Measuring Individual and Cultural Differences in Implicit Trait Theories

Please download to get full document.

View again

of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report
Category:

Health & Medicine

Published:

Views: 0 | Pages: 9

Extension: PDF | Download: 0

Share
Description
Journal of Personality and Social Psychology Copyright 2003 by the American Psychological Association, Inc. 2003, Vol. 85, No. 2, /03/$12.00 DOI: / Measuring Individual
Transcript
Journal of Personality and Social Psychology Copyright 2003 by the American Psychological Association, Inc. 2003, Vol. 85, No. 2, /03/$12.00 DOI: / Measuring Individual and Cultural Differences in Implicit Trait Theories A. Timothy Church, Fernando A. Ortiz, Marcia S. Katigbak, Tatyana V. Avdeyeva, and Alice M. Emerson Washington State University José de Jesús Vargas Flores and Joselina Ibáñez Reyes National Autonomous University of Mexico A new measure of implicit theories or beliefs regarding the traitedness versus contextuality of behavior was developed and tested across cultures. In Studies 1 (N 266) and 2 (N 266), these implicit beliefs dimensions were reliably measured and replicated across U.S. college student samples and validity evidence was provided. In Study 3, their structure replicated well across an individualistic culture (the United States; N 249) and a collectivistic culture (Mexico; N 268). Implicit trait and contextual beliefs overlapped only modestly with implicit entity theory beliefs and were predicted by self-construals in ways that generally supported cultural psychology hypotheses. Implicit trait beliefs were fairly strongly endorsed in both cultures, suggesting that such beliefs may be universally held. Current research on personality and culture is dominated by two theoretical perspectives, the trait perspective and the cultural psychology perspective. Cross-cultural trait psychologists argue that traits provide a meaningful basis for understanding and predicting behavior in all cultures and point to cogent theoretical perspectives and empirical findings that are consistent with the existence of universal trait dimensions (Church, 2000; McCrae, 2000). At the same time, some cultural psychologists have questioned the utility of the trait concept, at least for individuals in more collectivistic cultures, where the contextual nature of behavior is emphasized (Markus & Kitayama, 1998; Shweder, 1991). In short, these two perspectives raise a fundamental question: How important are personality traits versus contextual factors in understanding persons and their behavior in various cultures? In the present studies, we did not address the actual traitedness or contextuality of behavior. Rather, we report on the development and cross-cultural generalizability of a new measure of implicit beliefs about the traitedness versus contextuality of behavior. Why Measure Implicit Trait Theories Implicit theories, lay beliefs, or naïve psychologies play an important role in individuals explanations and predictions of A. Timothy Church, Fernando A. Ortiz, Marcia S. Katigbak, Tatyana V. Avdeyeva, and Alice M. Emerson, Department of Educational Leadership and Counseling Psychology, Washington State University; José de Jesús Vargas Flores and Joselina Ibáñez Reyes, Iztacala National School of Professional Studies, National Autonomous University of Mexico, Mexico City, Mexico. Portions of this article were presented at the XXV International Congress of Applied Psychology, Singapore, July The research was supported by National Institute of Mental Health Grant R We thank Dennis Bautista for assistance in data collection. Correspondence concerning this article should be addressed to A. Timothy Church, Department of Educational Leadership and Counseling Psychology, Cleveland Hall, Washington State University, Pullman, Washington behavior (Dweck, 2000; Norenzayan, Choi, & Nisbett, 2002). In addition, a number of psychologists have hypothesized that the cultural differences found in studies of self-concepts, dispositional inference, and self-enhancement biases may be the result of cultural differences in implicit theories. People in individualistic cultures who are thought to construe persons as independent agents (Markus & Kitayama, 1998; Triandis, 1995) are hypothesized to have stronger implicit beliefs regarding the traitedness of behavior. This, in turn, leads to stronger emphases on (a) trait attributes as an aspect of self-concept, (b) trait inference in the observation of others, and (c) self-enhancement in the evaluation of one s own traits (e.g., Heine, Lehman, Markus, & Kitayama, 1999; Kitayama, Markus, Matsumoto, & Norasakkunkit, 1997; Rhee, Uleman, Lee, & Roman, 1995). In contrast, people in collectivistic cultures who are thought to construe persons as interdependent and connected to in-groups are hypothesized to have weaker beliefs regarding the traitedness of behavior and stronger beliefs regarding the role of contextual factors. This, in turn, leads to (a) greater focus on roles and relationships as aspects of self-concept, (b) greater attention to situational