This article was downloaded by: [Illinois Wesleyan University]On: 11 October 2014, At: 03:59Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Journal of Social PsychologyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/vsoc20

Intergroup Threat and ExperiencedAffect: The Distinct Roles ofCausal Attributions and In-GroupIdentificationSandro Costarelli aa University of Trento, ItalyPublished online: 07 Aug 2010.

To cite this article: Sandro Costarelli (2009) Intergroup Threat and Experienced Affect:The Distinct Roles of Causal Attributions and In-Group Identification, The Journal of SocialPsychology, 149:3, 393-401, DOI: 10.3200/SOCP.149.3.393-401

To link to this article: http://dx.doi.org/10.3200/SOCP.149.3.393-401

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoeveras to the accuracy, completeness, or suitability for any purpose of the Content. Anyopinions and views expressed in this publication are the opinions and views of theauthors, and are not the views of or endorsed by Taylor & Francis. The accuracy ofthe Content should not be relied upon and should be independently verified withprimary sources of information. Taylor and Francis shall not be liable for any losses,actions, claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connection with, inrelation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms& Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

393

Address correspondence to Sandro Costarelli, Department of Cognitive Sciences and Education, University of Trento, Via M. Del Ben, 5, 38068–Rovereto (TN), Italy; sandro.costarelli@unitn.it (e-mail).

Replications and RefinementsUnder this heading are brief reports of studies providing data that sub-stantiate, challenge, or refine what we think we know. These notes con-sist of a summary of the study’s procedure and as many details about the results as space allows. Additional details concerning the results can be obtained by communicating directly with the author. Submissions to this section must provide a meaningful conceptual replication that extends the construct validity of the work.

Intergroup Threat and Experienced Affect: The Distinct Roles of Causal Attributions

and In-Group Identification

SANDRO COSTARELLIUniversity of Trento, Italy

ABSTRACT. Research shows that under manipulated conditions of intergroup threat, indi-viduals experience greater negative affect to the extent that low in-group identifiers make an in-group-internal attribution rather than an out-group-internal attribution, and high in-group identifiers make an out-group-internal attribution rather than an in-group-internal attribution for outcomes of intergroup comparison that threaten their social identity. The author predicted and found that under conditions of making an out-group-internal attri-bution, such an effect of in-group identification is mediated by the general proneness to perceiving in-group–out-group differences, or intergroup distinctiveness, at high, but not low, levels of in-group identification. Combining the findings of 2 different literatures, the author provides new insights into the distinct roles played by intergroup attributions as a predictor, in-group identification as a moderator, and intergroup distinctiveness as a mediator of the affective responses produced under conditions of social identity threat instantiated by individuals’ making out-group-internal attribution for the in-group unfavor-able outcomes of intergroup comparison.

Keywords: in-group identification, intergroup attributions, intergroup distinctiveness, intergroup emotions, social identity threat

The Journal of Social Psychology, 2009, 149(3), 293–301Copyright © 2009 Heldref Publications

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

394 The Journal of Social Psychology

FROM A SOCIAL IDENTITY PERSPECTIVE (Tajfel & Turner, 1979), re- searchers could argue that people for whom a group is important (i.e., high[er in-group] identifiers) experience emotions deriving from their group membership more strongly than do people for whom their group membership is less essen-tial (i.e., low[er] identifiers). In terms of attribution theory (Weiner, 1985), this suggests that for threatening comparison outcomes to have any psychological consequences, a sufficiently high degree of in-group identification is a neces-sary condition for sensitivity to in-group rather than individual responsibility to develop (cf. Terry & Hogg, 1996; Weiner; for a review and discussion, see Ellemers & Barreto, 2001).

Recent theoretical and empirical research suggests that this general pattern should be qualified by the valence of emotion considered. Concerning negative affective reactions, Branscombe, Ellemers, Spears, and Doosje (1999) recently suggested that high identifiers, “precisely because of the strength of their group identity, . . . are not always the most sensitive group members to threats to their social identity value” (p. 49; see also Ellemers, Spears, & Doosje, 2002). For example, die-hard soccer fans can be expected to aptly discount even the most inexcusable poor performance of the beloved team. In a similar vein, Doosje, Branscombe, Spears, and Manstead (1998) found empirical support for the argu-ment that group-based negative emotions “are only likely to be experienced by people who admit or accept that their group has done something wrong in the first place” (p. 879; see also Bizman, Yinon, & Krotman, 2001). As a consequence, those authors further argued that because high identifiers are typically unlikely to accept a negative interpretation of their group’s behavior, they use defensive means of dealing with such a group-threatening situation by “explaining away” their group’s negative behavior. In turn, this prevents high identifiers from expe-riencing the typically negative affective consequences of social identity threat. Doosje et al. also found empirical support for the reverse argument: Because low identifiers are typically more willing to accept the idea that their group has done something wrong, they are less likely to display defensive reactions and, there-fore, more likely to experience group-based negative emotions.

