An examination of sport fandom in the United Kingdom: a comparative analysis of fan behaviors, socialization processes, and team identification.

Parry, Keith D. ; Jones, Ian ; Wann, Daniel L. 等

Since Wann and Hamlet (1995) commented upon the lack of literature focused upon the sport fan, there has been a substantial increase in theoretical and empirical research on the subject, leading to considerable advances in knowledge and understanding. There is still, however, limited cross-cultural research, and according to Theodorakis and Wann (2008), a need to explore cultural differences in fandom. This is particularly true given that existing research indicates such differences to be present in the motivations and behaviors of sport fans and spectators from different ethnic groups (Bilyeu & Wann, 2002; Wann, Bilyeu, Breenan, Osborn & Gambouras, 1999), as well as different nationalities (Melnick & Wann, 2004, 2011; Theodorakis & Wann, 2008). To date, a limited number of cross-cultural studies have been undertaken. Wann, Melnick, Russell and Pease (2001) report findings from data collected from North American fans, and further empirical studies have tested Norwegian (Melnick & Wann, 2004), Greek (Theodorakis & Wann, 2008), and Australian fans (Melnick & Wann, 2011). The studies have all explored similar aspects of fandom: a) fan related behavior (such as watching sport on television, b) the role of various agents in the socialization process (such as parents or the media), c) gender differences, and d) the extent of fan identification with a favorite team. The following paragraphs highlight some of the more pertinent findings from these research endeavors.

Fan Related Behavior

Findings on fan related behavior demonstrate--in most cases--similarity across contexts, with males consistently demonstrating greater frequencies than females for all types of behavior. Males from all studies to date and Australian female fans demonstrate high levels of television spectatorship, with 81% of Norwegian male fans, 86% of Greek male fans, 97% of Australian male fans, and 86% of Australian female fans watching sport at least once a week. Female television viewing is lower for Norwegian (66%) and Greek female fans (34.8%). Listening to sport on the radio is, in most cases, comparatively rare when compared to television viewing. It is a much more popular activity amongst Greek males, of whom 61% listen at least once a week, and the entire Australian sample, where 59% of males and 38% of females listen to sport at least once a week, compared to only 15.9% of Greek females, 16% of Norwegian males, and 11% of Norwegian females.

The increased frequency of fan behaviors among Australian fans is also reflected in terms of live attendance at sporting events. Half of male Australian fans (50%) and 31% of female fans attend sporting events at least once a week according to Melnick and Wann's (2011) study, compared with 31% and 10% respectively of Norwegian fans, and only 10.1% of male and 4.5% of female Greek fans. All studies, however, demonstrate the importance of discussing sport with others, with 97% of Australian fans and 91% of Norwegian fans doing so at least once a week. Theodorakis and Warm (2008) found that male Greek fans follow this pattern, with 86.3% discussing sport once a week (however, only 54.7% of Greek females did so). The final behavior explored was that of accessing sport on the internet. This was only been undertaken by Melnick and Wann's (2011) Australian study, which demonstrates relatively high levels of internet use amongst Australian males, with 78% accessing sport at least once a week, compared to 44% of females.

The Role of Various Agents in the Socialization Process

Each study highlights friends and peers as key socialization agents. There is less consistency with regard to the role of parents, school and community in the socialization process. The North American data reported by Wann et al. (2001) highlights the importance of the school, followed by parents, and finally community. Greek fans report that parents and community are the second and third most important agents. The variability between each, however, is low. Australian fans, as with North American fans, also report the community as least significant; however they rate all agents highly, with little variation between them. Norwegian fans are shown by Melnick and Wann (2004) to be most strongly influenced by the local community, and both Greek and Norwegian fans cite the school as the least important socialization agent reflecting, to some extent, the different sporting structures within each context, such as the emphasis upon club sport in Greece and Norway, as opposed to the emphasis upon school based sport in the United States. In terms of the single most influential factor, all studies to date highlight the importance of the father as a socialization agent. The Greek and Australian studies both report 45% of respondents suggesting that their father had been the most influential agent, with 30% of Norwegian fans citing this as the case.

