International Journal of Psychology, 2014 Vol. 49, No. 6, 480–487, DOI: 10.1002/ijop.12073
Where are we coming from versus who we will become: The effect of priming different contents of European identity on cooperation Francesco La Barbera1 , Pia Cariota Ferrara1 , and Mihaela Boza2 1
Department of Political Science, University of Napoli Federico II, Napoli, Italy of Psychology and Educational Sciences, Alexandru Ioan Cuza University of Iasi, Iasi, Romania
2 Department
I
n two experiments, we investigated how priming European identity as common project versus common heritage affects participants’ cooperation in a social dilemma; an additional aim was to explore the mediators involved in the process. In the first experiment, 82 students played a public good dilemma with a European bogus partner and then completed self-report measures of identification with the European Union (EU), group-based trust and collective interest. Results showed that priming a common project-based but not a common heritage-based European social identity fostered cooperative behaviour; this effect was mediated by two sequential mediators: the common project prime increased participants’ strength of identification with EU (mediator 1) which, in turn, positively affected group-based trust (mediator 2), fostering greater cooperation. Experiment 2 was conducted with a similar procedure on a sample of 124 students, using a different measure of trust and changing the order of mediators. Results supported those of previous experiment: Priming a project-based EU identity content (compared to heritage-based one) had significant direct and indirect effects on cooperation. Keywords: European identity; Social identity; Cooperation; Trust; Common project.
Cooperation and competition have been extensively studied in social psychology. Because cooperation is fundamental to social functioning, understanding the antecedents of cooperative behaviour is important for researchers, professionals and policy makers (Brewer, 2000). Although a large body of research has been devoted to study the effect of sharing a group identity on cooperative behaviour (Kramer & Brewer, 1984; La Barbera & Cariota Ferrara, 2012), there is less research on the effect of the content of social identity (Deaux, 2000; Herrmann & Brewer, 2004; Tajfel, 1981) on cooperation. The content of social identity is important when presenting a task requiring cooperation from several parties or countries, as it is the case with the European Union (EU): How should “Europe” be defined to make Europeans from different nations more cooperative? Many political leaders have stressed the importance of the common culture, history and common roots of the EU—an identity
based on a common heritage. Others have stressed the importance of working together to build the future—an identity based on a common project. Which identity is more successful in promoting cooperation and why? In this study we addressed this question. Common heritage versus common project The positive effect of shared identity on cooperation could be relevant in the context of EU: people of different nationalities—ordinary citizens and politicians—may become more cooperative with each other if they share a strong identification with EU. In a recent study, indeed, La Barbera and Cariota Ferrara (2012) found that participants were more cooperative with a European partner when they scored high on European identification. In addition, besides sharing or not a common membership, scholars have also highlighted the importance of
Corresponding should be addressed to Francesco La Barbera, Dipartimento di Scienze Politiche, Università degli Studi di Napoli Federico II, via Rodinò 22, 80138 Napoli (NA), Italy. (E-mail:
[email protected]). We are very grateful to Marilynn Brewer and James Maddux for their insights and suggestions on an early version of the manuscript. We also thank the anonymous reviewers whose thoughtful work helped us in improving this article.
