Within-individual variability in self-concepts and personality states: Applying density distribution and situation-behavior approaches across cultures
Introduction
Evidence of behavioral variability led some psychologists to question the existence of traits and emphasize the situational determinants of behavior (Mischel, 1968, Ross and Nisbett, 1991). More recently, however, within-individual variability has featured prominently in efforts to synthesize alternative perspectives in the person-situation debate (Fleeson and Noftle, 2008, Fournier et al., 2008, Mischel and Shoda, 1998). This synthesis has arisen from the recognition that both trait and situationist perspectives are correct. Consistent with situationist perspectives, specific behaviors exhibit substantial within-individual variability. Once aggregated, however, consistent individual differences in behavioral tendencies are evident and indicative of stable traits (Epstein, 1979, Fleeson, 2001, Fournier et al., 2009, La Guardia et al., 2000). A limitation of this research is that it has been conducted largely in Western countries. The extent of within-individual variability may differ across cultures. In the two studies reported here, we examined within-individual variability in seven diverse cultures. We applied two theoretical approaches that have figured prominently in recent efforts to incorporate consistency and within-individual variability into integrated frameworks, the density distributions approach (Fleeson, 2001) and situation-behavior profiles (Mischel, Shoda, & Mendoza-Denton, 2002).
In Fleeson’s (2001) density distributions approach, traits are conceptualized as distributions of personality states, where personality states refer to momentary enactments of the relevant trait. Trait-related consistency is reflected in the stable means of the distributions (e.g., average levels of extraversion states), while within-individual variability is revealed in the considerable variability around individuals’ means. For example, using experience sampling methods, Fleeson (2001) showed that the amount of within-individual variability in Big Five personality states was about the same as between-person variation and nearly as great as within-individual variability in affect. In subsequent studies, Fleeson and colleagues (Fleeson, 2007, Fleeson and Leicht, 2006) showed that within-individual variability in personality states can be predicted from psychologically-active situational attributes. These studies are important because they show that within-individual variability is meaningful and systematic, not random or due to error (see also La Guardia et al., 2000, Ryan et al., 2005).
In the present study, we drew on Self-Determination Theory (SDT; Deci & Ryan, 2000) to identify attributes of everyday situations that might underlie within-individual variability in personality and affective states. SDT predicts that people express their traits differently in various situations as a function of how much their needs for autonomy, competence, and relatedness are satisfied in the situations (La Guardia & Ryan, 2007). The needs associated with SDT are of particular interest because SDT theorists have proposed that they are universally important across all cultures (Chirkov et al., 2005, Levesque et al., 2004). Indeed, in eight cultures, Church et al. (2012) found that need satisfaction in various roles partially accounted for cross-role variability in Big Five traits. Similarly, in three cultures, Lynch, La Guardia, and Ryan (2009) found that perceived autonomy-support accounted for within-individual variability across relationships in ideal-actual self-concept discrepancies. However, the impact of autonomy support was somewhat stronger in the United States than in Russia or China, raising the possibility that the situational attributes underlying within-individual variability may differ in their explanatory power across cultures.
In the situation-behavior approach, within-individual variability is represented as stable “if–then” situation-behavior profiles, or behavioral signatures (Fournier et al., 2008, Furr, 2009, Mischel et al., 2002, Smith et al., 2009). In Cognitive–Affective Personality System (CAPS) theory (Mischel, 2004, Mischel and Shoda, 1995), individual differences in these situation-behavior profiles are attributed to differences in the chronic accessibility and organization of cognitive–affective units such as goals, beliefs, expectancies, and competencies (Fournier et al., 2008, Mischel and Shoda, 1995). A few studies in the United States and Canada have derived behavioral signatures in children or adults (Fournier et al., 2008, Shoda et al., 1994). For example, Smith et al. (2009) found stable and distinctive situation-behavior profiles for youth coaches’ supportive and instructional behaviors across three game situations (winning, losing, or in a close game). Although situation-behavior profiles in these studies were used to characterize within-individual variability in behavior, they can also be used to describe within-individual variability in self-concepts (e.g., trait ratings in different roles) (Church et al., 2012, English and Chen, 2007).
In the present studies, we applied both the density distribution and situation-behavior approaches to investigate within-individual variability in self-concepts and personality states across cultures. Consistent with the density distribution approach, we operationalized behavior in terms of the Big Five personality states reported in daily situations or interpersonal interactions (Fleeson, 2001, Fleeson and Wilt, 2010). As in previous mono-cultural studies, we expected to find substantial and reliable within-individual variability and that this variability can be represented as stable if–then profiles or signatures. At the same time, we tested theoretical proposals offered by cultural psychologists, who have suggested that within-individual variability may be greater in some cultures than others.