The Extent of Identification with a Favorite Team

Previous cross-cultural research also targeted team identification, which is the extent to which a fan feels a psychological connection to a team (Wann et al., 2001). Levels of identification for Australian fans have been found to be very high, with fans supporting an Australian team scoring a mean of 40.11 (out of a possible 56) on the Sport Spectator Identification Scale (Wann & Branscombe, 1993), and those following a non-Australian team scoring 42.07. These are considerably higher than Greek (M= 36.9) and Norwegian fans (M= 31). Australians also were more likely to have a favorite team (92%), compared to 87.5% of Greek fans and 65.2% of Norwegian fans. Most Greek fans supported a team based in their homeland compared to 90% of Australian fans and only 64% of Norwegian fans. As Melnick and Wann (2011) suggest, this may be due to the absence of a professional sports industry in Norway, whereas both Greece and Australia possess high profile professional sport teams.

The Current Investigation

Johnes (2005) suggests that the United Kingdom (UK) was the birthplace of modern sport. Through the rationalization and codification of folk games in the nineteenth century (a process Elias, 1986, terms "sportization"), many of the sports that are played around the world saw their birth in the UK. Yet despite this sporting tradition, recent figures (Sport England, 2011) indicate that only slightly over one third of the adult population participate in regular (once per week) sporting activity. Although this percentage has risen since 2006, there has been a decrease in the number of young people (aged 16-19) who regularly participate in sport during the same period. Swimming, association football, athletics, and cycling are the most participated in sports as identified in the most recent (2010-11) survey of sports participation rates amongst adults (Sport England, 2011). In July 2005, London won the right to host the 2012 Olympic Games. The London 2012 bid was backed by the Government as it was believed that hosting the Olympic Games was one way to improve the physical and mental well-being of the nation, with additional benefits such as urban regeneration and a reduction in anti-social behavior (Hill, 2010).

However, although participation may be straggling, sport spectatorship remains high in the UK. The sporting environment for fans in the UK is dominated both in terms of attendance and media coverage by the English Premier League (association football). The combined attendance figures for top-flight English football teams were over thirteen million in the 2011-12 season (ESPN, 2012). Elsewhere, crowds at major sporting events appear to be increasing, with the flagship tennis and motorsports events (Wimbledon and the British Formula One Grand Prix) seeing record attendances in 2011 (PwC, 2011). Clearly, sport fandom and spectating play an important role in the culture and society of the UK and permeates many aspects of life.

The purpose of the current study was two-fold. First, we examined sport fandom and spectating within the United Kingdom, replicating the procedure used in the aforementioned cross-cultural examinations. Second, we provide a comparison with existing studies based upon the different cultures outlined above. Due to the lack of past research of this type on fans in the UK, establishing specific hypotheses was not appropriate (similar to the work by Melnick & Wann, 2004; Theodorakis & Wann, 2008). Rather, the current investigation attempted to examine the following research questions:

1. To what degree do male and female university students from the United Kingdom engage in behaviors commonly associated with sport fandom?

2. To what extent do parents, friends / peers, school and community contribute to their socialization into the sport fan role?

3. Which one of the aforementioned socialization agents is most influential?

4. How strongly do students identify with their favorite team?

Method

To facilitate comparison with previous studies, the methodology adopted was based upon the previous cross-cultural studies discussed above.

Participants

Participants were a convenience sample of 252 students based in the Department of Sport at a provincial university in the UK. The gender breakdown was 185 males (73.4%) and 67 females (26.6%). The mean age was 20.61 (SD = 3.26).

Measures

The questionnaire packet contained five sections. The first section asked participants to provide demographic information regarding gender, age, and year in school. In the second section participants indicated how often they attended sporting events in person, watched sporting events on television, listened to sporting events on the radio, and discussed sport with friends and relatives. Response options to these scale-items were: 1 = never, 2 = once a year, 3 = twice a year, 4 = once a month, 5 = twice a month, 6 = once a week, 7 = twice a week, and 8 = once a day.

