Nationalistic Collective Rituals, Intergroup Relations, and Legitimation of National Social Systems

The epigraph shows two perspectives on one of the most popular sports in Latin America and the rest of the world. In this article, we test related ideas based on literature on collective rituals, national identity, and social identity theory. Specifically, we explore if international football matches can be understood as a type of nationalistic collective ritual that can lead to prejudice toward immigrants and legitimation of national social systems.

Despite this overall pattern, the literature has suggested that national identification can take different forms with consequences for the study of outgroup prejudice and legitimation of national social systems. Some authors distinguish between conventional and constructive patriotism (Sekerdej & Roccas, 2016), blind and constructive patriotism (Livi et al., 2014), or nationalism and patriotism (Bar-Tal, 1993;Hoyt & Goldin, 2016). All these conceptualizations share a common definition. The former refers to an uncritical alignment with the nation, and the latter is the attachment focused on ingroup well-being. In the specific case of the conceptualization we use in this study, patriotism is conceived as a form of attachment, pride, and love for the nation. Nationalism, on the other hand, entails a perception of superiority over other countries or nationalities (Kosterman & Feshbach, 1989;Li & Brewer, 2004). Studies have shown that nationalism is predictive of negative attitudes toward outgroup members; but patriotism can be associated with positive outgroup attitudes given that it is coherent with critical perceptions of the national ingroup (Carter, Ferguson, & Hassin, 2011;Wagner et al., 2012). In the specific case of legitimation of national social systems, it has been found that nationalism -and not patriotism -predicted system justification and perception of meritocracy (Vargas-Salfate & Miranda, in press).

Collective gatherings and rituals
According to Durkheim's (1976) perspective, collective gatherings and rituals are defined as a set of actions and utterances aimed to achieve a social goal. In that sense, rituals are characterized by a convergence of individuals who manifest different actions in a formalized and repetitive way. These actions are not always instrumental or directly related to the desired goal, but they imply shared emotions, the expression of societal values and intended social relations within a community (Paez & Rimé, 2014). In the present article, we argue that shared exposure to international football competitions can be conceived as a form of collective rituals. Indeed, matches involve different conventions, including some that are not directly related to the main goal of this competition, such as playing the national anthem . In addition, people frequently watch these matches with other individuals, focusing their attention to the event and sharing their emotions about the national team performance. Finally, individuals are also exposed to the actual collective behavior of the audience by mass media . For all these reasons, we believe the exposition to international football competitions can be conceived as a collective ritual.
In collective rituals, the social identity is enhanced leading to a high sense of belonging to the ingroup (Drury & Reicher, 2009;Khan et al., 2016) and to a fusion of one's personal identity with the ingroup identity (Lobato & Sainz, 2019). Fusion of identity goes beyond group identification, strengthening the social cohesion between members (Buhrmester et al., 2012). This fusion affects mainly those individuals directly involved in the ritual. In the specific case of sport collective rituals, the atmosphere surrounding the event is made up of various ceremonies and parades, which are associated with a level of emotional understanding that only comes from these intense shared first-hand experiences (Páez et al., 2015).
An important feature in this neo-Durkheimian approach is that participating in collective rituals leads to experiencing collective emotions von Scheve, 2011;von Scheve & Ismer, 2013). This type of emotions is understood as those emotions that are shared by a group and are due to more reasons than the mere identification with that group such as sharing the same collective context (e.g., a dictatorship can lead to an emotional climate marked by fear among the population; Paez, Espinosa & Bobowik, 2013). Collective emotions explain why rituals affect collective attachment and identity fusion. In that sense, several authors have proposed that emotions arise in collective gatherings reinforcing social identity processes (Hopins et al., 2016). However, to our knowledge, none has studied its role on intergroup relations and legitimation of national social systems. Given that collective rituals enhance social identity over individual identity, we expect national rituals to enhance perceptions of fairness and legitimacy of the system. This association should be mediated by collective emotions.
Indirect evidence supporting our assumption for treating international football competitions as collective rituals comes from research showing that German spectators experienced a high level of national pride and identification with patriotic symbols during the 2010 South Africa Football World Cup  and that the 2006 Germany Football World Cup contributed to patriotism in the specific domain of sport-although it was unrelated to xenophobia (Kersting, 2007). In addition, literature on sociology shows that sport is an important source of national pride in Western societies (Broch, 2016;Buffington, 2012); pride in sport achievements is a predictor of national pride (Meier & Mutz, 2016); and in some countries-such as Australia and Germany-the flag (i.e., a national symbol) is associated with sports among their inhabitants (Becker et al., 2017).