Cultural psychologists have proposed that within-individual variability may be greater in non-Western cultures. One theoretical perspective addresses the distinction between independent and interdependent self-construals, which are viewed as more prevalent in individualistic and collectivistic cultures, respectively (Markus and Kitayama, 1991, Suh, 2002). People with independent self-construals are expected to exhibit greater consistency because of their greater need to express their unique traits. In contrast, for people with interdependent self-construals, roles and relationships are expected to impact behavior more than traits, increasing within-individual variability (Heine, 2001, Markus and Kitayama, 1998). A second theoretical perspective attributes greater within-individual variability in East Asian cultures to dialecticism, a system of thought based in Eastern philosophical traditions and characterized by tolerance of contradiction, expectations of complexity and change, and holistic thinking (Peng & Nisbett, 1999). Because of their dialecticism, East Asians are thought to be more able than Westerners to store inconsistent information in their self-concepts and to behave in more variable ways across situations (Choi and Choi, 2002, Spencer-Rodgers et al., 2010). A third theoretical framework addresses the tightness versus looseness of a culture (Gelfand et al., 2006, Gelfand et al., 2011). In tight cultures, where social norms and situational constraints on behavior are stronger, greater within-individual variability in both self-concepts and behavior might be expected.
Several studies have compared within-individual variability in self-concepts across cultures by assessing variability in trait ratings across roles or relationships. The range of cultures investigated has been quite limited, however. Suh, 2002, Boucher, 2011, and English and Chen, 2007, English and Chen, 2011 found that Asians (or Asian Americans) exhibit greater variability in trait ratings than Americans (or European Americans). Cultural differences in within-individual variability in these studies were attributed to cultural differences in either self-construals or dialecticism. Church et al. (2012) and Church, Anderson-Harumi et al. (2008) compared a more diverse set of cultures. For example, in a study in eight cultures (United States, Australia, Mexico, Venezuela, Philippines, Malaysia, China, and Japan), Church et al. (2012) found that only Japanese exhibited reliably greater cross-role variability in trait ratings than Americans. Dialecticism, but neither self-construals nor cultural tightness, mediated some of the cultural differences.
While several cross-cultural studies have examined self-concept variability, only two have compared within-individual variability in daily behavior or affect. Oishi, Diener, Scollon, and Biswas-Diener (2004) found that Japanese, Hispanics, and Asian Indians exhibited greater variability in affect across situations than did a multiethnic sample of Americans in Illinois. In a daily process study, Church, Katigbak et al. (2008) found some tendency for Filipinos, as compared to Americans, to exhibit greater cross-situational variability in Big Five behaviors. Clearly, additional studies are needed to enable confident conclusions about cultural differences in within-individual variability in actual behavior. In addition, we could not identify any cross-cultural studies that examined situation-behavior profiles or signatures in daily behavior.
Finally, there is also a need for studies that examine the convergence of self-concept and behavioral variability in various cultures. Role identity theory (e.g., Wood & Roberts, 2006) proposes that role identities (i.e., self-perceived traits in various roles) accurately reflect, to some extent, one’s actual role experiences (e.g., trait-relevant behaviors) in the respective roles. Similarly, ecological-realist perspectives propose that the self-perceived traits that comprise an important part of our self-concepts can be accurately inferred from behavior if observed in the context of trait-relevant activities (Baron and Misovich, 1993, Funder, 1995). From both perspectives, we would expect that individuals who exhibit greater behavioral variability will also exhibit greater self-concept variability and that situation-behavior profiles derived in self-concepts and behavior will converge to some extent in each culture.
We examined within-individual variability in two cross-cultural studies. In Study 1, which was based on the density distributions approach, participants in the US, Venezuela, the Philippines, China, and Japan rated their personality states and their positive and negative affects three times a day for 20 days. In Study 2, we applied both the density distribution and situation-behavior approaches to examine within-individual variability in self-concepts, personality states, and affective states in the US, Mexico, Malaysia, China, and Japan. Participants rated their Big Five traits (i.e., self-concepts) in five specific roles, then rated their personality and affective states in multiple interpersonal interactions or roles each day for 14 days.