The third section of the packet focused on fan socialization. Based on and consistent with the work of McPherson (1975; 1976), Melnick and Wann (2004) and Wann et al. (2001), four socialization agents were identified and selected as having the greatest potential impact on the sport fan socialization process: parents, friends, school and community. Respondents expressed the impact of each agent on an 8-point Likert-type scale anchored by 1 (the agent had no influence) to 8 (the agent had a great deal of influence). Additionally, an open ended question was included to allow participants to identify the most influential person or entity in their decision to be a sport fan. This additional item was justified due to the fact that forced-choice measures may be less optimal in situations assessing sport fan socialization (see Spaaij & Anderson, in press). Indeed, inconsistency between forced-choice and open ended responses have been noted (e.g., Melnick & Wann, 2011).

The fourth section contained the Sport Fandom Questionnaire (SFQ; Warm, 2002) to measure the degree of identification with the role of sport fan. The SFQ is a reliable and valid unidimensional scale comprised by five Likert-scale items. Response options ranged from 1 (strongly disagree) to 8 (strongly agree). Higher numbers represented greater levels of sport fandom. The fifth and final section of the packet focused on team identification. In this section, participants completed the Sport Spectator Identification Scale (SSIS; Wann & Branscombe, 1993). Respondents targeted their favorite team when completing the SSIS. The SSIS contains seven Likert-scale items with response options ranging from 1 (low identification) to 8 (high identification).

Procedure

After receiving permission from class instructors, questionnaires were completed by participants inside university classrooms. All subjects were informed that their participation was completely voluntary, participants could withdraw at any time, and the individual responses would be held in strict confidence. Upon completion of the study (approximately 20 minutes), the participants were debriefed and excused from the testing session.

Results

Current Sample--Sport Fandom in the UK

Initial analyses. The five items on the SFQ were combined to form a single measure of fandom (Cronbach's alpha = .90). Using a one-way analysis of variance, males (M = 34.74, SD = 4.80) reported significantly higher [F(l, 243) = 97.62,p < .001] levels of fandom than females (M= 26.40, SD = 8.04). The seven items of the SSIS were also combined (alpha = .92), with males (M= 44.01, SD = 8.91) demonstrating significantly higher [F(1, 249) = 55.50, p < 0.001] levels of identification than females (M = 33.90, SD = 11.00).

Sport fan related behavior. Males demonstrated greater frequency of participation in behaviors than females, with the exception of attending events in person (33% of females attended twice a week or more, compared to 22% of males). Females were, however, also more likely to be less regular attenders, with 36% attending less than once a month, compared to 22% of males. Males tended to watch significantly more sport on television, with over half (51%) watching every day, compared to only 8% of females. This pattern was repeated for accessing sport on the Internet, but not in terms of listening to sport on the radio (which was less common for all). Gender differences were also evident in terms of discussing sport, with 84% of males discussing sport every day, compared to 33% of females.

The relationships among fandom behaviors, team identification and sport fandom were examined using Spearman correlations. The findings demonstrated significant positive correlations (all p < .01) between team identification and all five fan behaviors (identification and attendance r = .24; watching sport on television r = .47; listening to sport on radio r = .36; accessing sport on Internet r = .43; discussing sport r = .46). Similarly, significant positive correlations (all p < .01) were found between sport fandom and the fan behaviors (fandom and attendance r = .20; watching sport on television r = .49; listening to sport on radio r = .36; accessing sport on Internet r = .54; discussing sport r = .53).

Regression analyses were undertaken to further explore the relationships among fandom, identification, and fan behavior. Each regression used one of the behaviors as the dependent variable while fandom and identification were predictor variables. Please see Table 1 for information regarding regression statistics for the predictor variables. With respect to attendance at sporting events, the combined effect of the two predictor variables was significant, F(2, 237) = 8.20, p < .01 (R = 0.25; adjusted [R.sup.2] = 0.06). With respect to independent contributions to attendance, team identification accounted for a significant proportion of unique variance while sport fandom did not. As for watching sport on television, the combined effect of the two predictor variables was again significant, F(2, 240) = 64.13, p < .001 (R = 0.59; adjusted [R.sup.2] = 0.34). With respect to independent contributions to watching televised sport, both team identification and sport fandom accounted for a significant proportion of unique variance. The regression on listening to sport on the radio revealed that the combined effect of the two predictor variables was significant, F(2, 240) = 24.51, p < .001 (R = 0.41; adjusted [R.sup.2] = 0.16). As for independent contributions to sport radio listening, both identification and fandom accounted for a significant proportion of unique variance. With respect to accessing sport on the Internet, the combined effect of the two predictor variables was significant, F(2, 241) = 64.80, p < .001 (R = 0.59; adjusted [R.sup.2] = 0.34). With respect to independent contributions to sport on the Internet, identification and fandom both accounted for a significant proportion of unique variance. And finally, for discussing sport, once again the combined effect of the two predictor variables was significant, F(2, 241) = 75.16, p < .001 (p = 0.62; adjusted [R.sup.2] = 0.38). With respect to independent contributions to sport discussions, both team identification and sport fandom accounted for a significant proportion of unique variance.

