Outgroup fanship in Australia and New Zealand

Abstract

The hypothesis was explored that, as a consequence of the asymmetrical relationship that exists between Australia and New Zealand, Australians will more consistently support New Zealand against third-party sports teams than will New Zealanders support Australia. The hypothesis was supported in two studies in which participants in each country rank-ordered a list of 12 teams (including Australia and New Zealand) from most to least favoured to win various sporting events. New Zealanders were more polarised in their rankings of Australia than vice versa, and the rankings were related to the similarity participants perceived between the two countries. The results are consistent with differences in the amount of attention the two countries pay to each other and with the way success or failure in third-party competitions may impact on the social identity and self-esteem of sporting fans in the two countries.

I support two teams: [New Zealand], and whoever is playing Australia

–New Zealand T-Shirt

Allegiance to sports teams is one of the most common and powerful social experiences for many individuals. People derive some of their self-esteem from the sports teams they support (e.g., Cialdini et al., 1976; Hirt, Zillman, Erickson, & Kennedy, 1992). Social identity theory (SIT; e.g., Tajfel & Turner, 1979), in particular predicts that to the extent people see a sports teams as part of their ingroup, the team's performance should have significant consequences not only for how they feel about the team, but also for their self-esteem and their sense of positive social identity. Hirt et al. (1992), for example, found that sports fans who witnessed the victory of their university's basketball team rated their own self-esteem and likelihood of success in future endeavours as higher than fans who witnessed their home team lose. These effects should be especially strong in sporting events involving traditional rivals.

However, how will people feel toward those rivals when they compete against other teams? This question, which we term “outgroup fanship”, is interesting in part because it arouses strong but competing intuitions. On the one hand, individuals might want their traditional rivals to fail, even if it is not at the hands of their home team, as expressed in the sentiment quoted above. This prediction follows from classic theories of cognitive consistency, such as cognitive dissonance (Festinger, 1957) and Heider's (1958) balance theory. In an empirical illustration of balance theory, Aronson and Cope (1968) found that participants offered more help to an individual who criticised a harsh experimenter, or who had praised a friendly experimenter, compared to an individual who offered feedback inconsistent with the nature of the experimenter. An extension of the principle that “my enemy's enemy is my friend” suggests that people will tend to favour any groups that are in conflict with their rivals or else risk uncomfortable cognitive imbalance.

The same conclusion has been reached by more recent research in group dynamics, which has documented outgroup derogation even when there is no direct benefit to the ingroup (e.g., Allen & Wilder, 1975; Brewer & Silver, 1978; Brown, Collins, & Schmidt, 1988; Forgas & Fiedler, 1996; Turner, Brown, & Tajfel, 1979). For example, Turner, et al. (1979) found that people assigned to arbitrary groups discriminated against an outgroup in a monetary distribution task without any rational reason for doing so. Indeed, they often preferred strategies designed to maximise the differences in profits assigned to ingroup and outgroup members even when these strategies resulted in a smaller absolute gain for their ingroup. In a similar way, in the context of an athletic competition, people may be motivated to maximise their rival's losses, regardless of whether they directly increase the success of their own team.

On the other hand, when their home team is not directly involved in a competition, people might hope for a rival team to succeed. One mechanism by which this reversal of fanship could occur is through recategorisation of the rival as ingroup when it competes against a third team. Haslam, Turner, Oakes, McGarty, and Hayes (1992), for example, manipulated Australians' expressed stereotypes about the United States, typically a negatively perceived outgroup, by manipulating the frame of reference in which judgments were made. When the United States appeared in the study with Australia, Britain, Iraq, and the Soviet Union, it was judged more positively than when the United States appeared with Australia and Britain alone, presumably because, in the context of non-English speaking and non-Western nations, the United States was temporarily recategorised as an ingroup, with corresponding benefits to its national prestige (cf. Gaertner, Mann, Murrell, & Dovidio, 1989). Thus, in the context of a third, mutual rival, people may treat even a negatively evaluated outgroup as an ally, and use it as a potential source of self-esteem when it succeeds.