When considered together, these previous works indirectly support the counterintuitive hypothesis that only to the extent that one does not identify with the group will an in-group-internal attribution for unfavorable outcomes of inter-group comparisons (e.g., being informed that one’s beloved soccer team lost a match because the players were just not capable of coordinating themselves while playing) not be efficiently discounted. As a consequence, for low—but not high— identifiers, such an attribution will be perceived as a more serious threat to their social identity, in turn leading to a greater negative affective response. Although recent research (e.g., Costarelli, 2007) has provided evidence supporting this idea, it also showed out-group target internality of attributions for in-group unfavorable comparison outcomes to produce a more negative affective response for high in-group identifiers than for low in-group identifiers. First, in-group identification

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

Costarelli 395

is positively associated with perceived salience of relevant out-groups that are present in the intergroup context (cf. Tajfel & Turner, 1979). As a consequence, high identifiers are particularly attuned, and therefore affectively vulnerable, to attributions highlighting in-group unfavorable comparison outcomes as being determined by out-group responsibility (merits) rather than by in-group respon-sibility (shortcomings). For die-hard soccer fans, for example, information under-lining that their beloved team lost a match because the rival team’s players were particularly good at coordinating themselves while playing should lead those fans to experience greater negative affect than information underlining that their beloved team’s players kept making serious strategy mistakes throughout the game. Consistent with this reasoning, Costarelli recently found that high, but not low, identifiers experience negative affect to the extent that they attribute in-group unfavorable outcomes of an intergroup comparison to a relevant out-group’s supe-rior performance (an out-group-internal attribution), rather than to the in-group’s poorer performance (an in-group-internal attribution).

To strengthen this argument, in the present study I tried to show that such an effect of in-group identification (expected for out-group vs. in-group internality of attributions) is mediated by individuals’ perceptions regarding in-group–out-group categorical difference (intergroup distinctiveness: Jetten, Spears, & Man-stead, 2001; for a discussion, see Jetten & Spears, 2003). I expected this to be the case because I anticipated this variable to be positively related to high identi-fiers’ greater negative affective responsiveness to the presence of an out-group in the intergroup context to whose characteristics or actions they can attribute a threatening intergroup status differential. Specifically, I based my hypothesis on the prediction of self-categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) that people minimize intragroup and maximize intergroup differences to the extent that they identify with the group. Thus, the stronger the in-group identification, (a) the more categorically different the perceptions of the in-group and the out-group and (b) the more legitimated by social reality those intergroup differences (on a given dimension of comparison) that instanti-ate an identity-threatening intergroup status differential correspondingly should be perceived.

In contrast, researchers can expect low in-group identifiers to be less responsive in the intergroup context to the presence of an out-group to whose characteristics or actions they can attribute a threatening intergroup status differ-ential. Previous researchers (e.g., Costarelli, 2007) have shown that low identi-fiers are particularly attuned and therefore affectively vulnerable to attributions highlighting in-group unfavorable outcomes of intergroup comparisons as being determined by in-group shortcomings (an in-group-internal attribution), rather than by out-group merits (an out-group-internal attribution). Consistent with that reasoning, researchers would not predict such effects of in-group identification on the negative affective response under conditions of in-group-internal attribution to be mediated by intergroup distinctiveness.

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

396 The Journal of Social Psychology

Method

Participants and Design

Participants were 50 female high school students (M age = 19.29 years, SD = 0.89 years) from Bolzano, Italy. In the present study, I capitalized on a linguisti-cally defined, naturally occurring in-group in selecting the experimental sample: The participants, or in-group members, belonged to the Italian-speaking, ethnic- linguistic population group living in Bolzano (the main town of the Italian South-Tyrol territory); the target out-group comprised high school students who were members of the German-speaking, ethnic-linguistic population group living in the same town.

The design was a 2 (attributions: in-group-internal vs. out-group-internal) × continuous measure of in-group identification. The two attribution conditions dif-fered on whether the nature of the threatening attribution provided to participants referred to in-group-internal or out-group-internal attributions for unfavorable outcomes of intergroup comparison.