The data analyzed for Study 1 were part of a larger study that compared the manifestation of Big Five traits in personality states and affect across cultures (Ching et al., submitted for publication). The data analyzed for Study 2 were part of a larger study that tested the cross-cultural generalizability of the Personality and Role Identity Structural Model (PRISM; Wood & Roberts, 2006) (Ching et al., 2013). Neither of these studies focused on within-individual variability or the hypotheses addressed here. For these studies, we selected a diversity of cultures expected to vary along the dimensions of individualism–collectivism, dialecticism, and tightness. Based on previous findings, we classified our United States samples (e.g., in planned contrasts) as individualistic and our samples in Mexico, Venezuela, the Philippines, Malaysia, China, and Japan as relatively collectivistic (Church, 1987, Díaz-Loving and Draguns, 1999, Hofstede, 2001). We classified the Asian samples, particularly those in China and Japan, as dialectical and the three non-Asian samples (US, Mexico, and Venezuela) as non-dialectical (Peng and Nisbett, 1999, Spencer-Rodgers et al., 2010). Finally, we classified Malaysia, China, and Japan as high in tightness and the US, Mexico, Philippines, and Venezuela as relatively loose (Gelfand et al., 2011). For the Study 1 and Study 2 samples in the present article, Ching et al. (2013) and Ching et al. (submitted for publication) reported results of mean comparisons on relevant cultural measures that generally supported these classifications, although somewhat less so for the individualism–collectivism dimension.
In both Study 1 and Study 2, we expected to find substantial within-individual variability in personality states and that this variability would rival the amount of variability exhibited for affects (Hypothesis 1). As argued by Fleeson (2001), affect variability provides a good comparison standard for variability in personality states, because affect is generally viewed as quite variable (e.g., Csikszentmihalyi and Hunter, 2003, Eaton and Funder, 2001, Eid and Diener, 1999, Kuppens et al., 2010). In Study 1, we expected perceived satisfaction of SDT needs in various situations (as well as needs for self-actualization and pleasure-stimulation) to partially account for within-individual variability in personality states and affects (Hypothesis 2). In Study 2, we anticipated that individuals who exhibit more cross-role variability in their personality states will also tend to show greater cross-role variability in their self-concepts (Hypothesis 3). In addition, we expected to find reliable if–then profiles across roles in both self-concepts and personality states (Hypothesis 4) and that the self-concept and personality state profiles will exhibit some degree of convergence (Hypothesis 5). Although we expected these hypotheses to be supported in all cultures, we also examined whether there are cultural differences in the extent of within-individual variability and, if so, whether they correspond to cultural differences in individualism–collectivism, dialecticism, or tightness.
Section snippets
Study 1
In Study 1, participants in five cultures rated their Big Five personality states and positive and negative affect three times a day for 20 days. We examined the extent of within-individual variability in personality and affective states (Hypothesis 1), whether cultures differ in the amount of within-individual variability, and whether within-individual variability can be accounted for to some extent by perceived satisfaction of needs in various situations (Hypothesis 2).
Study 2
In Study 2, we obtained new samples from different universities in the United States, China, and Japan, and added two new cultures, Mexico and Malaysia. In planned contrasts, the US was again classified as individualistic and Mexico, Malaysia, China, and Japan were classified as collectivistic (Díaz-Loving and Draguns, 1999, Hofstede, 2001). Based on previous findings, we classified Mexico as relatively loose and Malaysia as tight (Church et al., 2012, Gelfand et al., 2011). As a result,
General discussion
Research on within-individual variability can facilitate efforts to integrate structure and process approaches in the study of personality (Fleeson, 2004, Fleeson and Noftle, 2008, Mischel and Shoda, 1998). Whereas structure approaches focus on individual differences in typical levels of traits, process approaches emphasize within-individual variability and the situational factors that can account for this variability. We applied both density distribution (Fleeson, 2001) and if–then
Acknowledgments
The research was supported by National Science Foundation Grant 0953940. We thank Larissa Plotsky for assistance in data collection and Congcong Wang and Ming Lei for assistance in translation.
References (74)
- et al.
On delineating and integrating the study of variability and stability in personality psychology: Interpersonal trust as illustration
Journal of Research in Personality
(2006) - et al.
The interpersonal signature
Journal of Research in Personality
(2009) - et al.
Situational similarity and behavioral consistency: Subjective, objective, variable-centered, and person-centered approaches
Journal of Research in Personality
(2004) - et al.
Behavioral signatures at the ballpark: Intraindividual consistency of adults’ situation-behavior patterns and their interpersonal consequences
Journal of Research in Personality
(2009) - et al.
Extraversion and its positive emotional core
- et al.
On the nature of intraindividual personality variability: Reliability, validity, and associations with well-being
Journal of Personality and Social Psychology
(2006) - et al.
Dispositional knowing from an ecological perspective
Personality and Social Psychology Bulletin
(1993) The dialectical self-concept II: Cross-role and within-role consistency, well-being, self-certainty, and authenticity
Journal of Cross-Cultural Psychology
(2011)- Ching, C. M., Church, A. T., Katigbak, M. S., Locke, K. D., Vargas-Flores, J. d., Ibáñez-Reyes, J., et al. (2013)....
- Ching, C. M., Church, A. T., Katigbak, M. S., Reyes, J.A. S., Tanaka-Matsumi, J., Zhang, H., et al. (2013). The...