© 2014 International Union of Psychological Science
EUROPEAN IDENTITY CONTENT AND COOPERATION
the social identity content, that is, the shared representations of the collective self (Herrmann & Brewer, 2004). Nonetheless, the effect that different contents of social identity could have on cooperation has not been studied, neither in relation to EU, nor in other contexts. The aim of this study is to address this issue, because different identity contents could differently affect individuals’ strength of identification (Deaux, 2000; Li & Brewer, 2004), thus also affecting their willingness to cooperate. Therefore, we investigated if participants’ cooperation in a social dilemma (Kollock, 1998), with a European (bogus) partner, could be affected by priming them with two different contents of European social identity, which represent different perspectives about EU. The first, which we called common heritage, is built on social beliefs about common values, tradition and religion, which European people have inherited from the past and currently share. The second content, the common project, is built upon European people’s awareness of sharing social, economic and political goals to strive for and to achieve in the future. This dichotomy, which has been found in the common sense (Barro, 2004), also resembles several different theoretical approaches in defining the EU and European identity (Bruter, 2003; Reijerse, Van Acker, Vanbeselaere, Phalet, & Duriez, 2013). Brewer, Hong, and Li (2004) proposed the distinction between essence-based and agency-based perception of the ingroup: “Groups of any type may be viewed in terms of static common properties and similarities or in terms of their dynamic goals and purposes” (p. 32); they also suggested that the perception of agency-based ingroup entitativity enhances security, efficacy, ingroup identification and positivity. EU’s perceived entitativity is essence-based in the case of the common heritage-based European identity content, agency-based in the case of the common project-based identity content (Li & Brewer, 2004). In addition, the common project identity is presentand future-oriented, whereas the common heritage perspective is past-oriented (Faist, 2001). As Bilewicz (2007) showed, dwelling into the past could have a negative impact on intergroup attitudes, especially when the groups have a history of conflict. Europe has been a battlefield for most of its history. Reminding people of their history may also remind them of their conflicts. This could be a limitation of the heritage-based European identity. A project-based perspective, instead, being oriented to the future, should not suffer this burden. Finally, research has widely shown that common goals foster the strength of group identification and could have positive effects on intragroup and intergroup cooperation (Brewer, 2000). As we stated, the common project identity content—more than the common heritage—is based on common goals and common agency. Taken together, these arguments suggest that activating a common project-based European identity—more © 2014 International Union of Psychological Science
481
than a common heritage-based—should increase the individuals’ strength of identification with EU, and this, in turn, should positively affect their willingness to cooperate.
From ingroup identification to cooperation A second aim of the current study was to explore the influence of the different identity contents on cooperation, which could involve two mediation steps. First, as stated previously, priming different contents of EU identity could affect participants’ strength of identification with EU (Brewer et al., 2004; Deaux, 2000), which has been already shown to foster cooperation in a social dilemma with a European partner (La Barbera & Cariota Ferrara, 2012). Therefore, we expected the strength of identification to be the first mediator. Second, previous research about the identification influence on cooperation has proposed two possible interpretations of the process (De Cremer, van Knippenberg, van Dijk & van Leeuwen, 2008; Buchan et al., 2011). The goal amplification hypothesis proposes that common membership could foster group-based trust (i.e. the positive expectations about cooperative behaviour of other members), which could reduce one of the main obstacles for cooperation, namely the fear of exploitation. Therefore, a common social identity, fostering group-based trust, would promote cooperation. In the goal transformation hypothesis, the motive of collective interest, not trust, mediates the effect of social identity on cooperation; the individual’s self-categorisation as a member of a group makes personal and collective goals interchangeable. Hence, the effect of different contents of European social identity (common heritage vs. common project) on cooperation should be mediated, in a first step, by participants’ strength of identification with EU, and in a second step by trust and/or by motive of collective interest. We did not make any specific hypothesis about the second mediation step, because of the lack of research directly and formally contrasting the two possible mediators.
EXPERIMENT 1 In the experiment described below, we primed participants with two different texts built on the different conceptualisations of EU illustrated in previous section, with a procedure similar to that used by Li and Brewer (2004). We hypothesised that participants in the common project experimental condition would be more cooperative than those in the common heritage and control conditions. In
482
LA BARBERA, FERRARA, BOZA
addition, we hypothesised that this effect would be mediated, in the first step, by participants’ strength of identification with EU, and in the second step by trust and/or collective interest. Method Participants The experimental sample comprised 82 university students (58 female; M age = 19.77, SDage = 1.30) from a large Italian university who participated voluntarily. They were randomly assigned to the experimental conditions (common heritage, N = 26; common project, N = 27) and the control condition (N = 29).
were to be multiplied by 1.5, divided equally and redistributed to the two students, irrespective of the number of coins each had contributed. The number of coins allowed by each one of them would be kept secret. To make the game more realistic, they were informed that one pair of participants, assigned by random draw, would receive the money they earned in the game. On the next page, each participant indicated how many coins he/she wanted to contribute to the pool. Then, he/she completed the scales described below. After that, participants were thanked and debriefed. After about 1 month, one of the participants was drawn and was actually paid what she earned from the game (her bogus partner supposedly allowing 25 coins to the pool). Measures
Procedure Participants individually completed a questionnaire in which they were asked to perform a series of tasks. First, they were asked to carefully read a text about the EU. In the common heritage condition, the text stressed the aspects connected to commonality of values and tradition, and to the past. The European Union is based on a common heritage. European identity is a result of different countries having common roots. Being European citizens means coming together from a common past that stems from shared traditions. Being European means to identify oneself in this great past that brings people together, beyond all differences.