Procedure and Measures

After a regular lecture, a questionnaire was given to students volunteering to participate in a national survey. This was allegedly conducted by a governmental agency to compare school achievements of Italian-speaking students (the par-ticipants’ in-group) and German-speaking students (the participants’ out-group) attending high school in the town of Bolzano, Italy. First, in-group identification (as adapted for the present in-group; i.e., Italian-speaking high school students in Bolzano) was measured beforehand. Then the manipulation of social identity threat was instantiated by two different types of bogus information. Specifically, its content regarded differentially threatening counterfeit research results that were presented to participants across experimental conditions manipulating the perceived locus of the attribution made for the threatening information. Thus, in the in-group-internal attribution condition, participants were informed that in-group unfavorable outcomes of intergroup comparison (i.e., poorer school achievements of fellow in-group members) had been ascertained to result from the lower mean amount of time spent in doing homework and the lower mean amount of annually read books, by the local Italian-speaking high school stu-dents. In the out-group-internal attribution condition, participants were informed that in-group unfavorable outcomes of intergroup comparison (i.e., poorer school achievements of fellow in-group members) had been ascertained to result from the higher mean amount of time spent in doing homework and the higher mean amount of annually read books, by the local German-speaking high school stu-dents. Subsequently, manipulation checks were taken for the attributions made for the intergroup comparison outcome as being in-group internal or out-group internal. I then assessed the level of perceived intergroup distinctiveness (the

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

Costarelli 397

mediator). Next, I took dependent measures of level of experienced negative affect. Last, participants were asked to indicate their gender. After all participants had completed the questionnaire, they were debriefed and thanked. All items were measured on a 7-point Likert-type scale ranging from 1 (not at all) to 7 (very much).

Measures

In-group identification. Participants were first asked to answer five items, devel-oped by Cadinu and Reggiori (2002), which measured level of identification with professional groups (e.g., “I feel like a member of the category of Italian high-school student”; “I am proud to be an Italian high-school student”; “I often think of myself as an Italian high-school student”). The identification scale showed satisfactory internal consistency (Cronbach’s α = .85).

Manipulation check. Participants were asked to answer three items to measure whether attributions for the in-group unfavorable status differential made by the participants allocated to the in-group-internal and out-group-internal conditions were more in-group internal or out-group internal (1 = “The reasons behind the facts emerging from the data reported above have much more to do with reasons . . . that can be considered as being internal to Italian-speaking high school students in Bolzano [or] . . . for which Italian-speaking high school students in Bolzano can be considered as being responsible [or] . . . that can be considered as having much more to do with Italian-speaking high school students in Bolzano than with reasons . . . that can be considered as being internal to German-speak-ing high school students in Bolzano [or] . . . for which German-speaking high school students in Bolzano can be considered as being responsible [or] . . . that can be considered as having much more to do with German-speaking high school students in Bolzano”; 7 = “The reasons behind the facts emerging from the data reported above have much more to do with reasons . . . that can be considered as being internal to German-speaking high school students in Bolzano [or] . . . for which German-speaking high school students in Bolzano can be considered as being responsible [or] . . . that can be considered as having much more to do with German-speaking high-school students in Bolzano than with reasons . . . that can be considered as being internal to Italian-speaking high-school students in Bolzano [or] . . . for which Italian-speaking high school students in Bolzano can be considered as being responsible [or] . . . that can be considered as having much more to do with Italian-speaking high school students in Bolzano”; α = .82).

Intergroup distinctiveness. Perceived in-group–out-group categorical difference (Jetten et al., 2001; for a discussion, see Jetten & Spears, 2003) was measured by administering three items to the participants: (a) “I think Italian-speaking and

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

398 The Journal of Social Psychology

German-speaking students ... belong to different groups, (b) . . . can be considered as members of two distinct groups, (c) . . . should be regarded as part of two dif-ferent communities]”; α = .88) to the participants.

Affect. Respondents were asked to report the extent to which each of 10 emotional adjectives (the Positive Affect Negative Affect Schedule scale; Watson, Clark, & Tellegen, 1988) applied to how they were feeling at the moment. Participants were instructed not to think too much about their ratings but, instead, to give quick, gut-level responses. My focus of interest was on general, negative affective reactions. Accordingly, the negatively valenced and the reverse-scored positively valenced affective items were combined to provide a single index of negative affect on which higher scores reflected a more negative affective response. This choice was validated by the results of a principal-components analysis on the 10 emotion items specifying that the factors with eigenvalues greater than 1 be retained. A one-factor solution was extracted, accounting for 67% of the variance. All factor loadings exceeded .79. Furthermore, this combined measure showed a very good internal consistency (α = .86).

After all participants had completed the questionnaire, they were debriefed and thanked. All responses were given on a 7-point Likert-type scale ranging from 1 (not at all) to 7 (very much).

