This two-way interaction was further modified by a target x attribute x culture interaction. Although statistically marginal, the proportion of variance (eta square = .02) obtained is an important effect size for non-experimental studies (MCCLELLAND; JUDD, 1993). This three-way interaction suggests that cultural group differences in competency evaluations are weakly moderated by both the target and competency being considered. More specifically, AA obtained the highest competency ratings when considering both the ingroup and interdependent attributes, and the lowest when considering the self and independent attributes, findings which were reversed in EA.

 

 

Figure 1. Profile analysis of esteem scales for AA and EA. Mean scores standardized within each esteem scale are presented on the y-axis. Raw mean scores and standard deviations are presented in boxes for each point. Points represent the group mean score for each scale. Self-IND = Self Independent Esteem; SelfGrp-IND = Self-Group Independent Esteem; Self-INT = Self Interdependent Esteem; SelfGrp-INT = Self-Group Interdependent Esteem.

 

Associations with Positive Affect

We conducted a hierarchical multiple regression analyses to examine how different competency scores and esteem scores predict PA/JOY. We entered mean scores for the importance of independent competencies and interdependent competencies, and the number of years participants had lived in the U.S., the RSE and the CSES in the first step. The two centered global esteem scores, RSE and the CSES were included to examine if competency evaluation predicted PA/JOY above and beyond the contributions of global esteem scores. The importance scores were included because interdependent competencies were differentially important between the two cultural groups, and number of years in the U.S. was included as a brief reference of acculturation.

We then entered the four centered competency variables (Self-IND, Self-INT, SelfGrp- INT, and SelfGrp-IND) in the second step. The two-way interactions between culture and each of the centered competency variables (e.g., Self-IND x culture, Self-INT x culture) were entered in the third and final step.

Table 2. Models of Esteem across EA and AA Predicting Positive Affect

 

 

As can be seen in Table 2, RSE, CSES, and importance of independent competencies significantly predicted PA/Joy. However, in the final model, CSES was no longer statistically significant. Of the competency evaluation scores, SelfGrp-IND was not associated with PA/JOY and main effects for Self-IND and SelfGrp-INT were qualified by two-way interactions. Thus, as expected, the interaction between culture and competencies in the prediction of PA/Joy was statistically significant and explained 5 % of the variance F (4, 156) = 3.30, p < .05. The overall model explained 50% of the variance. Of the interactions between culture and the four competency scores, only Self-INT did not contribute to PA/JOY.

The three culture x competency interactions that predicted PA/JOY are depicted in Figure 2a-2c respectively following the procedure recommended by Aiken and West (1991). Higher levels of Self-IND were associated with higher levels of PA/JOY among EA but not among AA, whereas higher levels of SelfGrp-INT and SelfGrp-IND were associated with higher levels of PA/JOY among AA but not among EA.

 

 

Figure 2a. Interactions of three Esteem x Culture variables. Self-IND x Culture. Y-axis represents Positive Affect (PA/JOY). Self-IND = Self Independent Esteem.

 

 

Figure 2b. Interactions of three Esteem x Culture variables. SelfGrp-IND x Culture . Y-axis represents Positive Affect (PA/JOY). SelfGrp-IND = Self-Group Independent Esteem.

 

 

Figure 2c. Interactions of three Esteem x Culture variables. SelfGrp-INT x culture. Y-axis represents Positive Affect (PA/JOY). Self-INT = Self Interdependent Esteem; SelfGrp-INT = Self-Group Interdependent Esteem.

 