In the common project condition, the text emphasised common goals and the future. The European Union is based on a common project. European identity is a result of different countries having common goals. Being European citizens means building together a future scenario that comes from shared purposes. Being European means to identify oneself in this great future that brings people together, beyond all differences.
In the control condition there was no text to be read by participants. Then, participants were asked to play a two-person dilemma (adapted from Mulder, 2008) with a European (bogus) partner. Each participant was told that the research involved several EU’s countries and that he/she would be anonymously paired with a college student of an unspecified EU nation different from his/her own. Participants learned they possessed a personal endowment of 25 “virtual” coins (each one worth one Euro) and that each one of them must individually decide how many coins to invest in a “common pool” and how many to keep for him/herself. His/her European partner would do the same. All coins allowed to the pool
Cooperation. The level of cooperation was measured by the number of coins that each participant allowed to the pool (range: 1–25). Identification with the EU. The four-item identification scale by Spears, Doosje, and Ellemers (1997) was used to measure participants’ strength of identification with EU (α = .80). Participants answered on 7-point scales ranging from 1 (not at all) to 7 (very much). A single score was calculated, with higher values indicating stronger identification with EU. Trust. The level of participants’ trust was measured by two items with 7-point response scale, ranging from 1 (totally disagree) to 7 (totally agree): “I think that my European partner will be very cooperative with me” and “I have a positive expectation about the number of chips that my partner will allow to the pool” (Spearman’s ρ = .60). The items were averaged in a single score, with higher values indicating more positive expectations. Collective interest. The level of participants’ motive of collective interest was measured by two items (Mulder, 2008), which were answered on a 7-point response scale ranging from 1 (totally disagree) to 7 (totally agree): “My goal was to gain the higher possible number of chips for both of us” and “My goal was to reach an excellent profit for both of us” (Spearman’s ρ = .70). The items were averaged in a single score, with higher values indicating higher motive of collective interest. Results and discussion In this article analyses, the accepted significance level has been set on p < .05. Table 1 reports a summary of means, standard deviations and intercorrelation for the measures used in the experiment. © 2014 International Union of Psychological Science
EUROPEAN IDENTITY CONTENT AND COOPERATION
R2 = .10 .31
Identification
.35
R2 = .12 Trust
.45
.19
Project
483
R2 = .31 Cooperation
.24 -.48
Collective interest Heritage
Figure 1. Path analysis of the prime-to-cooperation effect. Note: In the figure, standardised regression weights are reported for significant direct effects. The portion of variance explained for each endogenous variables is also reported. TABLE 1 Experiment 1. Summary of intercorrelations, means and standard deviations for scores on the identification with EU (IDEU), trust (TRUST), collective interest (COLL) and cooperation (COOP) Measure
1
2
1. IDEU 2. TRUST
4.09 (1.33) .349**
3. COLL
.192
4.80 (1.14) .214
4. COOP
.190
.523**
3
5.09 (1.38) .339**
4
16.00 (5.97)
Note: The table shows Pearson’s r correlation coefficients. Diagonal cells report the means (SD in parentheses). *p < .05. **p < .01.