The moderator role of nationalism
Our argument states that nationalistic collective rituals might lead to outgroup prejudice and legitimation of national social systems because they make cognitively salient national identities and emotions related to them. Nevertheless, as it was exposed before, national identification can take different forms such as nationalism and patriotism. According to Roccas, Klar, and Liviatan (2006), glori-fication-the term these researchers use to refer to nationalism-entails not only a sense of superiority over other countries but also a high centrality of the national identity for the individual and collective self-concept. For that reason, glorification-or nationalism-also implies an uncritical alignment with national symbols, values, and leaders. In that sense, when national values are made cognitively salient to nationalistic individuals, they feel more compelled to behave in a coherent way with these values (Butz, Plant & Doerr, 2007). Given that social groups build their identities pursuing a positive self-concept (Brown, 2000), individuals tend to associate their countries with egalitarian values, which are contradictory with hostile behaviors toward minority groups. This argument was proposed by Butz, Plant, and Doerr (2007) to study the effect of the U.S. flag exposure on intergroup relations. These authors found that nationalism was not related to hostility toward Arabs and Muslims when American participants were exposed to the U.S. flag (for a different result, see Becker et al., 2012).
Contrary to Butz, Plant, and Doerr (2007), the independent variable of the present study is nationalistic collective ritual in the form of international football competitions. This type of social interactions is usually more competitive than mere flag exposure, with implications for non-nationalistic individuals. International football matches are competitive zero-sum situations. When the social context is highly competitive, intergroup conflict is likely, because groups dispute for scarce resources (Huddy, 2003). For individuals scoring low in nationalism, national values are less relevant to define their self-concept (Roccas, Klar & Liviatan, 2006), so the involvement in a competitive situation is more likely to lead to outgroup prejudice than flag exposure because of the lack of centrality of national values associated with egalitarianism and inclusiveness. Consistent with this idea, Piotrowski et al. (2019) found that different forms of national identification predicted prejudice toward refugees when they were associated with zero-sum thinking.
In summary, we expect that exposure to nationalistic collective rituals will lead to less outgroup prejudice and legitimation of national social systems among nationalistic individuals, but to more outgroup prejudice and legitimation of national social systems among non-nationalistic individuals.

The present research
In Study 1, we test our main conceptual assumption related to the treatment of international football competitions as a form of collective gathering or ritual. Although previous research has tested this hypothesis in other contexts , we extend these studies by showing this phenomenon in Latin America. Based on our theorization of collective rituals, in Study 1 we tested the following hypotheses using a correlational approach: H1: Nationalistic collective rituals will be associated to national identification. H2: The association between interest in nationalistic collective rituals and national identification will be mediated by collective emotions and fusion of identity with the national team.
In Study 2, we aim at analyzing the effect of participating in collective rituals on outgroup prejudice and legitimation of national social systems. Using an experimental design in a Chilean sample, we tested the following hypotheses: H3: Participating in nationalistic collective rituals will lead to outgroup prejudice and legitimation of the national social systems. H4: The effect of participating in nationalistic collective rituals on outgroup prejudice and legitimation of the national social systems will be moderated by nationalism. Specifically, among individuals high (low) in nationalism, participation in nationalistic collective rituals will lead to less (more) outgroup prejudice and legitimation of the national social systems. H5: The effect of participating in nationalistic collective rituals on outgroup prejudice and legitimation of the national social systems will be mediated by collective emotions.

Study 1
The first study was developed to test whether collective gatherings increase national identification through fusion of identity with the national team and the experience of collective emotions. In this study, we used a proxy measure related to interest in football previous to the 2014 Brazil Football World Cup among participants from Latin American countries and Spain. Interest in football, which is associated with exposure to previous activities of national teams, might be conceived as a sort of mediated participation in collective rituals through mass media, as it can be inferred from von .

Participants
Four hundred and fourteen individuals participated in Study 1 (59.2% women), with an age mean of 25.89 years old (SD = 9.86). Participants came from Chile (48.3%), Brazil (41.1%), and Spain (10.63%). All national teams from these countries were classified for the 2014 Brazil Football World Cup. A post-hoc sensitivity analysis was developed using the software G*Power v. 3.1.9.2 (Faul et al., 2009). Considering the sample size (N = 414), power of 0.80 (α = 0.05), and five predictors in a linear regression model, such as in our main analysis, the minimum effect size we could detect is f 2 = 0.03. This implies that our sample was able to detect a small effect size.

