Success through a cultural lens: perceptions, motivations, and attributions.

Feeny, Kimberly ; Wang, Qi

The interest in better understanding human action and behavior has been a focal point of research across disciplines for many decades. Motivation has been examined as a way to better understand human behavior and dynamics. Motivation leads us to success, which can be thought of as the achievement of something desired, planned, or attempted. People may pursue success for either a personal achievement (e.g. promotion at a job) or a group achievement (e.g. team victory). Regardless, the desire to succeed keeps us going, allowing us to constantly have something to work toward.

Success is often subjectively defined and perceived. For one person, performing well as a member of a team is a success even if the team did not win the match. For another, however, the team's victory is a success, in lieu of personal performance. What people value has an influence on what they attain and identify as a success. Individualism-collectivism has been identified as a cultural value that is rooted in Western and Eastern influences respectively (Hofstede, 1980). Individualism and collectivism lead to differences in how the boundaries among people are conceived, which, in turn, lead to differences between individualistic and collectivistic values (Markus & Kitayama, 1991). The primary individualistic view is that there are sharp boundaries between people, with each person being a complete unit. A person's identity in an individualistic society tends to be based mainly on one's personal experiences. By contrast, the primary collectivistic view is that people are not separate units, but rather are part and parcel of a larger group. A person's identity in a collectivistic society tends to be based on one's roles and experiences within the group context.

The basic individualistic and collectivistic view of people as either independent or interdependent leads to a different set of values (Markus & Kitayama, 1991). Consequently, people adopting these two different value systems may perceive, interpret, and react towards success through different lenses. Particularly, individualists perceive the world through self-orientated values and may view success as a personal achievement, whereas their collectivist counterparts most likely have group-orientated values and view success as their role or contribution in a collaborative, group achievement.

The purpose of this study was to look at cultural individualism-collectivism and its influence on perceptions of success, as well as attribution to that success, and motivation for that success. Moreover, because members in different cultures may adopt their dominant cultural values at different degrees, this study also looked into individuals' self-construal for the variations in success-related concepts (Markus & Kitayama, 1994). The following sections review cultural individualism-collectivism, self-construal, and success-related concepts. Hypotheses are proposed accordingly.

Individualism-Collectivism and Self-Construal

Individualism-collectivism has been identified as a major cultural dimension that differs among Western and Eastern cultures (Hofstede, 1980). Individualists stress human independence and the importance of individual self-reliance and autonomy. They also promote the exercise of individual goals and desires. Collectivists stress human interdependence and the importance of a collective rather than the importance of separate individuals. Collectivists focus on community and society and seek to give priority to group goals over individual goals. The collectivist view of the world is holistic, in that the world is greater than the sum of its parts (Markus & Kitayama, 1994). Nisbett (2003) attributes Westerners (referring to "primarily Europeans, Americans, and citizens of the British Commonwealth", p. xvi) as being motivated by positive comments and self-achievement, whereas Easterners (referring to "principally the people of China, Korea, and Japan" p. xvi) are motivated by negative comments and group goals. Westerners' behavior is guided by distinct, self and personal traits and Easterners are influenced by social roles and propriety. Individualism is a personal agency that values personal uniqueness and status. Collectivism is a collective agency that values cooperation, self-control, social order, and loyalty (Nisbett 2003).

Research on self-construal across cultures has identified problems with grouping all members of one culture under individualism and collectivism. Because culture is diverse, differences exist within it. Not all members of the same culture can be attributed to having the same values. Cross-cultural research on self has suggested that self-construal is an important mediator of cultural behavioral patterns. Self-construal is thought of as an individual-level cultural orientation and is theorized to mediate and explain the effects of culture on a variety of social behaviors (Ting-Tommey, 1989; Triandis, 1989). In discussing self-construal and relating the construct to individualism and collectivism, Markus and Kitayama (1991) used the terms interdependent and independent to identify those who endorse collectivist and individualistic values respectively.

Independent and Interdependent Self-Construal

In the independent construal (Markus & Kitayama 1991), most representations of the self have as their referent an individual's ability, characteristics, attributes, and goals. These inner traits and qualities are that person's primary regulator or motivator for behavior. The goal of individuals with an independent construal would thus be to stand out and to express their uniqueness. This orientation emphasizes the need to pursue personal self-actualization or to develop personal parts of the self. Goal seeking and success will then be motivated by ways in which the self can be developed and can be uniquely apart from the mass.

By contrast, in the interdependent construal, the self is connected to others and the principal components of the self are one's relationships and closeness to others (Markus & Kitayama, 1991). Internal traits are not void in a person who identifies with an interdependent self-construal. The difference is that this individual looks to enhance others belonging to his or her group with their own individual qualities and abilities. It is not how one can stand out, but how one can contribute to the whole, that counts. "Behavior is more significantly regulated by a desire to maintain harmony and appropriateness in relationships. Within such a construal, the self becomes most meaningful and complete when it is cast in the appropriate social relationship (Kim et al., 1996, p. 33)." For the interdependent individual, behavior is more guided by the want to meet the needs of others and promote others' goals or a group's goals, rather than their own, separate goals.