For testing the hypothesis of a significant effect of the identity priming on cooperation, an analysis of variance (ANOVA) was carried out, and the effect of prime was statistically significant, F(2, 79) = 3.91, p = .024, η2 p = .090. Pairwise comparisons showed that the mean of chips that participants contributed was significantly higher in the common project condition (M = 18.48, SD = 4.81) than in the common heritage condition (M = 15.27, SD = 7.05), t(51) = 1.94, p = .046, and in the control condition (M = 14.34, SD = 5.28), t(54) = 3.06, p = .009, with no significant difference between common heritage and control condition (t < 1). Therefore, in line with our expectations, priming a common project-based EU social identity fostered participants’ cooperative behaviour, whereas priming a common heritage-based EU identity did not. A path analysis was performed for studying the direct and indirect relations between prime, identification with 1
EU, trust, collective interest and cooperation. Two correlated dummy variables for the prime were included in the model. The variable “project” was made with value 1 for the project condition and value 0 otherwise. The variable “heritage” was made with the value 1 for the heritage condition and the value 0 otherwise (Hayes & Preacher, in press). Maximum likelihood method was used, with non-normal estimators correction by Bentler’s robust method (Bentler, 2006). In a first model, all direct effects of exogenous variables on the endogenous variables, as well as the effects of identification on trust, collective interest and cooperation, and the effects of trust and collective interest on cooperation were tested; the goodness-of-fit indices were not satisfactory: Normed Fit Index (NFI) = .97, Non-Normed Fit Index (NNFI) = .63, Comparative Fit Index (CFI) = .97, Root Mean Square Error of Approximation (RMSEA) = .15. In the final model (Figure 1), all non-significant paths were set at 0. The goodness-of-fit indices reached satisfactory levels (Hu & Bentler, 1999): NFI = .97, NNFI = .98, CFI = .99, RMSEA = .03. Figure 1 shows a diagram of the final model, which provides standardised weights (β) for significant direct effects. Project had a significant direct effect on identification and cooperation. On the contrary, heritage did not exert any significant effect. Both trust and collective interest affected cooperation, but only the former was affected by identification and project. Therefore, only trust could be included in the hypothesised causal chain that links the identity content and cooperation. Indeed, the indirect effect of project on cooperation via identification and trust was significant, β = .05, SE = .294, z = 2.27, p < .05, as well as the indirect effect of identification on cooperation via trust, β = .16, SE = .244, z = 2.62, p < .05; the indirect effect of project on trust via identification was also significant, β = .11, SE = .118, z = 2.52, p < .05.
The other dummy variable (heritage) was entered as a covariate (Hayes & Preacher, in press).
© 2014 International Union of Psychological Science
484
LA BARBERA, FERRARA, BOZA
A formal test based on 5,000 bootstrap samples (Hayes, 2013) confirmed the significance of the indirect effect of project1 on cooperation via identification (mediator 1 [M1]) and trust (mediator 2 [M2]), β = .06, Bias Corrected (BC) 95% CI (0.012, 0.171). A second bootstrap test (5,000 bootstrap samples) showed that the indirect effect of project (see Footnote 1) on cooperation via identification (M1) and collective interest (M2) was not significant, β = .02, BC 95% CI (−0.036, 0.131). EXPERIMENT 2 A second experiment was conducted for further testing the Experiment 1 findings on a larger sample, using a similar procedure. Taking into account that in previous experiment no significant difference between the heritage and control conditions was found, we reduced the number of experimental conditions to the crucial contrast between common heritage-based and common project-based messages. Also, collective interest, which had no significant mediating role, was excluded from the model. In addition, a different measure of trust was used, and the order of mediators was changed; participants completed the identification scale first, then the trust measure. Multiple regression models and bootstrapping mediation tests were used for testing the significance of the direct and indirect effects included in the causal chain. In summary, the second experiment was focused on the hypothesis that priming participants’ with a project-based message about EU, more than with a heritage-based one, positively affected participants’ willingness to cooperate with a European (bogus) partner in a social dilemma, and that the effect was mediated by identification with EU (M1) and trust (M2). Procedure and measures A total of 124 students (85 female; M age = 20.65, SDage = 2.25) from a large Italian university participated voluntarily in the experiment. They were randomly assigned to the experimental conditions (common heritage, n = 63; common project, n = 61). Participants were primed with the same two texts of previous experiment; then, they were asked to play the social dilemma game with a European (bogus) partner and to complete the measures of identification and trust. The identification measure was the same as in the first experiment (α = .84); participants answered on 10-point scales ranging from 1 (not at all) to 10 (very much). A different measure was used for trust (Buchan et al., 2011): a single item asked participants how much chips they believed their partner allowed to the common pool (range: 1–25). Finally, as we already explained, the order of mediators was reversed; participants completed the identification scale first, and then answered to the item about the number
TABLE 2 Experiment 2. Summary of intercorrelations, means and standard deviations for scores on the identification with EU (IDEU), trust (TRUST) and cooperation (COOP) Measure
1
1. IDEU 2. TRUST
5.70 (1.73) .389
3. COOP
.491
2
15.27 (6.30) .834
3
14.32 (4.67)
Note: The table shows Pearson’s r correlation coefficients. Diagonal cells report the means (SD in parentheses). All correlations are significant at p < .01 level.