Procedure
Participants were invited to participate in an online survey about emotions and experience related to the 2014 Brazil Football World Cup. We contacted undergraduate participants in different universities and nonstudent participants through social network. The questionnaire was administered one week before the beginning of this international competition (i.e., from June 5th to June 12th, 2014). The questionnaire contained an informed consent, measures of interest in football and identification with one's country. Then, we asked participants about collective emotions, social climate, and fusion of identity. Finally, we included a brief demographic questionnaire.

Interest in Football
We included a single-item measure of Interest in Football. Since we collected the data few days before the beginning of the international football competition, people interested in football are very likely to be exposed to mass media information related to national team activities, including their preparation for the international competition and previous matches. It is reasonable to think that these people saw some of these activities in collective contexts. This variable was measured with a scale ranging from 1 (not interested) to 10 (extremely interested).

National Identification
National Identification was measured by a four-item scale adapted and translated for assessing identification with the national ingroup (e.g., Generally, I feel fortunate to be from my country; Luhtanen & Crocker, 1992;Espinosa & Tapia, 2011), ranging from 1 (totally disagree) to 5 (totally agree). This scale was highly reliable (α = 0.85).

Positive Emotions
We selected ten items from the scale developed from Fredrickson (2009) to assess positive emotions felt right before the international football competition. The total score of this measure was conceived as an index of collective or ingroup related emotions because participants were explicitly asked to refer to these emotions when thinking about the competition and the moments previous to it (e.g., amusement, awe, gratitude, hope). All items were measured in a scale ranging from 1 (none) to 4 (a lot). This measure was highly reliable (α = 0.95).

Fusion of Identity
We included a pictorial measure to assess fusion of identity with the national team (Swann et al., 2009). Participants were asked to choose the diagram that best described the relation between them and their national football team. Specifically, each figure represented two circles: one for the participant and one for national team. Answers ranged from 1 (without overlap) to 5 (complete overlap).

Socio-demographic variables
In our models, we controlled for age, gender (1 female, 0 male), and country. In the case of country, we recoded this variable in two dummy variables (i.e., Brazil and Spain) with Chile as the reference category.

Results and discussion
Descriptive statistics and correlations matrix are shown in Table 1. As expected, interest in football right before the 2014 Brazil Football World Cup was associated with higher scores in Positive Emotions, r(414) = 0.56, p < 0.001, Fusion of Identity, r(414) = 0.29, p < 0.001, and National Identification, r(414) = 0.31, p < 0.001. In addition, female participants were less interested as compared to male participants, r(414) = -0.21, p < 0.001, and age was negatively related to Positive Emotions, r(414) = -0.17, p < 0.001.
We ran a multiple regression analysis with National Identification as the dependent variable and Interest in Football and socio-demographic factors (i.e., age, gender, and country) as independent variables. The results are shown in Table 2 and indicate that the regression was statistically significant, F(5, 408) = 23.60, p < 0.001, R 2 = 0.224, and Interest in Football was significantly and positively related to national identification. In addition, we estimated a path analysis with a bootstrap approach using 1,000 samples through the package lavaan for R (Rosseel, 2012), in order to test the mediator role of Positive Emotions and Fusion of Identity. Following the recent recommendations suggested by Yzerbyt, Muller, Batailler & Judd (2018), in addition to the joint-significance tests of the indirect effects, we also observed all the paths involved in the mediation. After adjusting for the same covariates, Interest in Football predicted  Supporting our first hypothesis, these results suggest that international football matches, at least in Chile, Brazil, and Spain, may be conceived as a form of nationalistic collective ritual because Interest in Football predicted National Identification. In addition, Study 1 showed that the association between Interest in Football and National Identification is mediated by Fusion of Identity-and not the experience of collective or in-group related positive emotions-providing partial support for our second hypothesis. Our theoretical argument suggested that these two variables might have accounted for the effect of collective rituals on national identity and related variables. One possibility is that interest in football by itself does not elicit ingroup related positive emotions because it is only an indirect proxy of the experience of participating in a collective ritual. In addition, it is not possible to rule out the possibility that fusion of identity and national identification causes interest in participating in nationalistic collective rituals, in general, and interest in football, in particular. To solve these issues, in Study 2 we directly manipulated the participation in collective rituals. We also tested our hypotheses related to the effect on outgroup prejudice and legitimation of national social systems.

