The evaluative face of a descriptive model: Communion and agency in Peabody s tetradic model of trait organization

Transcription

1 European Journal of Social Psychology Eur. J. Soc. Psychol. 38, (2008) Published online in Wiley InterScience ( The evaluative face of a descriptive model: Communion and agency in Peabody s tetradic model of trait organization GUIDO PEETERS* Catholic University of Leuven, Belgium Abstract An important strand of research on cognitive trait organization is connected with Peabody s (1967) tetradic model stressing two non-evaluative dimensions: tight loose and assertive unassertive. The relationship of this model with the communion/agency model involving two evaluative dimensions is discussed and empirically investigated. It is shown that the models are orthogonally related suggesting four trait categories: agentic tight loose, communal tight loose, agentic assertive unassertive and communal assertive unassertive. Implications are discussed regarding the interpretation of stereotype contents, the validity of the tetradic and communion/agency models and the development of a functionalist approach of social perception and judgment. Copyright # 2008 John Wiley & Sons, Ltd. It is evident from the present special issue and literature reviews (e.g. Abele & Wojciszke, 2007) that communion (warmth) and agency (competence) represent fundamental social-cognitive dimensions that are representative of a variety of analogous two-dimensional models. An outstanding feature of these models concerns the evaluative character of the dimensions, each dimension contrasting a positive good pole with a negative bad pole. However, Peabody (1967, 1985) advanced an alternative non-evaluative model. Perceived traits and stereotypes would be organized consistent with two descriptive basic dimensions labelled tight loose and assertiveness (assertive unassertive). Extreme descriptive values are negative, moderate values are positive. Hence each dimension is represented as a descriptive continuum marked by sets of four traits (tetrads) displayed in Tables 1 and 2. For instance, the tetrad stingy thrifty generous extravagant represents descriptive values ranging from extreme tightness to extreme looseness. Thrifty and generous mark moderate optimal values, stingy and extravagant mark extremes evaluated as too tight and too loose. Consistent with this Aristotelian association of positivity with moderation and negativity with extremity, Peabody (1967) found that, within tetrads, more contrast was experienced between negative traits (stingy, extravagant) than between positive traits (thrifty, generous). The tetradic model was advanced as a social-judgmental model (Peabody, 1967) and used as a stereotype content model (Peabody, 1985). Nevertheless, while meeting ample response from personality psychologists, the model was largely ignored by social psychologists. Even when dealing with stereotype contents, social psychologists preferred the evaluative communion/agency model, presumably because it was more suited to establish links with other psychological realms such as social motivation and emotion (e.g. Cuddy, Fiske, & Glick, 2007; Peeters, 1993; Phalet & Poppe, 1997). In addition, the distinction between communion-like and agency-like dimensions has been useful to define specific evaluative contents involved in various social psychological processes such as negativity effects in social judgment (Peeters & Czapinski, 1990), and attraction elicited by attitude similarity (Singh, Ho, Tan, & Bell, 2007). However, also the tetradic model holds trumps. According to Peabody (1985) it would enable to grasp the kernel of truth stereotypes may contain. If a nation is characterized as thrifty by its own citizens, and as stingy by its neighbours, then tightness may be advanced as a national characteristic reflecting a veridical aspect of the nation s culture. But even if one would not accept the kernel of *Correspondence to: Guido Peeters, Center for Social and Cultural Psychology, Catholic University of Leuven, K.U. Leuven, Tiensestraat 102, B-3000 Leuven, Belgium. guido.peeters@psy.kuleuven.be Copyright # 2008 John Wiley & Sons, Ltd. Received 11 September 2007 Accepted 28 February 2008