The concept of self-construal allows researchers to investigate the values of individualism-collectivism on an individual level, rather than a cultural level. Kim et al. (1996) indicated the problem of the cross-cultural research where cultural-level individualism-collectivism was assumed to be uniform upon everyone. "Individualism and collectivism rather appear to represent two dimensions along which people vary (Kim et al., 1996, p. 31)." Similarly, Triandis (1989) suggested that although the terms individualism and collectivism should be used to characterize cultures and societies, idiocentric and allocentric should be used to characterize individuals. These classifications are similar to Markus and Kitayama's (1991) independent and interdependent self-construals, representing the connectedness and relations found in collectivist cultures (interdependent) and the separateness and uniqueness of the individual stressed in individualistic cultures (independent). Markus and Kitayama (1991) argue that the different images of the self influence an individual's cognition, emotion, and motivation.

In the discussion of self-construal research, it is important to mention recent criticism that exists in the literature about self-construal scales. In an article which calls to question the validity of self-construal scales, Levine et al. (2003) found self-construal scales to be radically multidimensional and highly unstable within and across cultures. From these findings, they suggest that validity problems exist in cross-cultural research involving the use of self-construal scales. Gudykunst and Lee (2003) disregarded this conclusion and argued that the results of Levine et al.'s meta-analysis and priming studies did not raise problems with the validity of self-construal scales because their results were compatible with theorizing about self-construal. The priming studies that Levine et al. (2003) conducted should not be expected to influence the respondent's scores on self-construal scales: "Priming only influences the self-construal individuals' activity, not their scores on the scales" (p .266). Gudykunst and Lee (2003) also numerated the studies that have found support in Markus and Kitayama's theory and construct validity in the measurement of self-construal. In addressing this debate, our study examined whether an Eastern culture such as Chinese, and a Western culture such as American would still predict individualism-collectivism and independent and interdependent self-construals, as Markus and Kitayama (1991) proposed almost two decades ago.

RQ1: Will an Eastern cultural membership lead to higher collectivism and an interdependent self-construal? Will a Western cultural membership lead to higher individualism and an independent self-construal?

The variance between independent and interdependent self-construals sets the stage for differences in identifying success. Those who value independence will most likely identify a success that was achieved on the personal level and those who value interdependence will most likely identify a success that was a group success in which they had a contribution. As Nisbett (2003) and Markus and Kitayama (1991) discussed, individualism (independence) stresses self-sufficiency, whereas collectivism (interdependence) stresses reliance and cooperation with the group. Unlike individualists, competition is not something that is valued in collectivist societies, where people take more value in fulfilling their roles within a group. These differences suggest that individualist and collectivist cultures assign different valences to personal successes and group successes. On a related point, people who value individualism and collectivism differently may think about the reasons that cause success, and the motivation for success, differently.

Success in an Attribution Theorist Perspective

Attribution theorists investigate the perception of causality, or the judgment of why a particular incident occurred. The placement of cause and responsibility (i.e., attribution) guides subsequent behavior. As Kelley (1971) stated, "The attributor is not simply an attributor, a seeker after knowledge; his latent goal in attaining knowledge is that of effective management of himself and his environment (p. 22)." Under the lens of attribution theory, success is generally ascribed to some ability factor that includes both aptitude or capability and acquired skills, effort, the difficulty or ease of task, the personality and mood at the person, and help or hindrance from others (Graham, 1991). In achievement-related contexts, success may be attributed to high ability and effort, and failure is attributed to low ability or effort (Weiner, 1972). Therefore, success and its attributions are intertwined.

Attribution theory identifies locus, stability, and controllability as causal dimensions on which behavior falls (Weiner, 1985). The locus dimension defines the location of a cause as internal or external to the individual. Effort and ability are internal causal attributes because they mirror characteristics of the person. However, task difficulty and luck are external factors that exist within the environment. The stability dimension specifies causes as constant or varying over time. Ability is a stable factor, whereas effort and mood are unstable because individuals may differ from one situation to the next. The controllability dimension refers to personal responsibility, or whether a cause is subject to one's own volitional influence. Although effort is unstable, it is controllable because individuals are believed to be responsible for how hard they try. On the other hand, ability and luck are not within the control of the individual. These conceptual differences between causes are central to an attribution model of motivation because each dimension is uniquely related.

Nisbett (2003) has discussed differences in the ways Easterners and Westerners attribute causality (2003). Westerners engage in more causal attributions and attend primarily to the object engaging in a behavior. Easterners attend more, generally, to the environment and to the relations between the object producing a behavior and the environment that the object is in. "Westerners tend to assume that events are caused by the object and Asians are inclined to assign greater importance to the context" (Nisbett, 2003, p. 127). Therefore, Easterners are more likely to notice important situational factors and to realize that the environment plays a role in producing and influencing behavior. Westerners do not attribute causality to environmental factors; rather, they place responsibility of behavior on the individual.

Success in a Social Learning Theorist Perspective

Bandura's (1989) social learning theory and social cognitive theory focuses on mechanisms and effects of motivation. Through his work, Bandura (2001) discussed the core features of human agency and how they address the issue of what it means to be human. He identifies them as intentionality, forethought, self-reactiveness, and self-reflectiveness. Intentions center on a plan of action, thereby representing a goal, something which is visualized. Turning these visualized objectives into reality requires intentions that guide and keep one moving ahead (Bandura, 1991). Bandura (2001) maintained that intentions lead to forethought and visualization or formation of goals, which further lead to the plans of actions. By being represented in the present, anticipated future events are changed into current motivators and regulators of behavior. Thus, behavior is motivated directly by projected goals and anticipated outcomes rather than being pulled by an unrealized future state.