information in causal inferences about behavior, and (c) weaker tendencies to self-enhance (and perhaps even tendencies to self-criticize) in self-descriptions. Attempts to test these hypotheses will require instruments that measure individual and cultural differences in implicit theories regarding the traitedness versus contextuality of behavior, which we refer to here as implicit trait versus implicit contextual theories. Implicit trait theories may encompass at least the following component beliefs: (a) belief in the longitudinal stability of personality traits; (b) belief in the cross-situational consistency of trait-relevant behavior; (c) belief in the ability to predict individuals behavior from their traits; (d) the belief that traits can be readily inferred from relatively few behavioral instances; and (e) the belief that people can be accurately described and understood in terms of their traits. These component beliefs are suggested by theorists conceptions of lay dispositionism (e.g., Ross & Nisbett, 1991), empirical studies (Chiu, Hong, & Dweck, 1997; Dweck, 2000; Norenzayan et al., 2002), and logical considerations, which 332 MEASURING IMPLICIT TRAIT THEORIES 333 suggest that these component beliefs will tend to coexist in the same individuals. For example, believing that traits can predict or be inferred from behavior would seem to depend on beliefs in the longitudinal and cross-situational stability of traits. In contrast, when cultural psychologists refer to the contextual determinants of self or behavior, they refer to such factors as situations, roles, or the larger social context (Norenzayan et al., 2002, p. 110) or to social structures and interpersonal frameworks such as families, work groups, social roles, positions, or relationships (Markus & Kitayama, 1998, p. 70). Trait psychologists, who refer to transcontextual consistencies across time, situations, and social roles, have a similar view of what constitutes context (e.g., McCrae & Costa, 1984; Sheldon, Ryan, Rawsthorne, & Ilardi, 1997). Such conceptions of contextuality lead us to delineate and assess the following contextual beliefs, which contrast with each of the five implicit trait belief components: (a) longitudinal or temporal instability of traits; (b) cross-situational variability; (c) lack of predictive validity; (d) difficulty of trait inference; and (e) the perceived greater importance of contextual factors such as roles, statuses, and relationships, as compared with traits, in person description. We examined many implicit theory constructs and measures to see if they might assess these component beliefs, but only two were deemed directly relevant. Norenzayan et al. (2002) used paragraph descriptions of dispositional, situationist, and interactionist lay theories, which appear promising. However, these paragraphs combine or blend components of lay dispositionism or situationism into single descriptions rather than measuring the components separately. Thus, the authors assumed rather than empirically tested whether the component beliefs co-occur in the same individuals. Potentially most relevant, and thus warranting a more extensive comparison, is Dweck s (2000) implicit entity theory construct. Implicit Trait Versus Implicit Entity Theories Dweck and her colleagues have focused on implicit theories about whether attributes are fixed (the view of entity theorists) or malleable (the view of incremental theorists). Recent studies have examined these theories in the domains of morality and personality (for reviews, see Dweck, 2000; Levy & Dweck, 1998). These researchers have reasoned that individuals who believe in fixed traits will expect a high degree of consistency in trait-relevant behavior; thus, traits will be seen to have predictive value and will be readily inferred from sparse information about behavior. In contrast, incremental theorists, who view traits as more dynamic and malleable, will expect behavior to be more variable across time and situations; thus, traits will be seen as less predictive of behavior and less crucial or possible to infer from behavior. There is some resemblance between Dweck s (2000) entity theory construct and our implicit trait theory construct, but we also note some differences. First, Dweck s construct addresses directly only the longitudinal stability component of implicit trait beliefs (i.e., fixed vs. malleable traits). Thus, our implicit trait theory construct should lead to a more comprehensive assessment of the multiple aspects of lay dispositionism. Second, entity theorists apparently take a more rigid stance on the possibility of longitudinal change than most trait theorists would. For example, as stated by Dweck (2000, p. 88), simply put, entity theorists don t grant people the potential to grow. Indeed, the entity theory items (e.g., Everyone is a certain kind of person and there is not much that can be done to really change that. ) suggest a degree of inflexibility in those who strongly agree with them. This might account, in part, for some of the negative behavioral correlates that have been linked to entity