2 Evaluative and descriptive dimensions 1067 Table 1. Traits representing the descriptive dimension tight loose, expected and obtained OP (communion) and SP (agency) values and outcomes of contrast analyses testing the agreement between expected and obtained contrasts Too tight Optimally tight Optimally loose Too loose Stingy Thrifty Generous Extravagant Inhibited Self-controlled Spontaneous Impulsive Grim Serious Gay Frivolous Distrustful Sceptical Trusting Gullible Severe Firm Lenient Lax Inflexible Persistent Flexible Vacillating Choosy Selective Broad-minded Undiscriminating OP (communion) Expected: S 0 R 0 Obtained: S1.43 þ0.43 R1.71 S0.29 SP (agency) Expected: 0 R 0 S Obtained: þ0.36 R0.68 þ0.21 S0.93 OP (communion) contrast: F(1, 24) ¼ 67.56, p <.0001, MSE ¼ 0.52; SP (agency) contrast: F(1, 24) ¼ 06.84, p <.016, MSE ¼ Table 2. Traits representing the descriptive dimension assertive unassertive, expected and obtained OP (communion) and SP (agency) values and outcomes of contrast analyses testing the agreement between expected and obtained contrasts Too assertive Optimally assertive Optimally unassertive Too unassertive Aggressive Forceful Peaceful Passive Conceited Self-confident Modest Unassured OP (communion) Expected: S 0 R 0 Obtained: 1.88 þ0.38 R1.75 S0.38 SP (agency) Expected: 0 R 0 S Obtained: þ1.12 R2.62 S0.62 S2.50 OP (communion) contrast: F(1, 4) ¼ 13.14, p <.0015 MSE ¼ 0.21; SP (agency) contrast: F(1, 4) ¼ , p <.0001, MSE ¼ truth hypothesis, the tetradic model remains a challenge to communion/agency-like models that involve dimensions contrasting positive with negative poles. Hence it is worthwhile to investigate how the models are related. Would there be a relationship anyway, and if there is, would it be a one-to-one relationship between analogous dimensions? In search for an answer we should first have a closer look at how tight loose and assertiveness were established as independent descriptive dimensions. Peabody proceeded in the usual way. Social stimuli were rated against bipolar trait scales with descriptive and evaluative meanings simultaneously in opposition. Hence each tetrad yielded two scales, one contrasting the second with the fourth column of Tables 1 and 2 (e.g. thrifty extravagant), another contrasting the third with the first column (e.g. generous stingy). All scales were oriented the same way ranging from 7 (positive pole) to 1 (negative pole). Factor analysis of correlations between scales revealed the present tight loose and assertiveness dimensions in that scales based on tetrads from Table 1 did not load on the same factor as scales based on tetrads from Table 2. The non-evaluative nature of the dimensions was revealed in that within factors evaluatively equivalent scales did not carry equivalent loadings but opposite loadings. For instance, thrifty extravagant was opposed to generous stingy, the former loading positively, the latter negatively on the tight loose factor. In this way scales contrasting positive optimally tight traits with negative too loose traits clustered together and were opposed to scales contrasting positive optimally loose traits with negative too tight traits. Similarly scales contrasting positive optimally unassertive traits with negative too assertive traits were opposed to scales contrasting positive optimally assertive traits with negative too unassertive traits.