of chips they expected from their European fellow. Hence, the identification scale also worked as a delay and a distractor between the measurement of cooperation and trust, thus limiting the projection effect of participants’ choices on their expectations (Buchan et al., 2011). Results and discussion Table 2 provides the means, standard deviations and correlation between the measures used in the Experiment 2. The effect of identity priming on cooperation was significant, F(1, 122) = 23.23, p < .001, η2 p = .160: Participants in the common project condition, compared to those in the common heritage condition, scored significantly higher on cooperation (M project = 16.21, SDproject = 3.89; M heritage = 12.49, SDheritage = 4.66). This result was in line with our expectations and previous experiment’s findings. Direct and indirect effects of prime on cooperation were tested with linear regression models and bootstrapping tests, taking into account the recent approaches and guidelines proposed in the literature about mediation (Shrout & Bolger, 2002; Hayes, 2013). Table 3 provides results of three regression models used for testing the mediating role of participants’ strength of identification with EU and trust in the process of influence of prime on cooperation. The first model shows a significant effect of prime on identification. In the second model, trust was regressed on prime and identification with EU: identification proved to be a significant predictor, whereas prime was not significant. In the third model, prime, identification and trust were all entered as predictors of cooperation: trust was the main direct predictor of cooperation, but prime and identification also were significant. Taken together, results of the regression models are compatible with the hypothesis of a prime-to-cooperation effect mediated by identification with EU and trust (MacKinnon, Fairchild, & Fritz, 2007; Shrout & Bolger, 2002). A formal test of the indirect effects was performed by a bootstrapping analysis based on 5,000 bootstrap samples © 2014 International Union of Psychological Science
EUROPEAN IDENTITY CONTENT AND COOPERATION TABLE 3 Experiment 2. Mediation analysis of the effect of prime on cooperation Dependent variable
Predictor
B
t
R2
1 2
IDEU TRUST COOP
.42*** .11 .34*** .15** .14* .74***
5.13 1.20 3.73 3.00 2.58 14.77
.18*** .16***
3
PRIME PRIME IDEU PRIME IDEU TRUST
Model
.75***
Note: The variables entered in the regression models are prime (PRIME), strength of identification with EU (IDEU), trust (TRUST) and cooperation (COOP). *p < .05. **p < .01. ***p < .001.
(Hayes, 2013), which confirmed the significance of the overall indirect effect of prime on cooperation, β = .247, BC 95% CI (0.097, 0.382), and prime on cooperation via identification (M1) and trust (M2), β = .107, BC 95% CI (0.050, 0.179). Also the indirect effect of prime on cooperation mediated only by identification was significant, β = .058, BC 95% CI (0.004, 0.125), whereas the indirect effect with trust as the single mediator was not, β = .082, BC 95% CI (−0.041, 0.216). GENERAL DISCUSSION A large body of research has confirmed the positive effect that sharing a social identity has on cooperation, but this is the first study to investigate the effect of different contents of social identity on cooperative behaviour. Results of both the experiments showed that the identity prime had a significant effect on cooperation: Project-based European identity was more effective than heritage-based in fostering individuals’ identification, group-based trust and cooperation. The experiments involved two independent samples, were distant in time more than a year from each other and adopted slightly different procedures. Nonetheless, they showed a very similar pattern of results, which therefore seems to be quite robust. Our experiments also showed that the effect of prime on cooperation was mediated by participants’ strength of identification with the EU and group-based trust. The direct effect of prime on cooperation, as well as the indirect effect of identification, was significant in both experiments; in addition, the direct effect of identification on cooperation was significant in Experiment 2; hence, the effect of social identity on cooperation is mediated by group-based trust, but seems to go over and beyond it, because the content of social identity and the strength of social identification remain significant predictors of cooperation even controlling for trust (Buchan et al., 2011). This result also suggests that factors other than trust may mediate the effect of social identity on cooperation. Previous research has proposed collective interest as a possible © 2014 International Union of Psychological Science
485