However, the ease with which a rival can be recategorised may depend on a number of individual differences (e.g., chronic self-esteem, interest in sports) and national differences (e.g., size, history, international status, national character). The purpose of the current studies is to examine the latter, using as test cases traditional sports rivals Australia and New Zealand. We expected different patterns of outgroup fanship, based on the following reasoning. A consideration of the two countries' prominence in the world suggests that Australia looms much larger in the national consciousness of New Zealand than vice versa. As just one indication of the asymmetry of the Australia – New Zealand relationship one can consider the frequencies with which they are mentioned in each other's largest national newspapers, The Australian and The New Zealand Herald (Europa World Year Book, 2002). According to The Australian's online archives (http://www.theaustralian.news.com.au) “New Zealand” and its variants (e.g., “New Zealanders”) appeared in the newspaper on average 4,040 times per year during the 5 years (1997 – 2002) sampled. In contrast “Australia” and its variants appeared in the Herald 8,440 times per year during the same period (http://www.nzherald.co.nz).

Though admittedly limited, these frequency data suggest that New Zealand's culture, achievements, and concerns arouse comparatively less public interest in Australia, compared to those of Australia in New Zealand. The fact that New Zealanders pay more attention to Australia than vice versa is important, because attention plays a crucial role in judgments of similarity, influencing the relative weight given to both common and distinctive features in overall perceived similarity (e.g. Halberstadt & Niedenthal, 1997; Hogg & Turner, 1987, Krushke, 1992; Nosofsky, 1986). Because similarity in turn is the basis of most models of social categorisation (e.g., Niedenthal, Halberstadt, & Innes-Ker, 1999; Smith & Zarate, 1992) differential attention may affect the ability to recategorise a traditional rival. Smith and Zarate (1992), for example, illustrate how attention to racial differences increases the tendency to use racial categories by increasing both the similarities between members of the same race, and enhancing differences between members of different races, respectively. Analogously, we propose that New Zealanders' greater attention to their relationship with Australia will increase the weight given to the perceived similarities and dissimilarities between the two countries, and will make it harder for fans in the smaller country, New Zealand, to recategorise the larger and more prominent country, Australia, as ingroup when it competes against third parties. In contrast, it should be easier for Australian fans, who are less informed about and pay less attention to New Zealand, to recategorise New Zealand as ingroup when it competes against third parties. In other words, we predict that New Zealanders should report more extreme and polarised attitudes toward Australia because heightened attention increases the importance both of similarities and of differences between the two countries (Halberstadt & Niedenthal, 1997).

We tested the prediction that New Zealanders' attitudes to Australian sporting teams would be more extreme (i.e., more polarised) than Australians' attitudes to New Zealand's teams by asking Australians and New Zealanders to rank order a list of 12 countries, including Australia and New Zealand, on which they would want to see win particular international sporting competitions. Although we expected both New Zealanders and Australians to favour their own teams to win the hypothetical competitions, we predicted they would differ in their attitudes toward each other's success against third teams, with Australians giving more consistent support to New Zealand, but New Zealanders giving more polarised and variable support to Australia in such contests.

Study 1

Method

Participants

Thirty-six male and 105 female students at the University of Otago (Dunedin, New Zealand) participated as part of the internal assessment for their first-year psychology course. A separate sample of 64 male and 168 female students at the University of New South Wales (Sydney, Australia) were recruited as part of their coursework requirements in a social psychology class.

Materials

On a paper-and-pencil questionnaire participants were instructed to imagine that the Rugby World Cup, the Soccer World Cup, or the Davis (tennis) Cup were taking place, and to indicate which of 12 nations (Australia, Argentina, Canada, England, France, Germany, Ireland, Italy, New Zealand, Russia, South Africa, and the United States, all of which could plausibly field teams in any of the target sports events) they “hoped to win” the event by writing a “1” next to the country. Countries were listed in a single column down the center of the page. Next, participants were asked to imagine that their favoured team had already been eliminated from the competition, and to indicate which team they hoped would win in this situation, by writing a “2” next to the country. They continued this process until they had ranked all of the countries.

In addition, participants recorded their age, gender, citizenship, and length of time they had lived in Australia/New Zealand. They also indicated how many professional rugby/soccer/tennis games they had attended (“none”; “fewer than 10”; “between 10 and 50”; “more than 50”) and how interested they were in the outcome of the target sports event, on a scale from 1 (I don't care at all) to 10 (I care enormously).

Procedure

Participants were randomly assigned to one of the three sports events. New Zealanders completed their questionnaires in individual experimental cubicles, following several other, unrelated experimental tasks. The order in which the countries were presented (alphabetical vs. reversed alphabetical) was counterbalanced across participants. The correlation between the average country ranks given by the two orders was extremely high, r(12) = .97, and so Australian participants were given only the alphabetical version of the questionnaire, and subsequent analyses were collapsed across order. The procedure and instructions were the same for the two samples, except that Australians were tested in their laboratory groups.