3 1068 Guido Peeters Besides demonstrating the descriptive nature of dimensions, the abovementioned results revealed four categories of evaluative trait opposites: (1) optimally tight versus too loose, (2) optimally loose versus too tight, (3) optimally assertive versus too unassertive and (4) optimally unassertive versus too assertive. Intuitive appraisal of representative traits as well as comparison with the communion/agency literature suggests particular links of those categories with the communion/ agency model. For instance, within tight loose tetrads (Table 1), the optimally tight versus too loose contrast is represented by, among others, the trait opposites serious and frivolous that are representative of the agentic dimension intellectual evaluation (Rosenberg, Nelson, & Vivekanathan, 1968). Hence optimally tight versus too loose traits may be representative of agency, and if they are, the alternative optimally loose versus too tight contrast may fit the alternative communion dimension. As to the assertiveness tetrads (Table 2), modest and conceited (or vain) are not only representative of the optimally unassertive versus too assertive contrast but also of the communal dimension social evaluation (Rosenberg et al., 1968). Hence optimally unassertive versus too assertive traits may be representative of communion, leaving the representation of agency to contrasting optimally assertive versus too unassertive traits. The proposed links between tetradic and communion/agency models boil down to the expected þ/ contrasts displayed in Tables 1 and 2. It is worth mentioning that the independence of communion and agency is preserved the correspondent contrast weights being unrelated. Peabody (1970) did not fail to notice the affinities with Rosenberg et al. s (1968) variant of the communion/agency model. However, even after critical appraisals by De Boeck (1978) and Rosenberg and Olshan (1970) had demonstrated the psychological relevance of evaluation, Peabody and his colleagues did not switch to a communion/agency-like model. De Boeck (1978) had demonstrated that Procrustes rotation enabled to establish a general evaluative trait dimension contrasting all positive with all negative traits, indeed, but that rotation did not affect the descriptive tight loose and assertiveness dimensions. Hence isolating evaluation as a separate additional dimension became current practice in Peabody s studies (e.g. Peabody, 1978, 1984; Peabody & Goldberg, 1989; Saucier, Ostendorf, & Peabody, 2001). Nevertheless, also in those studies communion/agency-like solutions were obtained when varimax instead of Procrustes rotations were used. However those solutions were rejected arguing that evaluative and descriptive aspects were confounded. Confounded dimensions would be hard to interpret and lack the conceptual clarity required for sound theory. Moreover this lack of rigor seemed to have revealed itself in the proliferation of communion/agency-like models across the literature. However, one could object that also uncounfounded pure evaluation is neither clear nor easy to interpret. Linguistic analyses (Fillenbaum & Rapoport, 1971) as well as thorough examinations of semantic differential studies (Peeters, 1986) have revealed the fickle character of the sort of encompassing pure evaluation involved in De Boeck s (1978) evaluative dimension that ever since has been advanced by Peabody. In addition, the so-called confounded dimensions may meet the standards of conceptual clarity and unequivocal interpretability if they are not exclusively based on mathematical criteria such as amounts of variance explained and simple structure. The interpretations of mathematically grounded dimensions vary across studies because they are based on essentialist post-hoc readings of representative traits. In this way different interpretations may reflect different experimenters mindsets as well as different compositions of trait samples. However, dimensions can also be defined a priori as part and parcel of a conceptually clear functionalist theory. In this way Peeters (1976) defined two evaluative dimensions as a function of the perceived adaptive value of traits for respectively the possessors of the traits (self-profitability, (SP)) and for others in the environment of the possessors (other-profitability,(op)). Another functionalist theory defined two evaluative trait dimensions as a function of respectively perceived utility potential for society (social utility dimension) and desirability in interpersonal relationships (social desirability). The latter social-oriented theory seemed to complete the former individual-oriented theory, SP being associated with social utility and OP with social desirability (Beauvois & Dubois, 2000; Cambon, 2006; Dubois, 2003; Dubois & Beauvois, 2005). Above all, the dimensions of the functionalist approaches fit in well with the communion/agency model, communion being associated with OP, agency with SP (Abele & Wojciszke, 2007; Peeters, 1983). Hence it was decided to test the hypothesized contrasts displayed in Tables 1 and 2 by determining OP and SP values of the traits involved. DETERMINING OP AND SP A straightforward way to determine OP and SP values of traits may just require participants to rate in one way or another how advantageous or disadvantageous traits are for the possessors of the traits (SP) and for others in their environment