Discussion

In this study, we wished to explore how culturally-sanctioned values may function as one of the multiple sources of self-worth, self-regard, or self-esteem. Our interest in doing so was derived from numerous esteem and personality researchers who have exhorted the need to measure culturally-valued competencies rather than global universal competencies and define the target of evaluation as both the ingroup and the self (e.g., CROCKER et al., 1994; HIGGINS, 1996; SEDIKIDES et al., 2003). Nonetheless, it is typical for researchers interested in esteem to ask respondents for global assessments of themselves and their ingroup (e.g., how well or badly do you feel about yourself or your group). One potential problem with global assessments of collective esteem is that the language inherent in such evaluations may still reflect North American assumptions of general success (i.e., I am proud of my group’s achievements). In contrast, we asked respondents to indicate the degree to which the self and the ingroup possess a variety of independent and interdependent values where we linked each respondent’s selfidentified ingroup specifically to the values being measured. We approached this study based on the assumption that if one source of positive self-regard or esteem is derived from fulfilling culturally based competencies, measuring these competencies would provide us with information about one’s self-worth that should vary across cultural group members. We found that competency evaluations functioned similarly to what we would expect from culturally-variable positive regard in three important ways. In terms of selfenhancement, AA and Ea each self-enhanced on predicted competencies which were implicated in the respective group’s positive affect. Further support for these competency evaluations as a possible source of esteem were indicated by its relationship to global esteem: these evaluations contributed to positive affect above and beyond that of global esteem and appear to have a stronger contribution to positive affect than collective esteem altogether in this sample. Although more research is needed before we can claim that endorsing culturally-valid values and competencies may be one way of gauging culturally-variant self-worth, positive self-regard, or esteem, our findings are a modest step in that direction.

The debate of whether self-esteem is best conceptualized as a global sense of selfworth or as distinct specific entities continues after a century of discussion, with support for both views. Consequently, some but not all researchers will be inclined to think of our measures of competencies as aspects of esteem. Regardless of whether one is or is not inclined to view what we measured as indicative of esteem, we think what is particularly important is that our competency measures were associated with established esteem measures in expected ways and differed across cultural groups in theoretically predictable ways, not only in terms of mean levels but also in terms of associations with positive affect.

Our findings pertinent to self-enhancement were quite interesting and have potential implications for counseling and counseling research. As predicted, group differences in competency evaluation depended on what was being evaluated (independent versus interdependent competencies) and to a lesser extent who was being evaluated (the self versus the ingroup). For example, whereas EA rated themselves higher than did AA in the degree to which they possessed desirable independent competencies, AA rated their ingroups as possessing more desirable interdependent competencies than did EA. Furthermore, these self-enhancing tendencies appeared to have culture-specific effects on positive affect; whereas high levels of independent competency evaluation contributed to EA happiness, AA happiness was more influenced by interdependent competency evaluations. These results support a growing body of literature that indicates that Asians and Asian Americans self-evaluate positively but only on self-congruent domains (SEDIKIDES et al., 2003; YIK et al., 1998). In addition to demonstrating self-enhancement, competency evaluations contributed to happiness above that of global esteem, suggesting that this type of evaluation remains important for AA even after accounting for global esteem. Furthermore, once interdependent competency evaluations were considered, global collective esteem no longer predicted AA happiness suggesting that ingroup competency evaluations may be a more powerful predictor of happiness than global collective esteem. Perhaps global evaluations of one’s ingroup may be too westernized in its language or may not reveal subtle cultural differences in what collectivists consider desirable.

Several shortcomings of the present study should be noted. First, our competency measures were based on a small number of independent and interdependent values (SCHWARTZ, 1994). However, the values that we combined as either generally individualistic or collectivistic have been related to different specific types of individualism and collectivism (OISHI et al., 1998). Therefore, it will be important for future research to elucidate which specific types of independent and interdependent competencies are associated with cultural differences. Similarly, although we used an established measure of positive affect, future research should pay more attention to different types of positive affect (e.g., BERENBAUM, 2002) and their cultural relevance. Finally, the sample was ethnically heterogeneous and may have obscured group differences.

These issues of identity, ingroup, and positive feelings are useful to consider in multicultural counseling contexts. For example, asking if one’s ingroup is sufficiently polite rather than if one considers oneself to be worthy of one’s ingroup may better capture what is culturally successful, and thus, may better gauge positive self-worth. Certainly, consistent with this argument AA enhanced when they evaluated their ingroups on specific but not global competencies. Furthermore, it is unlikely that any individual endorses only one view of the self (HIGGINS, 1996). In cultures that are heterogeneous, such as the U.S. and Brazil, different ethnic minority groups are likely to endorse multiple selves (TSAI et al., 2001). Thus, it should not be surprising that the magnitude of the group differences was modest. Although mean differences exist at the cultural level, mean variation at the individual level is complex. Furthermore, while the groups studied were defined by race, these differences/similarities might exist between any number of cultural groups defined by ethnicity, race or religion. Thus, an important avenue for future counselors and counseling research is to explore the co-existence of multiple types of esteem that are potentially not complementary in multi-cultural societies. For example, how does individual who is both independent and interdependent negotiate different sources of esteem? Addressing such issues would enrich our understanding of the self, culture, and esteem and improve services to multicultural clients.