Study 2
In Study 2, we tested Hypotheses 3, 4, and 5 using an experimental approach. The main purpose of this methodology was to avoid self-selection biases associated with differences in national attitudes at the baseline level. In doing so, we expected to identify the effect of nationalistic collective rituals on prejudice toward immigrants and legitimation of national social systems and to explore the moderator role of nationalism.

Participants
One hundred and eighteen Chilean undergraduates participated in Study 2 (19.49% female), who were randomly assigned to experimental and control groups. In terms of social status, participants placed themselves near to the middle point of a scale ranging from 1 (the lowest position) to 10 (the highest position) with the mean of 4.91 (SD = 1.72). A sensitivity analysis was developed using the software G*Power v. 3.1.9.2 (Faul et al., 2009) considering power of 0.80 (α = 0.05) and our sample size (N = 118). This analysis showed that our sample size was able to detect a medium effect size in a t-test analysis (d = 0.52).

Procedure
All participants were informally contacted by a research assistant at the university campus. She briefly explained the cover purpose of the study (i.e., intergroup relationships) and asked them for their participation in exchange of being part in a retail gift card draw. If they agreed, they were conducted to a laboratory and assigned to a computer. After the informed consent, the baseline measures were included (i.e., nationalism and patriotism). Next, the website randomly directed participants to one of the two conditions. In the experimental condition, we asked participants to write a detailed description of an occasion in which they watched a male Chilean football match with other individuals. We emphasized they had to present as many details as possible including description of the match, the place they were, other individuals in that place, and the emotions they felt. In the control condition, we asked participants to do the same task but related to the last time they went to shop without the company of other individuals. The rest of study was similar for both conditions. Specifically, we then asked for the emotions they felt in the situation described. Next, we presented the dependent variables (i.e., system justification, perceived meritocracy, prejudice toward immigrants, prejudice toward Peruvian immigrants). Then, we included a brief demographic questionnaire. Finally, all participants were debriefed and thanked.

System Justification
As a measure of legitimation of the national social systems, we included the System Justification Scale (Kay & Jost, 2003) using a translated and adapted version for the Chilean context (e.g., In general, you find your society to be fair or Everyone has a fair shot at wealth and happiness). All eight items were measured in a scale ranging from 1 (strongly disagree) to 9 (strongly agree). We excluded one of the two con-trait items (i.e., Our society is getting worse every year) because even after recoded, it negatively correlated to the sale. The scale was highly reliable (α = 0.75).

Meritocracy
As a second measure of legitimation of the national social systems, we included perception of meritocracy (Jost & Hunyady, 2005). It was measured using a three-item scale developed specifically for the Chilean context (e.g., In Chile people are retributed by their effort; Castillo et al. 2019). The items were measured using a scale ranging from 1 (strongly disagree) to 5 (strongly agree). This scale was highly reliable in our sample (α = 0.75).

Prejudice toward Immigrants
We also included a seven-item scale of general prejudice toward immigrants (Wagner et al., 2006) being adapted for the Chilean context (e.g., When jobs get scare, the for- eigners living in Chile should be sent home). All items were assessed in a scale ranging from 1 (strongly disagree) to 7 (strongly agree). This scale was highly reliable in our sample (α = 0.83).

Prejudice toward Peruvian Immigrants
To measure prejudice toward Peruvian immigrants, we used the scales of Subtle and Blatant prejudice (Pettigrew & Meertens, 1995)

Prejudice as Behavioral Intention of Donation
At the end of the questionnaire, we presented a new page to participants explaining that experimenters had decided to draw a prize of 10,000 Chilean pesos. 1 In addition, we told them that if they desired, they could donate a part of that amount for an organization advocated to support immigrants in Chile. We specified that if they were winners in the draw, the experimenters will donate the exact amount indicated by the participant and the rest will be sent to them.

Nationalism
A six-item scale (Kosterman & Feshbach, 1989;Li & Brewer, 2004) measure was translated and adapted for the Chilean context (e.g., Foreign nations have done some very fine things but it takes Chile to do things in a big way). All items were assessed using a scale ranging from 1 (strongly disagree) to 7 (strongly agree). We dropped one of the items because of reliability analyses (i.e., It is NOT important that Chile be number one in whatever it does), although the final coefficient was not as high as the rest of the measures (α = 0.54).