theorists, such as tendencies to be more judgmental, punitive, and prone to social stereotyping than incremental theorists (e.g., Chiu, Dweck, Tong, & Fu, 1997; Hong, Chiu, Dweck, & Sacks, 1997; Levy, Stroessner, & Dweck, 1998). In contrast, in our conception of implicit trait theories, the longitudinal component is presented as more probabilistic than deterministic (e.g., People who are friendlier than others now will probably remain friendlier than others in the future as well ). Trait psychologists do not view traits as fixed or nonmalleable, but only as relatively stable over time, at least as reflected in the rank ordering of individuals. Finally, the entity theory construct has not been applied much across cultures, and when it has, the results have sometimes been inconsistent with the expectations of cultural psychologists. If cultural psychologists are correct that people focus more on traits in individualistic cultures than in collectivistic cultures, entity theorists should be more prevalent in individualistic cultures. Indeed, Norenzayan et al. (2002) found this to be the case in a comparison of American and Korean samples. However, neither Chiu, Hong, and Dweck (1997, Study 4) nor Chiu, Dweck, et al. (1997) found this to be true in comparisons of Hong Kong and American samples. In sum, although we do not question the value of the entity theory construct, nor the impressive empirical findings associated with it (Dweck, 2000), we determined that it might not capture the implicit theory distinction we were interested in, particularly given our focus on testing cross-cultural hypotheses. Overview of the Present Studies Like Levy et al. (1998), Norenzayan et al. (2002), and others, we assume that implicit theories can be directly assessed using selfreport. Although implicit beliefs may not be at the forefront of everyday consciousness, nor commonly articulated spontaneously, we assume that individuals can meaningfully endorse statements that do articulate such beliefs. We conducted three studies. In Studies 1 and 2, we (a) tried out items that assess the hypothesized components of implicit trait (and contextual) beliefs in U.S. samples; (b) investigated the structure of the implicit trait theory domain; and (c) provided initial validity evidence for the measure. In Study 3, we examined the structural equivalence of the measure across an individualistic culture (United States) and a collectivistic culture (Mexico) and tested hypotheses relating implicit theories to independent and interdependent self-construals. In Hofstede s (1983) value-based ranking of 50 cultures along the individualism collectivism continuum, the United States ranked first and Mexico ranked 31st. Many other researchers have described Mexican culture as being collectivistic, with its emphases on close family ties, long-term commitments and reciprocal obligations among friends, affiliative obedience, and interdependence (e.g., Diaz-Loving & Draguns, 1999). In each of our studies, we included an entity theory measure to determine the extent of overlap between the implicit entity and implicit trait theory constructs. Our specific hypotheses were the following: 334 CHURCH ET AL. Hypothesis 1: A coherent dimension (or dimensions) of individual differences contrasting those who believe that behavior is determined more by traits (implicit trait theorists) versus context (implicit contextual theorists) exists in all cultures. Hypothesis 2: The following belief components of implicit trait theorists will cohere on a single dimension (e.g., in factor analysis): (a) belief in the longitudinal stability of traits; (b) belief in the cross-situational consistency of trait-relevant behavior; (c) a belief in the ability to predict individuals behavior from their traits; (d) the belief that traits can be inferred from relatively few behavioral instances; and (e) the belief that people can be accurately described and understood in terms of their personality traits. Hypothesis 3: Implicit trait beliefs will be at most moderately associated with implicit entity theory beliefs. Hypothesis 4: In all cultures, persons with more independent self-construals have stronger beliefs about the traitedness of behavior (i.e., tend to be implicit trait theorists), whereas persons with more interdependent self-construals have weaker beliefs about the traitedness of behavior (i.e., tend to be implicit contextual theorists). Hypothesis 5: Persons in individualistic cultures have stronger implicit trait beliefs than do persons in collectivistic cultures. Study 1: Item Development and Structure In Study 1 we tried out items that assess the hypothesized five components of implicit trait theory beliefs, obtained an initial look at the structure of this domain, and examined the relationship between implicit trait and entity theories. Sample Method A total of 266 college students (74 men, 192 women) at Washington State University (n 242) and Chaminade University in Hawaii (n 24) provided complete data. Mean age was 22.6 years (SD 4.98). Students from all year levels (i.e., freshman, sophomore, junior, senior) and