4 Evaluative and descriptive dimensions 1069 (OP). However, OP and SP values obtained in that way are positively correlated (Abele & Wojciszke, 2007; De Boeck & Claeys, 1988). Reviewing and discussing OP and SP values obtained with different methods, Peeters (1992) concluded that the positive correlation was substantially due to secondary SP assigned to a good number of OP traits, as if participants expected that being beneficial or harmful to others would boomerang to the self. Distortions due to secondary SP may be reduced when SP values are averaged across large sets of traits. However, they may be problematic in the present study where a small number of traits are involved. Hence an alternative method was used that has been described and proven valid in previous studies (Peeters, 1992, 2007). The method involved the use of key traits, being representative OP and SP traits selected from the above study of De Boeck and Claeys, A trait is a representative SP trait if it is rated manifestly advantageous (þsp) or disadvantageous ( SP) for the self while being rated neutral for others. Selected þsp key traits (translated from Dutch) are: independent, willing to achieve, clever, inquisitive, having understanding. SP key traits are: insecure, quickly disappointed, self-embroiled, distracted, uninspired. Four judges (K.U.Leuven PhD students) determined SP values of Peabody s traits (translated in Dutch) by rating on a 7-point scale how well each trait fitted either the SP cluster ( 3) or the þsp cluster (þ3), using the middle zero value to indicate that the trait did not fit the one cluster more than the other. The same judges determined the traits OP values using the following key traits, also selected from De Boeck and Claeys, 1988 in an analogous way as the SP key traits. þop: helpful, forgiving, fair, gentle, sympathizing. OP: heavily criticizing, not sympathizing, high-handedly cold, finding others worthless, keeping on distance with others. The ratings, averaged across judges, were very reliable Cronbach as amounting to 0.93 for SP and OP. In order to warrant generalizability of SP and OP values, the procedure was repeated using four other PhD students as judges and the following key traits that were no longer drawn from De Boeck & Claeys study but advanced by Peeters (1992) on the basis of several previous studies. þsp: powerful, ambitious, self-confident, practical, quick. SP: weak, ambitionless, shy, clumsy, slow. þop: tolerant, generous, sensitive, reliable, trusting. OP: intolerant, selfish, insensitive, unreliable, suspicious. This time a amounted to.86 for SP and.85 for OP. Correlations between replications (over N ¼ 36 traits) amounted to.94 for SP and.90 for OP. Apparently, replications involving different judges and different sets of key traits yielded highly similar outcomes. Hence SP and OP values of traits were averaged across replications. RESULTS AND DISCUSSION Mean SP and OP values are presented in Tables 1 and 2. The patterns of positive and negative values fit the expectations, predicted contrasts being significant for tight loose traits as well as for assertiveness traits. Tight and assertive are OP (negative communion) if extreme and þsp (positive agency) if optimal. Loose and unassertive are SP (negative agency) if extreme and þop (positive communion) if optimal. Practical Implications A main practical implication is that research outcomes mapped in one model can be translated into the other. For instance, Peabody (1985) found that European nations were perceived as quite assertive, which in the light of the present study means that the positive side of Europeans is perceived as þsp (positive agency), and the negative side as OP (negative communion). As to the tight loose dimension, the above pattern was confirmed for northern Europeans who were perceived as rather tight. However the pattern was attenuated for southern Europeans. The latter were perceived as loose, which means that a dominant feature of the southern stereotype involves þop (positive communion) on the positive side and SP (negative agency) on the negative side. As SP and OP match dimensions of McClintock s (1972) model of social motivational value orientations (Peeters, 1983, 1993), the northerners stereotype can be characterized as most individualistic, leaning towards competition. The southerners stereotype could be characterized as individualistic as well, but less individualistic than the northerners and leaning more towards cooperation. In terms of stereotype content types (Cuddy et al., 2007; Fiske, Cuddy, & Glick, 2002; Phalet & Poppe, 1997), this means that northern stereotypes reflect the winner type, specifically the sinful rather than the virtuous winner (Phalet & Poppe, 1997), and they may elicit admiration and envy (Cuddy et al., 2007; Fiske et al., 2002). The southern stereotype reflects the virtuous winner type