Self-reactiveness involves not only the deliberative ability to make choices and action plans, but the ability to direct to appropriate courses of action. Finally, self-reflectiveness urges people to be not only agents of action, but self-examiners of their own functioning. In sum, intentionality, forethought, self-responsiveness and self-reflectiveness are core aspects that guide and influence behavior. Although these four elements offer a strong basis for understanding motivation, efficacy is another important concept that influences people's motivations to act.

Self-efficacy is an important determinant of human motivation, affect, and action (Bandura, 1989). It refers to a person's belief in his or her capacity to produce desired results through certain actions. Most people will not be motivated to pursue a goal unless they perceive the relevant self-efficacy. Meanwhile, the stronger the perceived self-efficacy, the higher the goals set and the firmer the commitment made (Bandura, 1989).

Social cognitive theory also discusses the perceived collective agency (Bandura, 1997). People's shared belief in their collective power to produce desired results is a key ingredient of collective agency. Group successes are achieved through shared intentions, knowledge, and skills of its members. "Because the collective performance of a social system involves transactional dynamics, perceived collective efficacy is an emergent group-level property, not simply the sum of the efficacy beliefs of individual members (Bandura & Cervone, 1983, p. 1023)." Beliefs of collective efficacy serve functions similar to those of personal efficacy beliefs and operate through similar processes (Bandura, 1997).

Bandura (2001) argues that self-efficacy then is not confined to the self, nor does it necessarily exalt the self or spawn an individualistic lifestyle, identity, or morality that slights collective welfare. If belief in the power to produce results is put in the service of relational goals and beneficial social purposes, it fosters a communal life rather than eroding it. A high sense of self- and collective-efficacy promotes a pro-social orientation characterized by cooperativeness, helpfulness, and sharing, with a vested interest in each other's welfare (Bandura, 1989, 2001). Therefore, both self-and collective-agencies may motivate people to pursue success.

Hypotheses

In summary, one important dimension along which cultural values vary is individualism-collectivism. Perceived personal efficacy contributes to productive functioning by members of collectivistic cultures just as it does to functioning by people raised in individualistic cultures (Earley, 1993). However, cultural norms and characteristics shape the ways in which efficacy beliefs are developed, the purpose to which they are put, and the socio-structural arrangements through which they are best exercised (Bandura, 2001). People from individualistic cultures feel most efficacious and work most productively under an individually orientated system, whereas those from collectivistic cultures judge themselves most efficacious and work most productively under a group-orientated system.

Cultural members, nonetheless, may adopt cultural values at different levels. However, people with independent self-construals reflect and act upon individualist values and people with independent self-construals do so with collectivist values. Looking at success, attribution, and motivation through a cultural lens, the research posits the following hypotheses:

H1: An individualist culture, high individualism, and independent self-construals will predict the perception of success more as a personal achievement than a group achievement, compared with a collectivist culture, high collectivism, and interdependent self-construals.

H2: An individualist culture, high individualism, and independent self-construals will predict more internal attributions to success than external, compared with a collectivist culture, high collectivism, and interdependent self-construals.

H3: An individualist culture, high individualism, and independent self-construals will predict more importance placed on internal motivations than external, compared with a collectivist culture, high collectivism, and interdependent self-construals.

Method

Sample

Participants were 101 undergraduate communication students from an east coast university in the United States and 181 undergraduate mass media and journalism students from a southwestern university in China. The two samples were demographically comparable. For the U.S. sample, the majority were female (73.3%), Caucasian (83.2%), sophomores and juniors (85.1%), from a family with married parents (82.2%), having no traveling experience outside of the U.S. over one year (94.1%), and in the 19-22 age range (M = 20.25, SD = 0.73). Thirty students (29.3%) worked part-time. For the Chinese participants, most were sophomores and juniors (98.9%), individuals from a family with married parents (85.1%), with no experience staying outside of the country for longer than one year (97.8%), and in the age range of 19 and 25 (M = 21.02, SD = 1.26). Ninety (49.7%) were females. Sixty-four students (34.4%) had a part-time job and the rest were students only. The Chinese sample included the Chinese ethnicity unanimously.

Instruments for Individualism-Collectivism

To obtain a measurement of individualism and collectivism, participants responded to a 5-point Likert type, 11-item individualism-collectivism scale (INDCOL) that Cai and Fink (2002) adapted from Hui and Triandis' (1988) original 66-item form. The shortened version is used for its brevity and its equivalence with the original version: Cai and Fink found moderate reliability (Cronbach's [alpha] = .76) and a large correlation of the shortened version with the original measure (r = .83). The 11-item scale measured individualism-collectivism from three aspects: family, neighbor, and colleague. In this study, the Cronbach's alphas of the INDCOL scale were .70 for the U.S. sample, .65 for the Chinese sample, and .67 for the overall sample. Following Cai and Fink's suggestion (2002), participants' scores were summed and averaged to represent their degrees of individualism-collectivism. Higher scores represented higher collectivism and lower individualism.