 

Conclusion

While this study is exploratory, it suggests that the ways in which we can measure esteem and examine its effects on well-being is complex.

Not only should we consider individual self-achievement as a source of esteem, for people whose identities are closely interlinked to their families and communities, esteem derived from the maintenance of these ties may be as, if not more relevant. Future studies should consider the importance of examining esteem linked to the dynamic interplay of individual and collective selves as we navigate complex personal identities in increasingly multi-cultural worlds.

 

References

AHLERING, R. Collective self-esteem data from four ethnic groups. Psychological Reports, v. 92, p. 562-564, 2003.        [ Links ]

AIKEN, L. S.; WEST, S. G. Multiple regression: testing and interpreting interactions. Thousand Oaks: Sage Publications, 1991.        [ Links ]

BAE, S-W.; BREKKE, J. S. The measurement of self-esteem among Korean Americans: a cross-ethnic study. Cultural Diversity & Ethnic Minority Psychology, v. 9, p. 16- 33, 2003.        [ Links ]

BENET-MARTINEZ, V.; KARAKITAPOGLU-AYGUN, Z. The interplay of cultural syndromes and personality in predicting life satisfaction. Comparing Asian Americans and European Americans. Journal of Cross-Cultural Psychology, v. 34, p. 38-60, 2003.        [ Links ]

BERENBAUM, H. Varieties of joy-related pleasurable activities and feelings. Cognition and Emotion, v. 16, p. 473-494, 2002.        [ Links ]

CHENG, H.; FURNHAM, A. Attributional style and self-esteem as predictors of psychological well being. Counseling Psychology Quarterly, v. 16, p. 121-130, 2003.        [ Links ]

CROCKER, J.; LUHTANEN, R.; BLAINE, B.; BROADNAX, S. Collective self-esteem and psychological well-being among White, Black, and Asian college students. Personality and Social Psychology Bulletin, n. 20, p. 503-514, 1994.        [ Links ]

FEATHER, N. T.; MCKEE, I. R. Global self-esteem and the fall of high achievers: australian and Japanese comparisons. In: PANDEY, J.; SINHA, D.; BWAWUK, D. P. S. (Eds.). Asian contributions to cross-cultural psychology. New Delhi: Sage Publications, 1996. p. 200-213.        [ Links ]

FLEMING, J. S.; COURTNEY, B. E. The dimensionality of self-esteem: II. Some results for a college sample. Journal of Personality and Social Psychology, v. 39, p. 921-929, 1984.        [ Links ]

GEGEO, K. A.; GEGEO, W. Shaping the mind and straightening out conflicts: the discourse of Kwara’ae family counseling. Conflict discourse in Pacific societies. Stanford, CA: Stanford University Press, 1990.        [ Links ]

HEINE, S.; LEHMAN, D. R.; MARKUS, H. R.; KITAYAMA, M. Is there a universal need for positive self-regard? Psychological Review, v. 106, p. 766-794, 1999.        [ Links ]

HIGGINS, T. The “self digest”: self-knowledge serving self-regulatory functions. Journal of Personality and Social Psychology, v. 71, p. 1.062-1.083, 1996.        [ Links ]

JAMES, W. The principles of psychology. New York: Holt, 1890.KWAN, V.; BOND, M. H.; SINGELIS, T. M. Pancultural explanations for life satisfaction: adding relationship harmony to self-esteem. Journal of Personality & Social Psychology, v. 73, p. 1.038-1.051, 1997.        [ Links ]

LUHTANEN, R.; CROCKER, J. A collective self-esteem scale: self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, n. 18, p. 302-318, 1992.        [ Links ]