5 1070 Guido Peeters (Phalet & Poppe, 1997), though less a winner than the northern stereotype. It involves prejudices that, as far as they would deviate from admiration, reflect paternalism rather than envy (Cuddy et al., 2007). Implications for the Tetradic Model The results show that there is no one-to-one relationship between Peabody s descriptive dimensions and the evaluative OP-communion and SP-agency dimensions. Rather Peabody s dimensions are conglomerates of OP and SP. Consequently the non-evaluative character of his dimensions can be questioned. For instance, in order to ground the non-evaluative character of dimensions based on factor analysis, Peabody regarded a factor contrasting thrifty with generous as non-evaluative because adjectives with opposite loadings carried the same valence. However he ignored that one pole of the dimension was marked with SP, the other pole with OP. So he assumed that opposite OP and SP valences would neutralize each other. For instance, combining OP (e.g. malicious) with þsp (e.g. self-confident) would communicate evaluatively neutral assertiveness. Nevertheless, þsp combined with OP may involve cognitive interactions (Birnbaum, 1974; Peeters, 1992; Vonk, 1996; Wojciszke, Bazinska, & Jaworski, 1998). They imply that þsp can add to the negative value of OP making that malicious combined with self-confident is not neutral but even more negative than just malicious. It should be mentioned, however, that cognitive interactions sometimes fail to occur (e.g. De Bruin & Van Lange, 1999). The distinction between OP and SP dimensions then seems blurred both dimensions coinciding into a single encompassing evaluative dimension as the dimension De Boeck (1978) added to the non-evaluative dimensions of the tetradic model. Hence Peabody s claim that a dimension that does not oppose valences but SP to OP should be qualified as non-evaluative may hold in conditions where OP and SP fuse together into one encompassing evaluative dimension. For instance, one-dimensional evaluative processing has been observed when participants take the perspective of uninvolved observers who are asked not to describe their own impressions, but to estimate evaluative impressions formed by others (Peeters, 2007; Peeters, Cornelissen, & Pandelaere, 2003). The distinction between OP and SP persists in the own impressions of which Aderman (1969) has shown that they are not formed from the perspective of an uninvolved observer but from the perspective of an actor who anticipates contact. Implications for the Communion/Agency Model Extremity and Mixed Social-perceptual Contents One interesting implication concerns the role of extremity. Tables 1 and 2 show that negative traits ( SP and OP) represent descriptive extremes. It means that the negative valence of a trait is preserved when descriptive extremity increases, but turns into positive when descriptive extremity decreases and reaches the adjacent positive trait. However, the positive valence of the adjacent trait does not belong to the same agentic or communal dimension. Specifically, SP (extravagant, passive) turns into þop (generous, peaceful) and OP (stingy, aggressive) into þsp (thrifty, forceful). Alternatively, when positive traits become more extreme, they turn into negative traits belonging to the alternative agentic or communal dimension, þsp turning into OP and þop into SP. An important implication is that, as far as descriptive meaning components are involved, OP (negative communion) is closer to þsp (positive agency) than to SP (negative agency). The latter is closer to þop (positive communion) than to OP (negative communion). This pattern is consistent with particular findings on mixed stereotype contents that have been explained as a compensatory process motivated by a norm of justice (Judd, James-Hawkins, Yzerbyt, & Kashima, 2005). The present results suggest an alternative semantic explanation. Splitting Communion and Agency Another implication is that within SP-agency and OP-communion dimensions a distinction can be made between traits that are descriptively related to the tight loose dimension and traits that are descriptively related to the assertiveness dimension. In this respect it is worthwhile that further differentiation within OP-communion and SP-agency dimensions