Instruments for Self-Construals

Gudykunst et al.'s (1996) 12-item self-construal scale was used to measure independent and interdependent self-construals. A 5-point Likert type scale was used where 1 = Strongly Disagree and 5 = Strongly Agree. This measure was chosen because it had demonstrated high validity and reliability in cross-cultural communication research (see Gudkunst & Lee, 2003, for a summary). In this study, Cronbach's alphas for the independent self-construal were .63, .71, and .69 for the U.S. sample, the Chinese sample, and the overall sample, respectively, and were .70, .74, and .73 for the interdependent self-construal for the above three compositions in the same order. Participants' scores on each self-construal scale were summed and averaged to represent their levels of independent and interdependent self-construals separately.

Micro-Narratives: Success, Attributions to Success, and Motivations of Success

Participants were instructed to provide micro-narratives that asked them to describe a situation where they were successful (success), why and how they achieved that success (attribution), and what motivated or caused them to pursue that success (motivation). Micro-narratives refer to the specific stories that people tell about their lives (Baumeister, Stillwell, & Wotman, 1990). Baumeister et al. argued that because people often organize information about self-concepts within a story frame instead of an adjective-list frame, micro-narratives have high external validity and is especially suitable for the exploration of people's subjective perceptions and motivations. "The cognitive processes that produce stories are difficult to fathom. On the other hand, the motivations are presumably the same whether they operate on encoding, recall, or selection, and so autobiographical account may be a valued technique for examining these motivations (Baumeister, et al, 1990, p. 995)."

Coding Procedures

Content analysis proceeded to code participants' micro-narratives. The unit of the coding was a complete micro-narrative instead of separate clauses that describe success, attribution, and motivation. This decision was made because an initial examination of nine responses indicated that attributions and motivations were (a) often interwoven in the story of success, and (b) complemented throughout the story. Therefore, the coders had to read the complete story to grasp the full picture of how success, attributions, and motivations were told. A thematic coding scheme was then developed based on Boyatzis's (1998) suggestion of the five-element theme and code development: "a label [of a theme], a definition of what the theme concerns, a description of how to know when the theme occurs, a description of any qualifications or exclusions to the identification of the theme, and examples (p. 53)."

Specifically, success was defined as the achievement of something desired, planned, or attempted. To distinguish whether a participant told a story mainly about personal success or group success where the individual was a group member, a bipolar scale was used where -2 = group success (e.g., my team won), -1 = group primary and individual secondary (e.g., my team won and I helped a little), 0 = equal success shared by group and individual (e.g., my team won and I felt winning like others), 1 = individual primary and group secondary (e.g., it was my victory though the team won also), and 2 = individual success (e.g., I won). The key point was to distinguish whether a sense of team was mentioned in the story and to what degree. When no clear demarcation was detectable between personal and group success amounts, 0 was assigned. Themes of success were coded based on an inductively built-up list from an examination of all the stories that the participants told (see Boyatzis, 1998, for an inductive method of theme development). The success stories included seven themes: adventurous experience, extracurricular or habit-related success, academic or major-related success, work or job-related success, leadership success, social relationship success, and other.

Next, attribution to success was defined as the perception of causality of success, or the judgment of why the success occurred. The similar 5-point bipolar scale was used to code attributions, where -2 = external attribution (e.g., my success totally came from my family's support), -1 = external primary and individual secondary (e.g., my success mainly came from my family's support and my efforts counted too), 0 = equal amounts of external and internal attributions (e.g., my success came both from my family's support and my efforts), 1 = internal primary and external secondary (e.g., my success mainly came through my hard work and my family supported too), and 2 = internal attribution (e.g., I succeeded because I worked very hard). Further, four themes were used to categorize attributions based on Weiner (1985): ability, task difficulty, effort, and luck. An "other" category was used to guarantee exhaustion.

Finally, motivation was defined as the cause of the pursuit of success. The above 5-point bipolar scale was again adopted: -2 = external motivation (e.g., my pressure to win came from my parents), -1 = external primary and internal secondary (e.g., my pressure to win came mainly from my parents, and in some part from myself too), 0 = equal amounts of internal and external motivations (my pressure to win came from both my parents and myself), 1 = internal primary and external secondary (e.g., my pressure to win mainly came from my own needs for distinction and I know my parents will be happy too), and 2 = internal motivation (e.g., I wanted to win). Participants' motivations were then schematized based on Maslow's hierarchy of needs (1954): physiological needs, security needs, acceptance needs, esteem needs, and self-fulfillment needs. An "other" category was added to guarantee exhaustion.

Intercoder Reliability

Because this study involved two samples speaking two different languages, special caution was given to ensure intercoder reliability. First, double-coding procedure was used, where the coders rated identical responses and resolved all differences to reach high intercoder reliability (Boyatzis, 1998). One author, who is bilingual, trained both the Chinese coder (also bilingual) and the American coder in their native languages respectively. After explaining instructions and answering all the questions from the coders, the bilingual author discussed and coded 10 narratives from each culture with the respective coder and resolved all inconsistencies. Second, the bilingual author and the coders from each culture independently coded 10 narratives and compared the results across the six coding points (i.e., numeric values on the bipolar scale of success, attribution, and motivation, and numeric values representing the themes of the above three concepts) on each narrative. Because each bipolar scale (for success, attributions, and motivations, respectively) could be considered nominal or in some sense, ordinal, and the thematic categorization was nominal, intercoder reliability was calculated through percentage agreement (Boyatzis, 1998). For the Chinese sample, the intercoder agreement percentage was 91.7%. For the American sample, the intercoder agreement percentage was 90.0%. The bilingual author and the two coders discussed all inconsistencies and reached agreement. Third, the Chinese coder coded 10 narratives from the U.S. sample independently, and compared results with the American coder. The agreement percentage was 86.7%. Finally, the bilingual author discussed with the coders again to resolve inconsistencies, and then the two coders carried on coding the rest of the narratives in their own cultural group independently.