Patriotism
As a control variable, we included patriotism using a scale developed altogether with nationalism scale (Kosterman & Feshbach, 1989;Li & Brewer, 2004). We translated and adapted this measure to the Chilean context (e.g., The fact I am Chilean is an important part of my identity) using a scale ranging from 1 (strongly disagree) to 7 (strongly agree). This measure was highly reliable in our sample (α = 0.72).

Collective Emotions
Finally, as both the manipulation check and the mediator variable, we included Pride as the collective emotion felt in the situation described at the beginning of the study (i.e., football match vs. Shopping, depending on the condition). We used a scale developed to assess the intensity of pride (Harth, Kessler, & Leach, 2008) using three items (i.e., pride, success, and happiness) ranging from 1 (with low intensity) to 9 (with high intensity). We chose this scale because it included items related to emotions that were likely to be experienced after participating in nationalistic collective rituals. This scale was highly reliable in our sample (α = 0.79).

Results and discussion
We compared experimental and control groups regarding nationalism and patriotism at the baseline, and collective emotions after the manipulation using t-test analyses. As expected, the two groups were not significantly different before the manipulation, in terms of Nationalism, t (116)  Next, to address whether nationalism moderated the effect of participation in collective rituals on the outcome variables (Hypothesis 4), we ran a series of linear regression models with Nationalism, the experimental condition and their interaction as independent variables, adjusting for the effect of Patriotism to isolate the ingroup glorification aspect of nationalism as it is usually done in similar studies (Roccas, Klar, & Liviatan, 2006). All continuous independent variables were grand-mean centered. Nationalism appeared as a significant predictor of system justification, b = 0.43, t(113) = 3.28, p = 0.001, but neither condition, b = 0.04, t(113) = 0.24, p = 0.813, nor the interaction between condition and Nationalism were significant predictors, b = -0.20, t(113) = -1.16, p = 0.247. Nationalism also significantly predicted meritocracy, b = 0.61, t(113) = 3.64, p < 0.001, but condition did not, b = -0.06, t(113) = -24, p = 0.812. The interaction between condition and Nationalism was marginally significant, b = -0.39, t(113) = -1.74, p = 0.084. A simple slope analysis (Aiken & West, 1991) showed that the effect of condition was not significant at either high scores of Nationalism, +1 SD, b = -0.46, t(113) = -1.40, p = 0.160, or low scores of Nationalism, -1 SD, b = 0.35, t(113) = 1.06, p = 0.290, although the trends were positive and negative, respectively.