Results and Discussion

Manipulation Checks

An analysis of variance (ANOVA; attribution manipulation effect coding: –1 = in-group internal, 1 = out-group internal; in-group identification: continuous regressor) on the attribution manipulation check scores revealed a statistically significant main effect of attribution, F(1, 49) = 40.71, p < .001. No other effects were significant, Fs(1, 49) < 0.40, ps < .83. This confirmed that participants’ attributions for the outcome of the in-group unfavorable intergroup comparison differed across conditions. Consistent with the manipulation, participants in the out-group-internal attribution condition (M = 5.17, SD = 0.70) reported making more out-group-internal attributions than did participants in the in-group-internal condition (M = 2.57, SD = 0.74).

Affect

As found in previous research (e.g., Costarelli, 2007), I expected that in-group identification would moderate the impact of causal attributions on experienced negative affect. A previous hypothesis was that after facing an in-group-internal attribution, individuals with weaker group identification would

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

Costarelli 399

experience greater negative affect. In contrast, stronger group identification was predicted to be associated with greater negative affect after individuals made an out-group-internal attribution. To test these predictions, I conducted an ANOVA with the attribution manipulation entered as a categorical factor (effect coding: –1 = in-group-internal, 1 = out-group-internal) and in-group identification entered as a continuous regressor. I tested this model with negative affect as a dependent variable.

In line with those predictions, inspection of the parameter estimates revealed that the Attribution × Identification interaction was significant: B = –0.37, SE B = 0.14; t(49) = –2.49, p < .05. Neither main effect was significant: for attribu-tion condition: B = –0.14, SE B = 0.45; t(49) = –0.98, ns; for identification, B = –0.02, SE B = 0.19; t(49) = –0.24, ns. Tests of simple slopes for the regression of attribution on negative affect were performed at two levels of the manipulation variable (cf. Aiken & West, 1991). Consistent with predictions and replicating the effects observed in previous research (e.g., Costarelli, 2007), for participants in the in-group-internal attribution condition, greater negative affect was associated with weaker identification, B = –0.76, SE B = .027; t(23) = –2.83, p < .05. In contrast, as expected, for participants in the out-group-internal attribution condi-tion, greater negative affect was associated with stronger identification, B = 0.47, SE B = 0.14; t(27) = 3.27, p < .01.

Mediation Analysis

I conducted further analyses to test the following hypothesis: For high iden-tifiers, the increase in degree of experienced negative affect produced by being confronted with out-group-internal attributions, rather than in-group-internal attributions, for group unfavorable outcomes of intergroup comparison is medi-ated by perceived intergroup distinctiveness—a variable that is specifically rel-evant to high identifiers’ greater sensitivity to the performing presence of an out-group in the intergroup context in comparison with low identifiers. Accordingly, mediation by intergroup distinctiveness of the respective effects of identification on negative affect observed (as previously indicated) for high and low identifiers was expected for high identifiers in the out-group-internal condition, but not for low identifiers in the in-group-internal condition. Thus, using the procedure indicated by Baron and Kenny (1986) for simple mediation, my prediction that intergroup distinctiveness would mediate the slope of identification on negative affect in the out-group-internal condition but not in the in-group-internal condi-tion was tested separately for participants in the in-group-internal condition and in the out-group-internal condition.

For participants in the in-group-internal condition, two separate simple-regression analyses revealed a significant effect of identification on negative affect, B = –0.76, SE B = 0.27; t(23) = –2.83, p < .05, and on perceived inter-group distinctiveness (the mediator), B = 0.72, SE B = 0.42; t(23) = 2.26, p <

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

400 The Journal of Social Psychology

.05. When the effect of distinctiveness was controlled for by entering it into the regression model, as expected, the path from distinctiveness to negative affect was not reliable, B = –0.01, SE B = 0.15; t(23) = –0.14, p < .88. In contrast, the direct path from identification to negative affect remained significant, B = –0.76, SE B = 0.28; t(23) = –2.71, p < .05. Together, consistent with predictions, these results indicate that no evidence of mediation by perceived in-group–out-group distinctiveness was found for participants in the in-group-internal condition in which the attributional focus was on in-group characteristics or actions rather than out-group characteristics or actions.

For participants in the out-group-internal condition, two separate simple-regression analyses revealed a significant effect of identification on negative affect, B = 0.47, SE B = 0.14; t(27) = 3.27, p < .01, and on perceived intergroup distinctiveness, B = –0.90; SE B = 0.37, t(27) = –2.47, p < .05. When the effect of distinctiveness was controlled for by entering it into the regression model, as pre-dicted, the path from distinctiveness to negative affect was significant, B = 0.60, SE B = 0.10; t(27) = 2.78, p < .05. However, the direct path from identification to negative affect was no longer reliable, B = 0.23; SE B = 0.14, t(27) = 1.63; p < .14. A subsequent test ascertained that, as expected, the reduction to nonsignificance of the predictive effect of identification on negative affect due to the indirect effect of the proposed mediator was significant, z = –2.25, p < .05.