a variety of major fields of study were sampled. Ethnic background was as follows: White/Caucasian (n 197), Asian/Pacific Islander (n 25), Latino (n 8), African American (n 3), Native American (n 2), multiracial (n 20), other (n 4), and not reporting (n 7). We eliminated 2 participants who rated their ability to read an English questionnaire as poor, leaving 264 participants. Instruments Personality Beliefs Inventory (PBI). Items were written to assess trait versus contextual beliefs for each of the five belief components, which we refer to as longitudinal stability, cross-situational consistency, predictive validity, trait inference, and general understanding components. Some items refer to personality traits in general, whereas others refer to specific traits. Because implicit beliefs might vary for different traits (Gidron, Koehler, & Tversky, 1993; Rothbart & Park, 1986), we selected both positive and negative traits from each of the Big Five domains (Goldberg, 1990). Similarly, some contextual items refer to situations or longitudinal change in general, whereas others refer to specific situations, roles, statuses, and relationships, consistent with the conception of contextuality described earlier. Representative items for each component are shown in the Appendix. For Study 1, we tried out items for each component, with a balanced number of implicit trait and contextual items. Participants responded to the items using a 6-point bipolar scale of agreement: strongly disagree, somewhat disagree, slightly disagree, slightly agree, somewhat agree, and strongly agree. Entity theory measure. We administered a five-item version of Dweck s (2000) domain-general entity theory measure. Three items measure entity beliefs (e.g., The kind of person someone is, is something very basic about them and it can t be changed very much ); two reverse-keyed items measure incremental beliefs (e.g., Everyone, no matter who they are, can significantly change their basic characteristics ). We modified the verbal anchors of Dweck s 6-point agreement scale to the 6-point scale used with the PBI. Using principal-axis factor analysis, we confirmed that a single bipolar dimension provided an adequate representation of the scale s structure; all five items loaded between.67 and.75 on the first factor. Dweck and her colleagues have reported alpha reliability values ranging from.70 to the.90s and test retest reliability values in the.80s (e.g., Dweck, Chiu, & Hong, 1995; Levy et al., 1998). Alpha reliability in the present sample was.85. Procedure Student volunteers were recruited in classes, completed the two questionnaires on their own time, and returned them to class for pick-up by the researchers. Students completed the PBI, then the entity theory measure, and received extra credit for participation. Results Structure of Implicit Trait Beliefs We factor analyzed the PBI item pool using principal-axis solutions with oblique (oblimin) rotations. The pattern of eigenvalues suggested that up to five factors might be meaningful (the first 10 eigenvalues were 12.27, 6.61, 5.10, 4.07, 3.16, 2.34, 2.28, 2.19, 2.00, and 1.94). We first examined the one-factor solution to determine if a single bipolar dimension contrasting implicit trait and contextual beliefs would emerge. The factor was a general trait beliefs factor and was only weakly bipolar; most of the contextual beliefs items had negative loadings only in the.10 to.25 range. In the two-factor solution, the two factors were interpretable as general trait and contextual beliefs dimensions. Forty-seven trait beliefs items, representing all five belief components, had loadings of.30 or higher on the first factor. Thirty-four contextual beliefs items, again representing all five belief components, had loadings of.30 or higher on the second factor. The two factors were essentially orthogonal (r.06). In the three-, four-, and fivefactor solutions, the number of additional items with high loadings was small. The third factor was defined primarily by the trait beliefs items from the general understanding component, which split off from the general trait beliefs factor. The fourth and fifth factors were not very interpretable. 1 1 In two of the three studies in this article, we compared factor pattern matrices obtained with the total U.S. sample and with the more ethnically homogeneous White/European American subsample. Congruence coefficients (Tucker, 1951) between matched factors all exceeded.97. That is, inclusion of the small number of ethnic minority participants did not affect the factor structure of the PBI items. Therefore, only the factor solutions for the total U.S. samples are discussed in each study. MEASURING IMPLICIT TRAIT THEORIES 335 Relating Implicit Trait and Entity Theories We correlated regression-method factor scores for the implicit trait and contextual beliefs dimensions with the entity theory measure. Estimated alpha reliabilities for the trait and contextual beliefs dimensions were.92 and.85, respectively. We found modest tenden
Recommended
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x