mediator, but there is a lack of studies testing directly and formally its significance. In the current study, the mediating role of collective interest was not significant. This could be due to an inadequate operationalisation of the concept, but could also suggest that the interchangeability of personal and group goals, that is, the basis of the goal transformation hypothesis, could be more typical of very cohesive groups, whose members share a very high identification with the ingroup, which allows them to “go beyond the self.” The strength of individuals’ identification with the EU is generally not very high (Eurobarometer, 2012; La Barbera & Cariota Ferrara, 2012). Also in our experiments, participants’ identification average score was not statistically different from the scale midpoint (t < 1). Future research could investigate whether the motive of collective interest could be a significant mediator of the common identity effect on cooperation only among “high identifiers.” At the same time, other mediators could be relevant in the case of the EU, such as individuals’ moral/normative concern and/or future expectations, which are considered to play an important role in relation to cooperation (Kollock, 1998). However, our results about mediational processes should be viewed as very preliminary and require further testing, especially in relation to the order of the experimental flow. Indeed, although the order of the mediators was reversed in the second experiment, our procedure did not allow us to measure the mediators before the dependent variable. Also, the topic of identification (the first mediation step) needs further investigation. In the current experiment, participants’ identification with the EU was measured as a unidimensional factor, because we were interested in investigating the impact of different representations of the EU on individuals’ strength of identification with the EU, and if this would mediate the effect of identity content on cooperation. Nonetheless, recent approaches to social identity have defined it as a multidimensional construct (Leach et al., 2008). Therefore, an important extension of the current work should provide for the use of a more complex measure of identification, looking for a more complete understanding of the influence of identity content on cooperation. Indeed, the different components of European identification could have a different mediating role in this process (La Barbera & Cariota Ferrara, 2012). Furthermore, a cross-national comparison of the effects of the different conceptualisations of EU could be an interesting research path. In particular, it would be interesting to explore the effect of project-based and heritage-based European identity on willingness to cooperate in the case of high/low status European member states, as well as in the case of new accession countries. In addition, determining if the effect of identity content on cooperation is moderated by the different kind of group membership (e.g. national, occupational and
486
LA BARBERA, FERRARA, BOZA
informal) could clarify if the identity content difference we highlighted here (project vs. heritage) is relevant just for one kind of membership, or if its relevance could be more general. From a practical perspective, the present findings, especially if replicated with larger and cross-national samples, could be important at a time in which the EU’s image and popularity are at low level in the European countries (Hobolt, 2009; Eurobarometer, 2012). According to Herrmann and Brewer (2004), if we agree about the important and active role that institutions should have as agents of change and builders of identity, “deliberate efforts can be undertaken to build social identification” (p. 14). In this sense, our findings could open new perspectives about “identity building”: In the case of EU, the rhetoric of common roots and past glories might be less effective in promoting identification, trust and cooperation among people from different countries than the idea of EU as a common project. Manuscript received July 2013 Revised manuscript accepted March 2014 First published online May 2014
REFERENCES Barro, M. (2004). I giovani e l’Europa. Rappresentazioni sociali a confronto [Young people and Europe: Comparing social representations]. Milan, Italy: FrancoAngeli. Bentler, P. M. (2006). EQS 6 Structural equation program manual. Encino, CA: Multivariate Software. Bilewicz, M. (2007). History as an obstacle: Impact of temporal-based social categorizations on Polish-Jewish intergroup contact. Group Processes & Intergroup Relations, 10, 551–563. doi:10.1177/1368430207081540. Brewer, M. B. (2000). Superordinate goals versus superordinate identity as bases of intergroup cooperation. In D. Capozza & R. Brown (Eds.), Social identity processes: Trends in theory and research (pp. 117–132). London, U.K: Sage. Brewer, M. B., Hong, Y., & Li, Q. (2004). Dynamic entitativity: Perceiving groups as actors. In V. Yzerbyt, C. M. Judd, & O. Corneille (Eds.), The psychology of group perception: Contributions to the study of homogeneity, entitativity, and essentialism (pp. 25–38). New York, NY: Psychology Press. Bruter, M. (2003). Winning hearts and minds for Europe: The impact of news and symbols on civic and cultural European identity. Comparative Political Studies, 36, 1148–1179. doi:10.1177/0010414003257609. Buchan, N. R., Brewer, M. B., Grimalda, G., Wilson, R. K., Fatas, E., & Foddy, M. (2011). Global social identity and global cooperation. Psychological Science, 22, 821–828. doi:10.1177/0956797611409590. De Cremer, D., van Knippenberg, D., van Dijk, E., & van Leeuwen, E. (2008). Cooperating if one’s goals are collective-based: Social identification effects in social dilemmas as a function of goal transformation. Journal of Applied Social Psychology, 38, 1562–1579. doi:10.1111/j.15591816.2008.00359.x.