Results

Given the importance of national identification in this study, only data from New Zealanders and Australians who were born in their respective countries (ns = 116 and 134, respectively) were used in the primary analyses, which concerned their rankings of their own country and of their rival country (“home” and “rival,” respectively). The original 12 ranks were collapsed into four categories (ranks 1 – 3, ranks 4 – 6, ranks 7 – 9, and ranks 10 – 12, re-coded 1 – 4, respectively) to reduce the considerable error variance associated with a complex serial rank-ordering task. p was set to .05 for all statistical tests in both studies.

The average rankings given to home and rival teams were analysed in a 2 (Team: home vs. rival)×2 (Participant nationality: Australian vs. New Zealander)×3 (Sport) mixed-model analysis of variance (ANOVA), with the first factor treated as a repeated measure. The analysis revealed main effects of team, F(1,244) = 425.24, and sport F(2,244) = 12.58, as well a Team×Sport interaction, F(2,244) = 4.47. Participants in both samples ranked their home teams higher than their rival teams in all sports, but the difference was greatest for soccer. The mean rankings appear in Table I.

Table I. Mean home and rival team rankings in three sports by Australian and New Zealander participants: M(SD)

	Australian participants		New Zealander participants
	Home team	Rival team	Home team	Rival team
Rugby (Study 1)	1.03 (.17)	2.13 (1.12)	1.02 (.15)	2.23 (1.24)
Tennis (Study 1)	1.09 (.29)	2.23 (1.11)	1.00 (.00)	1.78 (1.11)
Soccer (Study 1)	1.05 (.22)	2.86 (.95)	1.29 (.71)	2.60 (1.24)
Soccer (Study 2)	1.06 (.23)	1.97 (.82)	1.40 (.86)	2.29 (1.21)
Note. Data represent the mean category participants used to rank their home team (i.e., Australia for Australians and New Zealand for New Zealanders) and rival teams (New Zealand for Australians; Australia for New Zealanders) among 12 nations in the indicated sporting competitions. Category 1 = ranks 1 – 3; Category 2 = ranks 4 – 6; Category 3 = ranks 7 – 9; Category 4 = ranks 10 – 12. Lower numbers indicate better rankings.

Chi-square analyses were used to examine the frequency distribution of ingroup and outgroup rankings as a function of experiment sample. Australians and New Zealanders did not differ in their ingroup rankings, χ²(3) = 3.65, p > .05, with the vast majority (94.4% and 95.5%) ranking their home countries in category 1 (91.4% and 88.1% in fact hoped their home country would win the target event). However, the two groups showed different patterns of outgroup rankings, χ²(3) = 11.54, p < .05. As seen in Figure 1, Australians gave consistently high rankings of New Zealand, whereas New Zealanders' rankings of Australia were more polarised. Levene's test for equality of variance indicated that New Zealanders' rankings were marginally more variable than Australians' overall, and significantly more so in the case of soccer ratings.

Figure 1. Distribution of Australians' rankings of New Zealand, and of New Zealanders' rankings of Australia (%), on a list of 12 teams, averaged across three sports, Study 1. Lower numbers reflect better rankings

Finally, to examine potential individual differences in rival rankings, these rankings were correlated, for the two samples separately, with participants' interest ratings and the number of games they reported having attended. Games and interest were correlated, rs = .36 and .59 for Australians and New Zealanders, respectively, but only interest correlated with rival rankings, rs = .27 and .19. That is, the more interested participants were in a sport, the worse they ranked their country's rival.

Discussion

Previous research and theory led to competing intuitions regarding fanship toward a traditional rival when in competition with another team. We argued that support for a rival team would depend in part on the relationship between the countries involved, and in particular on the importance of the rivalry to the individuals in those countries. Australia and New Zealand provide an interesting test case due to their relative size and salience: New Zealanders are more conscious of, and more concerned with, Australians than vice versa. We predicted this cultural attentional difference to be reflected in more polarised rankings of Australia by New Zealanders than of New Zealand by Australians.

Indeed, although the average rankings did not differ between the two countries, an analysis of the distribution of those of rankings revealed very different patterns in the two samples. New Zealanders were more polarised in their rankings of their rival than were Australians, a phenomenon further evidenced by greater variance in the former's judgments.