Translation Procedure

To obtain authentic results, the questionnaire was translated into Chinese for the Chinese participants. The back translation procedure was employed (van de Vijver & Leung, 2001; Werner & Campbell, 1970). First, the bilingual author translated the entire questionnaire into Chinese. Second, a bilingual professor in the English Department in the southwestern Chinese university where the Chinese sample was enrolled translated the Chinese version back to English. The two translators compared incongruities, and constructed the final Chinese language version. The above justification was done so that the two versions should arrive at equivalence as much as possible.

Data Collection Procedure

The U.S. participants answered the questionnaire online. The Chinese participants answered the questionnaire on printed pages in four group settings because of the lack of the convenience to access the internet. This procedural difference should not incur noticeable differences in the length of the micronarratives because the same number of lines was given for the participants to fill out.

Results

Micronarratives

Overall, the majority of the participants told a story about their personal success, regardless of cultural origin. No more than 5.0% of participants narrated a story where they felt successful because their group won. A large percentage of participants leaned toward personal attributions and motivations of their successes, though the Chinese responses appeared to vary more between external and internal polar ends than those of Americans' (Table 1).

As for the themes of success stories, the two samples were similar in that academic or major-related areas, work areas, and extracurricular-related areas were the most frequent themes. Three noticeable differences appear in the two samples. One, a sharp contrast appears in the percentages of the academic-related area: Whereas 63.0% of Chinese stories were around this area, only 41.6% of American stories were around it. Another noticeable difference was that American stories (13.9%) reported much more leadership success than the Chinese counterparts (1.7). Finally, the American stories were much more variable than the Chinese stories (Table 2). A Chi-square test indicated a significant cultural difference in the themes of success stories, [chi square] (6) = 30.04, p < .001.

<01_TB013>

<01_TB014>

The types of attributions to success, interestingly, seemed to be similar in the Chinese and American responses. Most of the participants attributed their successes to efforts. Whereas the attribution to luck appeared more in the Chinese, the attribution to ability appeared more in the Americans. An overall culture by attribution effect type was found between the two samples, [chi square] (4) = 15.27, p < .001.

For the types of motivations to success, both American and Chinese revealed more acceptance and esteem needs than the other motivation categories, though the Chinese appeared to cluster more around the former and the Americans, the latter. One interesting phenomenon was that more Chinese than Americans talked about realizing their dreams or life goals (coded as self-fulfillment). This finding may be a reflection of the Chinese long-term oriented values as compared with the Americans' short-term oriented values (Ting-Toomey & Chung, 2005). Again, an overall culture by motivation type effect was found between the two samples, [chi square] (4) = 122.13, p < .001.

Research Questions and Hypotheses Tests

The research question asked whether an Eastern culture such as Chinese and a Western culture such as American would still predict individualism-collectivism, and independent and interdependent self-construals as Markus and Kitayama (1991) proposed almost two decades ago. T-tests were conducted to address this research question. For individualism-collectivism, no significant cultural differences were found between Chinese (M = 4.01, SD = 0.43) and Americans (M = 3.99, SD = 0.42); t (258) < 1. However, significant cultural differences were found on independent and interdependent self-construals, and in the direction that Markus and Kitayama (1991) proposed. Chinese (M = 4.13, SD = 0.57) scored significantly higher than Americans (M = 3.93, SD = 0.50) on the interdependent self-construals, t (269) = 2.93, p < .001. Chinese (M = 3.99, SD = 0.42) scored significantly lower than Americans (M = 4.35, SD = 0.40) on the independent self-construals, t (275) = 2.88, p < .001. The effect sizes of the culture to predict the interdependent and independent self-construal were .04 and .03, respectively. Table 3 reports the correlations among individualism-collectivism, independent self-construal, and interdependent self-construal.

To test Hypotheses 1 through 3, discriminatnt analysis was appropriate because the independent variables include both continuous and categorical variables and the dependent variables were categorical (Hair, Anderson, Tatham, & Black, 1995). As shown in Table 1, the original 5-point bipolar ends to code success, attribution, and motivation incurred considerably unequal cell sizes and some cells had 0 input. Therefore, the 5-point categories were reduced to three categories for each of the three variables. For success, the group success (coded as -2) and group primary and individual secondary (coded as -1) were combined, and the individual success (coded as 2) and individual primary and group secondary (coded as 1) were combined. The equal amounts of individual and group success (coded as 0) remained the same. Consequently, the three categories represented success stories emphasizing a group, an individual, or both. Similarly, for success and attribution, external (coded as -2) and external primary and internal secondary (coded as -1) were combined, and internal (coded as 2) and internal primary and external secondary (coded as 1) were combined. The equal amounts of internal and external attribution or motivation (coded as 0) remained the same. The resulted three categories for attribution represent causes of success emphasizing external factors, internal factors, and equal amounts of external and internal factors. For motivation, the three categories reflect causes of the pursuit of success emphasizing needs driven by external factors, internal factors, or both. When testing the hypotheses, only bipolar ends were used in the discriminant analyses, because the hypotheses did not make predictions about the midpoint.