Deaux, K. (2000). Models, meanings, and motivations. In D. Capozza & R. Brown (Eds.), Social identity processes: Trends in theory and research (pp. 1–14). London, U.K.: Sage. Eurobarometer (2012). No. 77. Public opinion in the European Union. Retrieved from http://ec.europa.eu/public_opinion/ index_en.htm. Faist, T. (2001). Social citizenship in the European Union: Nested membership. Journal of Common Market Studies, 39, 37–58. doi:10.1111/1468-5965.00275. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press. Hayes, A. F., & Preacher, K. J. (in press). Statistical mediation analysis with a multicategorical independent variable. British Journal of Mathematical and Statistical Psychology. doi:10.1111/bmsp.12028. Herrmann, R. K., & Brewer, M. B. (2004). Identities and institutions: Becoming European in the EU. In R. K. Herrmann, T. Risse, & M. B. Brewer (Eds.), Transnational identities: Becoming European in the EU pp. (1–22). New York, NY: Rowman & Littlefield. Hobolt, S. B. (2009). Europe in question: Referendums on European integration. U.K.: Oxford University Press. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. doi:10.1080/10705519909540118. Kollock, P. (1998). Social dilemmas: The anatomy of cooperation. Annual Review of Sociology, 24, 183–214. doi:10.1146/annurev.soc.24.1.183. Kramer, R. M., & Brewer, M. B. (1984). Effects of group identity on resource use in a simulated commons dilemma. Journal of Personality and Social Psychology, 46, 1044–1056. doi:10.1037/0022-3514.46.5.1044. La Barbera, F., & Cariota Ferrara, P. (2012). Being European in a social dilemma: The effect of European identity on cooperation. Testing, Psychometrics, Methodology in Applied Psychology, 19, 165–175. doi:10.4473/ TPM19.3.2. Leach, C., van Zomeren, M., Zebel, S., Vliek, M., Pennekamp, S., Doosje, B., ... Spears, R. (2008). Group-level self-definition and self-investment: A hierarchical (multicomponent) model of in-group identification. Journal of Personality and Social Psychology, 95, 144–165. doi:10.1037/ 0022-3514.95.1.144. Li, Q., & Brewer, M. B. (2004). What does it mean to be an American? Patriotism, nationalism, and American identity after 9/11. Political Psychology, 25, 727–739. doi:10.1111/ j.1467-9221.2004.00395.x. MacKinnon, D. P., Fairchild, A. J., & Fritz, M. S. (2007). Mediation analysis. Annual Review of Psychology, 58, 593–614. doi:10.1146/annurev.psych.58.110405.085542. Mulder, L. B. (2008). The difference between punishments and rewards in fostering moral concerns in social decision making. Journal of Experimental Social Psychology, 44, 1436–1443. doi:10.1016/j.jesp.2008.06.004. Reijerse, A., Van Acker, K., Vanbeselaere, N., Phalet, K., & Duriez, B. (2013). Beyond the ethnic-civic dichotomy: © 2014 International Union of Psychological Science
EUROPEAN IDENTITY CONTENT AND COOPERATION
Cultural citizenship as a new way of excluding immigrants. Political Psychology, 34, 611–630. doi:10.1111/ j.1467-9221.2012.00920.x. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7, 422–445. doi:10.1037/1082-989X.7.4.422.
© 2014 International Union of Psychological Science
487
Spears, R., Doosje, B., & Ellemers, N. (1997). Self-stereotyping in the face of threats to group status and distinctiveness: The role of group identification. Personality and Social Psychology Bulletin, 23, 538–553. doi:10.1177/0146167297 235009. Tajfel, H. (1981). Human groups and social categories. U.K.: Cambridge University Press.
