The importance of traits , group memberships , and relationships

An experiment was conducted to replicate and extend previous findings that traits are considered to be more important than group memberships (Trafimow & Finlay, 2001). In the present experiment, relationships were evaluated as well as traits and group membership. Participants listed five of their traits, five of their group memberships, and five of their relationships. They then ranked (relative to each other) and rated (in absolute terms) the importance of these traits, groups, and relationships. Consistent with Trafimow and Finlay (2001), traits were considered more important than group memberships. A new finding was that relationships were also considered more important than group memberships. Further, the data also indicated that, whereas the perceived importance of traits and groups is only weakly related to their cognitive accessibility, the perceived importance of relationships is significantly more related to their accessibility. Taken together, the findings suggest that even in individualist cultures, people think their relationships with others are more important than the literature on individualism might suggest. © 2003 Elsevier Inc. All rights reserved. There is a considerable amount of evidence that people have at least three cognitive structures where self-cognitions are stored. These three cognitive structures are the private self (where cognitions about traits, states, and behaviors are stored), the collective self (where cognitions about one's groups are stored), and the relational self (where cognitions about one's relationships are stored). Support for these three selves has been obtained with a variety of methods including priming studies (Brewer & Gardner, 1996; Finlay & Trafimow, 1998; Gardner, Gabriel, & Lee, 1999; Kuihnen & Hannover, 2000; Kiihnen, Hannover, & Schubert, 2001; Trafimow, Silverman, Fan, & Law, 1997; Trafimow & Smith, 1997; Trafimow, Triandis, & Goto, 1991; Ybarra & Trafimow, 1998), conditional probability analyses (Trafimow & * Corresponding author. Tel.: +1-505-646-4023; fax: +1-505-646-6212. E-mail address: [email protected] (D. Trafimow). 0362-3319/$ see front matter © 2003 Elsevier Inc. All rights reserved. doi: 10.1016/S0362-3319(03)00041-7 . .. 1. 448 D. Trafimowg L Madson/Thie Social Science Journal 40 (2003) 447-457 Finlay, 2001; Trafimow & Smith, 1997; Trafimow et al., 1991, 1997), factor analyses (Cheek & Briggs, 1982; Singelis, 1994), and others (Hazan & Shaver, 1994; Kashima et al., 1995; Reis & Shaver, 1988). Unfortunately, the fact that people have private, collective, and relational selves provides little information about which self-structures they consider to be of greater or lesser importance relative to each other. Given the voluminous literature testifying to the importance of the "self' (or at least to researchers' beliefs in its importance), it seems important to know exactly what aspects of the self, that is, which kinds of self-cognitions, are of particular importance. The main goal of the present article is to investigate this issue. To our knowledge, there is only one study that explores the relative importance of different types of self-cognitions. Trafimow and Finlay (2001) asked Mexican-Americans and Anglo-Americans to list five private self-cognitions (cognitions about one's traits, states, or behaviors) and five collective self-cognitions (cognitions about one's group memberships). Participants later rated the importance of their cognitions on an absolute scale from 0 to 99, and ranked the relative importance of these cognitions. Trafimow and Finlay (2001) found that private self-cognitions were rated and ranked as more important than collective self-cognitions. They also found that, as participants went from the most to the least important cognitions, there was a much steeper drop in rated and ranked importance of collective self-cognitions than private self-cognitions. For example, the most important private self-cognition was given a mean rating of 94.59, and the least important private self-cognition was given a mean rating of 47.75, for a difference of 46.85. In contrast, the corresponding means for collective self-cognitions were 91.20 and 31.85, for a difference of 59.35. A similar pattern was obtained in the ranking data in that the decrease for private self-cognitions was less steep (e.g., 47.75) than for collective self-cognitions (e.g., 59.35). Thus, these researchers concluded that whereas people have several private self-cognitions that they consider important, they have fewer (only two or three at most) collective self-cognitions that they consider important. There were two additional findings of interest. First, the pattern described in the previous paragraph was more pronounced for the Anglo-Americans than Mexican-Americans. Mostly, this is because the decrease in importance of collective self-cognitions was steeper for Anglo-Americans than for Mexican-Americans (mean differences in ratings from most important to least important self-cognitions were 68.05 and 50.65 for Anglo-Americans and Mexican-Americans, respectively). Second, there was a significant, but small, correlation between the accessibility of collective self-cognitions and their rated importance (the correlation for private self-cognitions was not statistically significant). Assuming that there are three types of self-cognitions, Trafimow and Finlay's (2001) findings are limited in a crucial way because of their failure to test the importance of relational self-cognitions (cognitions about one's relationships). Different perspectives suggest different predictions about the importance of relational self-cognitions. Much of the literature on individualism/collectivism assumes that people in individualist cultures (like the U.S.) think more about their own traits and characteristics than they think about their relationships with other people. From this perspective, we would predict that relational self-cognitions should be considered less important than private self-cognitions (see Triandis & Trafimow, 2001a, 2001b for recent reviews of the literature). This "self-centered" perspective does not make a prediction about the relative importance of relational versus collective self-cognitions. I. ................