Further, before testing the hypotheses, potential multicollinearity among the three continuous independent variables was examined. For any of the three outcome variables (success, attribution, and motivation), the tolerance levels varied between .82 and .87. Therefore, no multicollinearity threat was detected.

H1 predicted more personal than group success stories by the American culture, high individualism (and low collectivism) on the individualism-collectivism (I-C) scale, high independent self-construals (IND-SC), and low interdependent self-construals (INTER-SC). This hypothesis was not supported. Discriminant analysis results revealed no significant effect from the four predictors, Wilk's Lambda = .99, [chi square] (4) = 3.24, p > .05. The canonical correlation for the first discrimnant function was .12. The correctly classified cases amounted to 97.9%. The reason for the independent variables to lack discriminant functions may be that most of the participants told individual success stories in lieu of cultural memberships.

H2 stated that the American culture, high individualism (and low collectivism) on the I-C scale, high IND-SC, and low INTER-SC would predict more internal than external attributions to success. This hypothesis was partially supported. Together, the four predictors significantly predicted people's likelihoods to make internal or external attributions, Wilk's lambda = .90, [chi square] (4) = 19.20, p < .001. The canonical correlation was .31, resulting in about 10% variance explained in the outcome variable. Both national culture (F[1, 192] = 12.11 p < .001) and individualism-collectivism (F[1, 192] = 7.70, p < .01) contributed uniquely to the discriminant prediction. The correct classification percentage was 80.9%. As reflected in the standardized canonical discriminant function, the Chinese culture (coded higher than the U.S. culture in the numeric value) and higher collectivism (and lower individualism) predicted the higher probability of external attributions.

[Z.sub.D1] = .77*[Z.sub.culture] + .61*[Z.sub.I-C] + .31[Z.sub.IND-SC] - .15[Z.sub.INTER-SC]

The asterisks indicate the significant coefficients, Z represents the standardized z-score, and D1 refers to the first discriminant function. A [chi square]-test was further performed to test the cultural effect on attribution type. Consistent with the discriminant results, culture significantly predicted attribution type, [chi square] (2) = 16.02, p < .001. Figure 1 indicates that whereas in both cultures, the majority attribute their successes to internal reasons, more Chinese attribute their successes to external reasons or to both internal and external reasons than Americans.

[FIGURE 1 OMITTED]

H3 predicted more internal motivations from American cultural membership, high individualism (and low collectivism) on the I-C scale, high IND-SC, and low INTER-SC. This hypothesis was not supported except for the unique effect of individualism-collectivism. The four predictors together did not predict the likelihood of internal vs. external motivations, Wilk's lambda = .96, [chi square] (4) = 7.67, p > .05. The canonical correlation was .19. The I-C scale, however, had significant discriminant effect on motivation type, F (1, 191) = 5.30, p < .05. As reflected in the standardized discriminant function below, the higher the collectivism (and hence the lower the individualism), the more likely the external attribution.

[Z.sub.D1] = .35[Z.sub.culture] + .99*[Z.sub.I-C] - .28[Z.sub.IND-SC] - .22[Z.sub.INTER-SC]

Discussion and Conclusion

This study took interest in understanding how people perceive success, what they attribute to success, and what they consider to be the drives for the pursuit of success. The study further looked into cultural differences as the explanatory agents. Following the literature in cultural individualism-collectivism and self-construals (e.g., Markus & Kitayama, 1991) and in cultural differences in cognition (e.g., Nisbett, 2003), the study hypothesized that a collectivist culture such as Chinese should reinforce its members with grouporiented cognitive constructs about success, whereas an individualist culture such as American should reinforce its members with individual-oriented cognitive constructs about success. The hypothesis was partially supported. The most distinct support was found in the cultural differences on attributions, where more Chinese make external than internal attributions as compared with their American counterparts.

This finding supported Nisbett and colleagues' research about cognitions in Eastern and Western cultures. Nisbett (2003) argued that because Eastern cultures value a collectivist mindset, think holistically, and perceive the world as made of continuation of substance, they tend to make external attributions in understanding self, other, the environment, and events. Western cultures tend to use a personal agent, think analytically, and perceive the world as made of different categories of objects, and therefore are more likely to make internal attributions to observed phenomena. In this study, people were asked to give a success story. As reflected in the data, most people recalled a personal story, yet still, the Chinese participants reported more external attributions to their successes as compared with the Americans. Constructivism proponents argue that cognitive constructs through which people perceive and interpret the world, namely, personal constructs (Kelly, 1955), are greatly influenced by social interactions. Because the Chinese culture reinforces group values and American culture reinforces individual values through socialization, their cognitive constructs may be activated or connected differently when asked to recall an event.