Interestingly, fanship was not strongly predicted by the idiosyncratic importance of the sports, as found in some previous research. Reeves and Tesser (1985), for example, using a methodology similar to the current paradigm, found that students more strongly preferred close opponents of their university (based on self-reports) to win in a high-relevance sport (football) than a low-relevance sport (water polo). Branscombe and Wann (1992) reported outgroup derogation and increases in blood pressure in U.S. participants who watched a boxing match between an American and a Russian, but only for those who strongly identified as Americans. Wann and Branscombe (1995) found that arousal caused ingroup favouritism with respect to the University's basketball team, again only for participants reporting high identification with the team. In the current study, the psychological relevance of a sport, measured as a person's interest in its outcome, weakly predicted worse rankings of rivals, but this was true for both samples and so cannot explain the differences in the distributions. Furthermore, if individual differences account for some of the variance, the question remains as to the origin of those individual differences and how they translate into the polarised outgroup rankings observed in New Zealand.

A stronger factor in the current results, we believe, is the cultural differences in the attention given to the rivalry between the two countries, which influences perceived similarity and in turn the differential ease with which a traditional rival can be recategorised as ingroup when competing against third countries. Although New Zealanders' polarised rankings are consistent with their greater inability or unwillingness to recategorise Australia as ingroup in such contests, no direct measure of perceived similarity was included in the study. Therefore, we report a second study that sought more direct evidence for the mechanism of differential attention that may be implicated in Australians' and New Zealanders' different attitudes towards their rivals. Study 2 is a replication of Study 1 but also incorporates direct assessment of international similarities; we predict the perceived similarity between Australia and New Zealand to be more variable among New Zealanders than among Australians, and that similarity judgments will predict outgroup rankings, providing direct support for the role of similarity and categorisation in outgroup fanship.

Study 2

Method

Participants

Seventy-five (21 male, 54 female) students in New Zealand and 61 (12 male, 48 female, one unidentified) at the University of New South Wales were recruited as described in Study 1.

Materials

A three-part paper-and-pencil questionnaire asked participants, first, to consider the countries of Australia, England, Ireland, New Zealand, and the United States. All participants were asked to rate the overall similarity between all 10 possible pairs of the five countries, using a 1 – 10 scale anchored at not similar at all and very similar. The second part of the questionnaire was identical to the ranking task used in Study 1, except that only soccer (the sport associated with the largest difference between home and rival rankings in Study 1) was used as a target sport. Countries were presented in four columns of three countries each. The third part of the questionnaire asked participants, as in the first study, to record their age, gender, citizenship, and length of time they had lived in Australia/New Zealand and to indicate how many professional soccer games they had attended and how interested they were in the outcome of the soccer World Cup.

Procedure

The procedure was identical to that used in Study 1, except that the countries were presented in alphabetical order for all participants.

Results

As in Study 1, only data from New Zealanders and Australians who were born in their respective countries (ns = 65 and 37 respectively) were analysed. A t test on the critical Australia – New Zealand similarity ratings indicated that although New Zealanders' judgments were both higher and more variable than Australians (M = 8.29 vs. 7.86; SD = 1.64 vs. 1.21) neither difference reached significance. Rankings, recoded again into four categories, were analysed in a 2 (home vs. rival)×2 (participant nationality) mixed-model ANOVA, with the first factor treated as a repeated measure. Mean rankings appear in Table I. The analysis revealed main effects of team, F(1,90) = 47.56, reflecting better rankings for home than for rival teams (1.28 vs. 2.18).

As in Study 1 separate chi-square analyses were used to examine the frequency distribution of home team and rival rankings between the two samples. As in Study 1, Australians and New Zealanders did not differ in their home team rankings, with the majority of participants ranking their home countries in category 1. However, the two groups had different distributions of outgroup rankings, χ²(3) = 9.98. As seen in Figure 2, Australians' rankings were more consistent in their support for New Zealand than vice versa, and Levene's test for equality of variance confirmed that New Zealanders' rankings were significantly more variable than Australians', F = 18.23. Furthermore New Zealanders, but not Australians' outgroup rankings were correlated with their similarity ratings, r(63) = −.30 versus r(29) = −.12, such that greater perceived similarity predicted better team ranking. These rankings were not correlated in either sample with either interest in the outcome of the world cup or number of soccer games attended.