General discussion
Based on the literature on collective rituals (Durkheim, 1976;Fischer et al., 2013), in the present paper, we argued that major sport events are likely to enhance social identity over individual identity among its participants. Given this effect on identity, we expected these events to increase perceptions of fairness and legitimacy as a form to preserve a positive collective self-concept. In line with social identity theory (Tajfel, 1974), this mechanism is also hypothesized to lead to ingroup bias (i.e., preferring ingroup members over outgroup members ;Tajfel, 1969) and outgroup derogation. Consequently, such collective rituals are expected to increase prejudice toward immigrants.
In Study 1, we provided empirical evidence supporting that national football matches can be conceived as a form of nationalistic collective ritual. Indeed, interest in football predicted national identification a week before an international tournament. We also found that this association was mediated by socio-cognitive processes (i.e., fusion of identity) but not emotional processes (i.e., positive collective emotions). In Study 2, we manipulated participation in collective rituals and observed that these rituals increased collective pride. However, we did not find support for direct effect of this collective ritual on prejudice toward immigrants and legitimation of national social systems.
In addition, in Study 2, nationalism was tested as a moderator of the effect of collective ritual on prejudice and system justification. On the one hand, international football matches in South America are highly competitive because they involve a situation in which different teams compete for scarce resources. Under these zero-sum situations, intergroup conflict and prejudice are likely to arise (Huddy, 2003). In such a context, making salient the national ingroup should result in outgroup prejudice (Piotrowski et al., 2019). On the other hand, previous studies have shown that flag exposure led to less outgroup prejudice among highly nationalistic individuals (Butz, Plant, & Doerr, 2007). In Study 2, the interaction between nationalism and participation to collective rituals was not significant on most of the measures. However, it was significant on prejudice assessed as donation. Indeed, making salient the participation in nationalistic collective rituals led nationalistic individuals to behave as non-nationalistic people. These results are coherent with research in flag exposure (Butz, Plant & Doerr, 2007). Specifically, when exposed to the U.S. flag, nationalistic citizens from that country were less hostile toward Arabs and Muslims. The rationale behind this attitudinal response is that nationalistic individuals are highly motivated to behave in a way coherent with national values because their national ingroups are more central to the definition of their selfconcept (Roccas, Klar & Liviatan, 2006). Given that most part of social groups are defined using positive featuresbecause of a need to preserve a positive self-conceptthese values are usually associated with integration and nondiscrimination (Brown, 2000). Indeed, Becker et al. (2017) found that national flags were mostly associated with positive concepts across 11 countries. The main difference between our study and those conducted by Butz, Plant, and Doerr (2007) is that we focused on basic collective rituals. If we consider this difference, we may conclude that nationalistic collective rituals are maybe not as competitive as we previously assumed. Although there is not much research testing this prediction, using correlational data, Kersting (2007) showed that 2006 Germany Football World Cup contributed to patriotism in sport, but it did not predict xenophobia.
The reason why such an interaction is observed on prejudice measured as intention of donation, and not on other measures of prejudice might be that this measure was included at the end of the questionnaire-even after the demographic data. In that sense, participants might have been less concerned with social desirability biases that with other measures. In addition, prejudice assessed as donation is a measure of a behavioral intention-and not a mere attitude as the rest of the scales included in the study. Although attitudes are usually related to behaviors, this association is far from being perfect (Fabrigar & Wegener, 2010). Furthermore, behaviors can be also influenced by normative variables (Ajzen & Fishbein, 1973), which in our study might be reflected in the fact that even though nationalistic individuals have negative attitudes toward immigrants, they are willing to support an organization advocated to support them because national values compel them to do so.
It is worth noting that participation in a collective ritual has also increased emotions such as pride, which can be conceived as a form of reaffirmation of social self-integrity and positivity. As Cohen and Sherman (2014) posit, a common definition of self-affirmation is an act that evidence self's adequacy. We argue that the same definition might be used when treating collective processes. In that sense, experiencing positive emotions related to the success of the national football team can affirm the collective sense of adequacy. Indeed, studies show that self-affirmation about reductions in defensiveness reaffirmed participants needs to show less ingroup bias and also less outgroup prejudice (for a review, see Cohen & Sherman, 2014). We suspect that our results might have been driven by the same processes at the collective level, particularly because groupaffirmation increases the salience of social identity (Badea & Sherman, 2019). Nevertheless, future research should provide stronger and direct evidence of this process.

Limitations and future research
One important limitation of the present research is that the interaction between nationalism and participation to collective ritual was only observed for the subtler form of prejudice (i.e., prejudice as a donation to an organization advocated to support immigrants). Given these results, we encourage future research to use less direct attitudinal measures than the one we used in the present research (e.g., the implicit association test, Greenwald, McGhee, & Schwartz, 1998). Moreover, we cannot ensure our results are due to participating in nationalistic collective ritual and not mere nationalistic priming or individual affirmation through enhancing the sense of belonging to a specific group such as family or friends. This is an important limitation because current research has shown that positive emotions usually associated with collective rituals have been observed also when people share with other people in small groups in daily activities (Gabriel et al., 2019). In order to discard these alternative explanations, we should conduct research including a nationalistic nonritual condition and manipulating the composition of the group that participates in the ritual.
Finally, alternative explanations need to be ruled out in future research. Particularly, participants who once experienced group-affirmation were probably less motivated to express outgroup derogation given that the need for a positive social identity was already met. This argument is coherent with the argument proposed by Brewer (1999) who indicated that ingroup bias does not necessarily lead to outgroup derogation. In the case of the minimum group paradigm, for instance, participants do not have any other mean to achieve a positive social identity than expressing outgroup derogation, but that was not the case in the present study. Indeed, a brief content analysis revealed that about 75% of participants in the experimental condition referred to football matches in which the Chilean national team resulted as winner. For this reason, future research should test the effect of collective rituals that do not involve a "wining" situation. Flag exposure could be a quite efficient way to do so (Butz, Plant & Doerr, 2007).
Finally, because of our sample sizes we could not test social status as a potential moderator of the above effects. Indeed, previous findings indicated that high-status groups-such as white and men in the U.S.-are perceived as more prototypical of national identity than low status groups (Carter & Pérez, 2016). In that sense, it seems reasonable to argue that the above effects could be stronger among high social status individuals than among low social status individuals.