Besides national culture, the individualism-collectivism dimension also predicted attribution and motivation types; the higher the collectivism and the lower the individualism, the more likely people make external attributions and motivations. This finding implies that depending on the people's value systems, different motivational agents are effective in leading people to pursue success. People who value collectivism are more likely to recall external motivations. In China, filial piety is a highly commendable value. Children work hard to please parents and teachers. One goal of studying hard and working hard is to secure a stable life so that parents could be attended and family could be glorified. The Chinese participants' micronarratives focusing on academic-related success also illustrates this point. In the United States, people are encouraged to be responsible for their own lives, and the pursuit of success is largely to serve personal needs rather than group needs. The different educational systems in the two cultures may indoctrinate different individualism-collectivism values, and therefore, lead to different motivational agents for success.

Moreover, this study supported the link from culture (American vs. Chinese) to independent and interdependent self-construals, which was proposed nearly two decades ago by Kitayama and Markus (1991). Americans are still more independent and less interdependent as compared with Chinese on average. The influence of cultural membership on people's self-concepts appears to be a valid prediction. Nisbett (2003) discussed two possibilities regarding the relationship between globalization and cultures. Cultures could converge because of the increased contact and information exchange, facilitated by technology. However, cultures could diverge. Despite that people may be able to adapt to different cultures behaviorally, once they have a chance to perceive the deep-rooted differences in values, beliefs, and worldviews, they could feel an unsurpassable gap between peoples. Therefore, whereas certain values may collapse across cultures, others remain to affect peoples differently.

This study employed micronarratives to explore people's perception of success. The method had the merit to allow people to tell fully about the success story and concepts and themes around it. As Baumeister et al. (1990) argued, people's constructs may be organized in the story or event format, rather than a listing format. This method could be more natural in eliciting people's information and perceptions about a concept as compared with scales made up of statements or bipolar adjectives. Future research could further explore this method, especially in cognition and emotion-related research questions.

This study is limited in several aspects. First, the study's use of micronarratives and the involvement of two languages made the coding procedure challenging. Although the coders were trained closely, the American coder did not speak Chinese, therefore a cross-validation was done unilaterally. Second, the instruments used in this study only had moderate reliabilities, especially the shortened INDCOL scale. Therefore, the study cannot speak strongly in the debate about whether cultural individualism-collectivism still makes predictions about Eastern and Western cultural differences (e.g., Oyserman, Koon, & Kemmelmeiyer, 2002). Finally, the study did not focus on the examination of cultural values as they relate to the themes of success, attribution, and motivation, or make predictions about them. The differences found in the two cultures about these themes may be further pursued. For example, more Chinese attributed success to luck than Americans. Although both luck and group attributions are external, people who made these attributions may have different cognitive constructs about them.

This study contributes to the field of cross-cultural as well as cognitive communication research. Through the comparison of the cognitive constructs about success, the study reveals how cultural values influence people's perceptions and explanations of this concept. Educators may use the results from this study to find effective motivational methods for learners. Intercultural communicators may benefit from this study by knowing to be prepared that the other may be seeing the same object but will be using very different cognitive schemes to perceive and explain it.

References

Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1-26.

Bandura, A. (1991). Self-regulation of motivation through anticipatory and self-reactive mechanisms. In R. A. Dienstbier (Ed.), Perspectives on motivation: Nebraska symposium on motivation (pp. 69-164). Lincoln, NE: University of Nebraska Press.

Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist. 44, 1175-1184.

Bandura, A., & Cervone, D. (1983). Self-evaluative and self-efficacy mechanisms governing the motivational effects of global systems. Journal of Personality and Social Psychology. 45, 1017-1028.

Baumeister, R. F., Stillwell, A., & Wotman, S. R. (1990). Victim and perpetrator accounts of interpersonal conflict: Autobiographical narratives about anger. Journal of Personality and Social Psychology. 59, 994-1005.

Boyatzis, R. E. (1998). Transforming qualitative information: Thematic analysis and code development. Thousand Oaks, CA: Sage.

Cai, D. A., & Fink, E. L. (2002). Conflict style differences between individualists and collectivists. Communication Monographs, 69, 67-87.

Earley, P.C. (1993). East meets West meets Mideast: Further explorations of collectivistic and individualistic work groups. Academic Management Journal, 36, 319-348.

Graham, S. (1991). A review of attribution theory in achievement contexts. Educational Psychology Review, 3, 5-39.

Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K., & Heyman, S. (1996). The influence of cultural individualism-collectivism, self-construals, and individual values on communication styles across cultures. Human Communication Research, 22, 510-543.

Gudykunst, W. B., & Lee, C. M. (2003). Assessing the validity of self-construal scales: A response to Levine et al. Human Communication Research, 29, 253-274

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1995). Multivariate data analysis. Upper Saddle River, NJ: Prentice Hall.

Kelley, H. H. (1971). Attributions in social interactions. Morristown, NJ: General Learning Press.

Kim, M. S., Hunter, J. E., Miyahara, A., Horvath, A., Bresnahan, B., & Yoon, H. (1996). Individual--vs. culture-level dimensions of individualism and collectivism: Effects on preferred conversational styles. Communication Monographs, 63, 29-49.

Levine, T. R., Park, H. S., Lapinski, M. K., Wittenbaum, G. M., Shearman, S. M., Lee, S. Y., Chung, D., & Ohashi, R. (2003). Self-construal scales lack validity. Human Communication Research 29, 210-252.

Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253.