Figure 2. Distribution of Australians' rankings of New Zealand, and of New Zealanders' rankings of Australia (%), on a list of 12 teams in the Soccer World Cup, Study 2. Lower numbers reflect better rankings

Discussion

As in Study 1, New Zealanders showed less consistent and more polarised support for Australia in third-party competitions, whereas Australians more consistently supported New Zealand in such contests. Furthermore, similarity judgments predicted outgroup rankings, providing a mechanism for the effects. Due to differences in importance and weight given to New Zealand's relationship with Australia, New Zealanders pay more attention to, and are more likely to be conscious of both similarities and differences between the two countries. This in turn makes it more difficult for New Zealanders to recategorise Australia consistently as ingroup when it faces third-party opponents. Interestingly, the relationship between similarity and outgroup ranking was stronger in New Zealand than it was in Australia, a finding possibly related to the reduced power of the Australian sample, which should be explored in a replication of the current studies. It also remains for future research to take the logical next step: to link attention directly to ingroup and outgroup judgments. Previous theory and research on the similarity – categorisation relationship clearly suggest the presence of this link, which should, in turn, lead to corresponding benefits (e.g., biased allocation of resources) of ingroup membership.

General discussion

There is abundant evidence for the importance of sporting teams as ingroup representatives, and for the self-esteem benefits of supporting those teams in competitions against salient outgroups, but there is no prior research on what determines support for traditional opponents when they compete against third teams. Although a simple account of outgroup fanship, suggested by the quote at the start of this article, as well as cognitive consistency theories, predicts support for any competitor against a traditional rival, the fact that outgroup status is itself flexible suggests that the relationship between rivals is more complicated. We argued that, among other factors, the relative importance of the existing rivalry for ingroup identification and esteem should predict outgroup fanship. Australia and New Zealand were used as test cases because these countries appear to have an asymmetric relationship on this dimension: comparisons between the two countries are more salient and important in New Zealand than in Australia. Cognitive models of attention and categorisation predict that differential attention should enhance both similarities and differences between stimuli. Self-categorisation theory predicts that perceived similarity will affect the likelihood of recategorisation of a former rival as ingroup, and in turn the likelihood that the rival will be supported in competitions against other teams. As predicted, in both studies New Zealanders had more polarised attitudes toward Australia than Australians did toward New Zealand.

Future research could use more direct mathematical models (e.g., multidimentional scaling) to quantify and assess the role of attention in social categorisation and outgroup fanship. It is also important to replicate the current results in other cultures with a similar asymmetric relationship (e.g., the United States and Canada) to confirm that they are indeed due to attentional differences associated with asymmetries in cultural perceptions, rather than particular features of Australia and New Zealand. These studies, which should include direct measures of variables such as the identification with and status of the ingroup, the psychological closeness of the outgroup, and the idiosyncratic importance of the target sports, would help identify the cognitive, motivational, and individual difference variables that more precisely predict the extent of outgroup fanship.

The current studies were designed to examine the effects of cultural differences in attention on outgroup rivalry; an explanation of the origins of those cultural differences goes beyond the scope of this paper. One interesting possibility lies in the power Australia and New Zealand hold with respect to each other. Fiske (1993) has argued that power itself is associated with less attention to individuating information and greater use of stereotypes. People, she argues, have little need or motivation to attend to less powerful individuals because their control over them does not depend on it. Powerless individuals, on the other hand, need to attend to powerful others in order to predict their behaviour. Although Fiske's research focuses on individuals, the dynamics of power she identifies may have an analogy at the group level. When their group membership is salient, as in the current studies, New Zealanders may feel more need to attend to their relationship with their more powerful neighbour than is the case in reverse.

However attentional differences come about, they have important implications for cross-cultural attitudes, stereotyping and conflict. The more salient the comparison to a particular country, the stronger the opinions toward that country – positive or negative – are likely to be. At the same time, the current studies indicate that opinions toward particularly salient outgroups should be relatively malleable via changes in perceived similarity, among other factors. Thus, the bigger obstacle to peaceful international relations may not be negative attitudes per se, but rather a lack of concern that those attitudes are important. Sports fanship provides a rich, benign, real-world testing ground for such speculations, as well as for group and individualist theories of self-esteem management in general.

Acknowledgments

We thank Steve Catty and Emma Witt for help with data collection.