MARKUS, H. R.; KITAYAMA, S. Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, v. 98, p. 224-253, 1991.        [ Links ]

MCCLELLAND, G. H.; JUDD, C. M. Statistical difficulties of detecting interaction and moderator effects. Psychological Bulletin, n. 114, p. 376-390, 1993.        [ Links ]

OISHI, S.; SCHIMMACK, U.; DIENER, E. The measurement of values and individualismcollectivism. Personality and Social Psychology Bulletin, n. 11, p. 1.177-1.189, 1998.        [ Links ]

PELHAM, B. W. Self-investment and self-esteem: evidence for a Jamesian model of self-worth. Journal of Personality and Social Psychology, v. 69, n. 6, p. 1.141- 1150, 1995.        [ Links ]

ROSENBERG, M. Society and the adolescent self-image. Princeton, NJ: University Press, 1965.        [ Links ]

SCHWARTZ S. H. Beyond individualism and collectivism: new cultural dimensions of values. In: KIM, U.; TRIANDIS, H. C.; KAGITCIBASI, C.; CHOI, S-C; YOON, G. (Eds.). Individualism and collectivism: Theory, method, and applications. Newbury Park: Sage, 1994. p. 85-122.        [ Links ]

______; BILSKY, W. Toward a theory of universal content and structure of values: extensions and cross-cultural replications. Journal of Personality and Social Psychology, v. 58, p. 878-891, 1990.        [ Links ]

SEDIKIDES, C.; GAERTNER, L.; TOGUCHI, Y. Pancultural Self-Enhancement. Journal of Personality and Social Psychology, v. 84, p. 60-79, 2003.        [ Links ]

SINGELIS, M. T. The measurement of independent and interdependent self-construals. Personality and Social Psychology Bulletin, n. 20, p. 580-591, 1994.        [ Links ]

SOLOMON, S.; GREENBERG, J.; PYSZCYNSKI, T. A terror management theory of social behavior: the psychological functions of self-esteem and cultural worldviews. In: BERKOWITZ, L. (Ed.). Advances in experimental social psychology, San Diego: Academic Press, v. 24, p. 93-159. 1991.        [ Links ]

TRIANDIS, H. C. Individualism and collectivism. Boulder: Westview Press, 1995.        [ Links ]

______. The self and social behavior in differing cultural contexts. Psychological Review, v. 96, p. 506-520, 1989.        [ Links ]

TRIANDIS, H. C.; GELFAND, M. J. Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology, v. 74, p. 118-128, 1998.        [ Links ]

TSAI, J. L.; YING, Y. W.; LEE, P. A. Cultural predictors of self-esteem: a study of Chinese American female and male young adults. Cultural Diversity and Ethnic Minority Psychology. v. 3, p. 284-297, 2001.        [ Links ]

WATSON, D.; CLARK, L. A. Preliminary manual for the PANAS-X: Positive and Negative Affect Schedule-Expanded Form. Unpublished manuscript, Southern Methodist University, Iowa, 1991.        [ Links ]

WATSON, D.; CLARK, L. A.; TELLEGEN, A. Development and validation of brief measures of positive and negative affect: the PANAS scales. Journal of Personality and Social Psychology, v. 54, p. 1.063-1.070, 1988.        [ Links ]

WOOD, J. V.; HEIMPEL, S. A.; MICHELA, J. L. Savoring versus Dampening: self-esteem differences in regulating positive affect. Journal of Personality and Social Psychology, v. 85, p. 566-580, 2003.        [ Links ]

YIK, M. S. M.; BOND, M. H.; PAULHUS, D. L. Do Chinese self-enhance or self-efface?: it’s a matter of domain. Personality and Social Psychology Bulletin, n. 24, p. 399- 406, 1998.        [ Links ]

 

 

Endereço para correspondência
Contato:
Chitra Raghavan, Ph.D.
Assistant Professor Department of Psychology
John Jay College of Criminal Justice
City University of New York
New York, New York, 10019
e-mail: craghavan@jjay.cuny.edu

Tramitação
Recebido em maio de 2006
Aceito em agosto de 2006