Together, the results of this study replicate the findings of prior research (e.g., Costarelli, 2007) that low and high identifiers experience greater negative affect when they face an in-group-internal attribution and an out-group-internal attribution for threatening outcomes of intergroup comparison, respectively. It is more important that the present study provides novel insights into the psychologi-cal underpinnings of high identifiers’ negative responses as being mediated by a factor that is conceptually consistent with, although different from, the one that earlier research has revealed as a mediator of the investigated affective responses (e.g., Costarelli)—that is, intergroup distinctiveness. Specifically, as suggested by the results of the present mediational analysis, when high identifiers attribute to the out-group the causes of group unfavorable outcomes of intergroup com-parison, their general proneness to perceiving in-group–out-group differences renders particularly strong the threat posed by the group unfavorable intergroup status differential. In turn, this leads such group members to experience strong negative affect. Conversely, this mediation effect is not found to drive low identi-fiers’ stronger affective responsiveness to attributing in-group unfavorable out-comes of intergroup comparisons to in-group shortcomings (an in-group-internal attribution) rather than out-group merits (an out-group-internal attribution). The observed asymmetry in the mediational role of intergroup distinctiveness sug-gests that low identifiers’ affective reactions under such conditions are attribution processes that are driven by (self or group) blame rather than by intergroup locus of causality. Future researchers should explore whether this idea is empirically viable by pitting the mediational effects of these two factors against each other.

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014

Costarelli 401

AUTHOR NOTES

Sandro Costarelli is a member of the Department of Cognitive Sciences and Educa-tion at the University of Trento, Italy. He has conducted social psychology research in the areas of intergroup emotions, attitude ambivalence, prejudice, and discrimination.

REFERENCES

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interac-tions. Thousand Oaks, CA: Sage.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.

Bizman, A., Yinon, Y., & Krotman, S. (2001). Group-based emotional distress: An extension of self-discrepancy theory. Personality and Social Psychology Bulletin, 27, 1291–1300.

Branscombe, N. R., Ellemers, N., Spears, R., & Doosje, B. (1999). The context and content of social identity threat. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social identity context, commitment, content (pp. 35–58). London: Blackwell.

Cadinu, M., & Reggiori, C. (2002). Discrimination of a low-status outgroup: The role of ingroup threat. European Journal of Social Psychology, 32, 501–515.

Costarelli, S. (2007). Intergroup threat and experienced affect: The distinct roles of causal attributions, ingroup identification, and perceived legitimacy of intergroup status. Per-sonality and Social Psychology Bulletin, 33, 1481–1491.

Doosje, B., Branscombe, N. R., Spears, R., & Manstead, A. S. R. (1998). Guilty by asso-ciation: When one’s group has a negative history. Journal of Personality and Social Psychology, 75, 872–886.

Ellemers, N., & Barreto, E. (2001). The impact of relative group status: Affective, percep-tual, and behavioural consequences. In R. Brown & S. Gaertner, (Eds.), The Blackwell handbook of social psychology, (Vol. 4, pp. 324–343). Oxford, England: Blackwell.

Ellemers, N., Spears, R., & Doosje, B. (2002). Self and social identity. Annual Review of Psychology, 53, 161–186.

Jetten, J., & Spears, R. (2003). The divisive potential of differences and similarities: The role of intergroup distinctiveness in intergroup differentiation. European Review of Social Psychology, 14, 203–241.

Jetten, J., Spears, R., & Manstead, A. S. R. (2001). Similarity as a source of discrimination: The role of group identification. European Journal of Social Psychology, 23, 313–325.

Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations. Monterey, CA: Brooks & Cole.

Terry, D. J., & Hogg, M. A. (1996). Group norms and the attitude behaviour relationship: A role for group identification. Personality and Social Psychology Bulletin, 22, 776–793.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Redis-covering the social group: A self-categorization theory. Oxford, England: Blackwell.

Watson, D., Clark, L. A., & Tellegen, L. G. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.

Weiner, B. (1985). An attributional theory of achievement, motivation, and emotion. Psy-chological Review, 4, 548–573.

Received November 19, 2007Accepted February 17, 2008

Dow

nloa

ded

by [

Illin

ois

Wes

leya

n U

nive

rsity

] at

03:

59 1

1 O

ctob

er 2

014