Markus, H. R., & Kitayama, S. (1994). The cultural construction of self and emotion: Implications for social behavior. In S. Kitayama & H. R. Markus (Eds.), Emotion and culture (pp. 89-130). Washington D.C.: American Psychological Association.

Maslow, A. (1954). Motivation and personality. New York: Harper.

Nisbett, R. E. (2003). The geography of thought: How Asians and Westerners think differently and why. New York, NY: Free Press.

Oyserman, D., Coon, H. M., & Kemmelmeier, M, (2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-72.

Ting-Toomey, S. (1989). Identity and interpersonal bonding. In M. K. Asante & W. B. Gudykunst (Eds.), Handbook of international and intercultural communication (pp. 163-185). Newbury Park, CA: Sage.

Ting-Toomey, S., & Chung, L. C. (2005). Understanding intercultural communication. Los Angeles: Roxbury.

Triandis, H.C. (1989). The self and social behavior in differing cultural context. Psychological Review, 24, 366-383.

Weiner, B. (1985). An attributional theory of achievement and emotion. Psychological Review, 92, 548-573.

Weiner, B. (1972). Attribution theory, achievement motivation, and the educational process. Review of Educational Research, 42, 203-215.

Kimberly Feeny & Qi Wang

Villanova University

Correspondence: Kimberly Feeny is a graduate student and Qi Wang (Ph.D., University of Maryland) is an assistant professor at Villanova University. This paper was partially funded by the 2008 Faculty Summer Communication Research Grant from the Department of Communication at Villanova University. An earlier version won the Top Four Paper Award at the Association of Chinese Communication Studies division at the National Communication Association convention in Chicago, November 2009. Questions and comments should be sent to Qi Wang at q.wang@illanova.edu or 610-519-4439 (phone).

Table 1
Perceptions of Success, Attribution, and Motivation by Culture

 Success

 U.S. China Overall
 (n) (n) (N)

Group or 3 4 7
External (3.0%) (2.2%) (2.5%)

Group or 0 1 1
External as (0.6%) (0.4%)
Primary

Equal 1 4 5
 (1.0%) (2.2%) (1.8%)

Individual or 8 19 27
Internal (7.9%) (10.5%) (9.6%)
Primary

Individual or 88 151 239
Internal (87.1%) (83.5%) (84.8%)

 Attribution

 U.S. China Overall
 (n) (n) (N)

Group or 4 7 11
External (4.0%) (3.9%) (3.9%)

Group or 3 29 32
External as (3.0%) (16.0%) (11.3%)
Primary

Equal 16 45 61
 (15.8%) (24.9%) (21.6%)

Individual or 20 60 80
Internal (19.8%) (33.1%) (28.4%)
Primary

Individual or 58 38 96
Internal (57.4%) (21.0%) (34.0%)

 Motivation

 U.S. China Overall
 (n) (n) (N)

Group or 13 5 18
External (12.9%) (2.8%) (6.4%)

Group or 5 35 40
External as (5.0%) (19.3%) (14.2%)
Primary

Equal 20 39 59
 (19.8%) (21.5%) (20.9%)

Individual or 17 58 75
Internal (16.8%) (32.0%) (26.6%)
Primary

Individual or 43 42 85
Internal (42.6%) (23.2%) (30.1%)

Table 2
Themes for Success, Attribution, and Motivation by Culture

 U.S. China

Success Adventure 3 (3.0%) 10 (5.5%)
 Extracurricular 22 (21.8%) 20 (11.0%)
 Academic or major 42 (41.6%) 114 (63.0%)
 Work 12 (11.9%) 21 (11.6%)
 Leadership 14 (13.9%) 3 (1.7%)
 Social relations 5 (5.0%) 3 (1.7%)
 Other 3 (3.0%) 8 (4.4%)

Attribution Ability 20 (19.8%) 27 (14.9%)
 Task difficulty 3 (3.0%) 5 (2.8%)
 Efforts 75 (74.3%) 112 (61.9%)
 Luck 1 (1.0%) 11 (6.1%)
 Other 2 (2.0%) 24 (13.3%)

Motivation Physiological needs 0 13 (7.2%)
 Security 5 (5.0%) 16 (8.8%)
 Acceptance 11 (10.9%) 82 (45.3%)
 Esteem 75 (74.3%) 20 (11.0%)
 Self-fulfillment 7 (6.9%) 48 (26.5%)
 Other 3 (3.0%) 2 (1.1%)

 Overall

Success Adventure 13 (4.6%)
 Extracurricular 42 (14.9%)
 Academic or major 156 (55.3%)
 Work 33 (11.7%)
 Leadership 17 (6.0%)
 Social relations 8 (2.8%)
 Other 11 (3.9%)

Attribution Ability 47 (16.7%)
 Task difficulty 8 (2.8%)
 Efforts 187 (66.3%)
 Luck 12 (4.3%)
 Other 26 (9.2%)

Motivation Physiological needs 13 (4.6%)
 Security 21 (7.4%)
 Acceptance 93 (33.0%)
 Esteem 95 (33.7%)
 Self-fulfillment 55 (19.5%)
 Other 5 (1.8%)

Table 3

Correlations Between Individualism-Collectivism and Self-Construal

 IC ISC
Individualism-Collectivism (IC)
Independ Self-Construal (ISC) .19 *
Interdependent Self-Construal .35 * .27 *

* p < .01