6 Evaluative and descriptive dimensions 1071 has been proposed in order to advance the fit between the individual-oriented and social-oriented approaches described in the introduction (Peeters, 2004). For instance, the perceived social utility potential of SP may vary among SP traits. Particularly, tight loose related SP may involve more perceived social utility than assertiveness related SP. Indeed, having added more traits, Peabody (1984) found optimal tightness associated with competence-related traits that may represent a high social utility potential, for example capable, foresighted, hardworking, intelligent, organized, thorough. Optimal assertiveness, on the other hand, was associated with traits that seem more profitable for the individual than for society, for example individualistic, ambitious. In addition, a person s SP may only be perceived as utile for society if it is combined with a sort of OP that is not only expected to impact on relations with other individuals but also on society as a whole. A relevant OP distinction may be Trafimow and Trafimow s (1999) splitting of the communal morality dimension into traits reflecting Kantian perfect duties (honest, loyal) and imperfect duties (kind, charitable). The perfect duties may represent OP-communion with more societal relevance than the OP-communion represented by imperfect duties. However, after inspection of Tables 1 and 2, it is not obvious that Trafimow s distinction would reflect the distinction between tight loose related and assertiveness related OP. Moreover, in their confrontation with personality psychology Peabody and colleagues have advanced a distinction between agreeableness and integrity as separate dimensions that are analogous to Trafimow s imperfect and perfect duties, and those dimensions stand apart from tight loose and assertiveness (e.g. Peabody & De Raad, 2002). Apparently the split of OP-communion proposed to advance the integration of individual-oriented and social-oriented functionalist approaches seems not a matter of tight loose related and assertiveness related variants of OP-communion. Future research may sort out the issue. REFERENCES Abele, A. E., & Wojciszke, B. (2007). Agency and communion from the perspective of self versus others. Journal of Personality and Social Psychology, 93, Aderman, D. (1969). Effects of anticipating future interaction on the preference for balanced states. Journal of Personality and Social Psychology, 11, Beauvois, J.-L., & Dubois, N. (2000). Affordances in social judgment: Experimental proof of why it is a mistake to ignore how others behave towards a target and look solely at how the target behaves. Swiss Journal of Psychology, 59, Birnbaum, M. H. (1974). The nonadditivity of personality impressions. Journal of Experimental Psychology, 102, Cambon, L. (2006). Social desirability and social utility: Two evaluative dimensions of personality traits. Revue Internationale de Psychologie Sociale International Review of Social Psychology, 19, Cuddy, A. J. C., Fiske, S. T., & Glick, P. (2007). The BIAS map: Behaviors from intergroup affect and stereotypes. Journal of Personality and Social Psychology, 92, De Boeck, P. (1978). On the evaluative factor in the trait scales of Peabody s study of trait inferences. Journal of Personality and Social Psychology, 36, De Boeck, P., & Claeys, W. (1988). What do people tell us about their personality when they are freed from the personality inventory format.paper presented at the 4th European Conference on Personality, Stockholm, June De Bruin, E. N. M., & Van Lange, P. A. M. (1999). Impression formation and cooperative behavior. European Journal of Social Psychology, 29, Dubois N. (Ed.). (2003). A sociocognitive approach to social norms. London: Routledge. Dubois, N., & Beauvois, J.-L. (2005). Normativeness and individualism. European Journal of Social Psychology, 35, Fillenbaum, S., & Rapoport, A. (1971). Structures in the subjective lexicon. New York: Academic Press. Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, Judd, C., James-Hawkins, L., Yzerbyt, V., & Kashima, Y. (2005). Fundamental dimensions of social judgment: Understanding the relations between judgments of competence and warmth. Journal of Personality and Social Psychology, 89, McClintock, C. G. (1972). Social motivation a set of propositions. Behavioral Science, 17, Peabody, D. (1967). Trait inferences: Evaluative and descriptive aspects. Journal of Personality and Social Psychology Monographs, 7, (Whole No. 644). Peabody, D. (1970). Evaluative and descriptive aspects in personality perception: A reappraisal. Journal of Personality and Social Psychology, 16, Peabody, D. (1978). In search of an evaluative factor: Comments on De Boeck. Journal of Personality and Social Psychology, 36, Peabody, D. (1984). Personality dimensions through trait inferences. Journal of Personality and Social Psychology, 46, Peabody, D. (1985). National characteristics. Cambridge: Cambridge University Press.

7 1072 Guido Peeters Peabody, D., & Goldberg, L. R. (1989). Some determinants of factor structures from personality-trait descriptors. Journal of Personality and Social Psychology, 57, Peabody, D., & De Raad, B. (2002). The substantive nature of psycholexical personality factors: A comparison across languages. Journal of Personality and Social Psychology, 83, Peeters, G. (1976). Evaluative context effects upon trait-adjectives and nonlinear combination of information. XXIst International Congress of Psychology (p. 210). Paris: Société Française de Psychologie. Peeters, G. (1983). Relational and informational patterns in social cognition. In W. Doise, & S. Moscovici (Eds.), Current issues in European social psychology (pp ). Cambridge: Cambridge University Press. Peeters, G. (1986). Good and evil as softwares of the brain: On psychological immediates underlying the metaphysical ultimates. A. contribution from cognitive social psychology and semantic differential research. Ultimate Reality and Meaning. Interdisciplinary Studies in the Philosophy of Understanding, 9, Peeters, G. (1992). Evaluative meanings of adjectives in vitro and in context: Some implications and practical consequences of positive negative asymmetry and behavioral-adaptive concepts of evaluation. Psychologica Belgica, 32, Peeters, G. (1993). Self-other anchored evaluations of national stereotypes. Internal Report No 5. Leuven: K.U.Leuven, Laboratorium voor Experimentele Sociale Psychologie. Peeters, G. (2004). Review of the book a sociocognitive approach to social norms, edited by N. Dubois. European Bulletin of Social Psychology, 16, Peeters, G. (2007). Good and bad for self and other: From structural to functional approaches of fundamental dimensions of social judgment. Paper prepared for the Small Group Meeting on Fundamental Dimensions of Social Judgment: A View from Different Perspectives, Namur, 7 9 June 2007, and for Trente Années de Psychologie Sociale avec Jean-Léon Beauvois: Bilan et Perspectives, Paris, June Peeters, G., & Czapinski, J. (1990). Positive negative asymmetry in evaluations: The distinction between affective and informational negativity effects. In: W. Stroebe, & M. Hewstone (Eds.), European review of social psychology (Vol. 1, pp ). Chichester: Wiley. Peeters, G., Cornelissen, I., & Pandelaere, M. (2003). Approach-avoidance values of target-directed behaviors elicited by target-traits: The role of evaluative trait dimensions. Current Psychology Letters, 2, 11, Consulted on 1 June 2007, document396.html Phalet, K., & Poppe, E. (1997). Competence and morality dimensions of national and ethnic stereotypes: A study in six Eastern-European countries. European Journal of Social Psychology, 27, Rosenberg, S., & Olshan, K. (1970). Evaluative and descriptive aspects in personality perception. Journal of Personality and Social Psychology, 16, Rosenberg, S., Nelson, C., & Vivekanathan, P. S. (1968). A multidimensional approach to the structure of personality impressions. Journal of Personality and Social Psychology, 9, Saucier, G., Ostendorf, F., & Peabody, D. (2001). The non-evaluative circumplex of personality adjectives. Journal of Personality, 69, Singh, R., Ho, L. J., Tan, H. L., & Bell, P. A. (2007). Attitudes, personal evaluations, cognitive evaluation and interpersonal attraction: On the direct, indirect and reverse-causal effects. British Journal of Social Psychology, 46, Trafimow, D., & Trafimow, S. (1999). Mapping imperfect and perfect duties onto hierarchically and partially restrictive trait dimensions. Personality and Social Psychology Bulletin, 25, Vonk, R. (1996). Negativity and potency effects in impression formation. European Journal of Social Psychology, 26, Wojciszke, B., Bazinska, R., & Jaworski, M. (1998). On the dominance of moral categories in impression formation. Personality and Social Psychology Bulletin, 24,