Sport fan socialization. To examine the impact of the four socialization agents, scores were analyzed using a 2 (Gender: male or female) X 4 (Socialization Agent: parents, friends, school and community) mixed factor analysis of variance. Means and standard deviations appear in Table 2. Results indicated that the gender main effect was significant (Wilks' Lambda = .934, F(4, 245) = 3.93,p < .01). However, subsequent univariate tests indicated significant gender differences only in terms of friends/peers, who were more important for male fans F(1, 249)= 13.01, p < .001), with no gender differences in terms of parents, school and local community. Table 3 reveals responses for the greatest influence on participants becoming a sport fan. Fathers (49%) were, by some margin, the most significant socialization agent, followed by friends (8%). Fewer females (33%) identified their father as most influential, however.

Team identification, favorite team, and favorite sport. Wann et al. (2001) suggest that a score of 35 or above on the SSIS can be indicative of high identification. The mean score from the SSIS was 41.31, thus demonstrating a very highly identified sample, especially among males (M = 44.01). Females were, as noted earlier, significantly less identified (M= 33.90). Most fans identified with an association football (soccer) team (77%). Rugby union (10%), cricket (2%), basketball (2%) and American football (1%) were also listed. Most respondents listed a UK-based team (96% of those who listed a team), with only 4% listing a non-UK team.

Comparison of British Sport Fans to Previous Samples

The results of the current study were compared with previous cross-cultural studies (i.e., Melnick & Wann, 2004, 2011; Theodorakis & Warm, 2008; Wann et al., 2001). Comparisons were drawn between levels of sport fandom, team identification, sport fan behaviors, and socialization agents (note that Wann et al., 2001, did not examine levels of fandom and identification).

Sport fandom. An analysis of SFQ scores showed that British fans appear to be more comparable with Australian fans rather than their nearer neighbors in Europe (see Table 4). British fans had the highest average total score, and the highest score from all groups was among British male fans (the Norwegian sample reported the lowest total mean). It is interesting to note, however, that female fans from Britain had a mean score lower than those of Australian females (26.40 compared to 30.34). This finding may be linked to falling interest in sport in the UK, with lower participation by females than males, a gender gap that appears to become more pronounced with age (Cox, Coleman, & Roker, 2006). Further, a gender difference was found in each sample, with males reporting higher levels of fandom than females.

Team identification. As shown in Table 4, British fans were highly identified with their favorite teams. Although they identified more strongly than their European counterparts, British female fans had a lower mean score than Australian females, a finding consistent with the aforementioned data on fandom. Also similar to fandom, British males had the highest mean score on team identification and, once again, gender differences were noted in each sample (although magnitude of the difference was more pronounced in the British and Greek samples).

Sport fan behavior. British fans were more likely to watch sport in person at least twice per week than their counterparts. British fans were similar to most other samples in their high level of televised sport consumption. Overall, 39% of British fans and 51% of males from this sample reported watching sport on television on a daily basis. These totals were the highest levels among all samples. With respect to comparisons of listening to sport on the radio, subjects in the current sample were less likely to do so on a weekly basis than those from Greece and Australia. The proportion of British males who discussed sport each day was considerably higher than the other cultures here with 71% of respondents indicating that they do so. Data for Internet use was only available for Australian and British samples. It was clear that British fans use the Internet to follow their teams to a greater extent than fans in the Australian study.

Socialization into the sport fan role. With respect to category rankings of the four primary socialization agents, an examination of Tables 2 and 5 reveals that there were no exact matches between the British sample and the other fan groups. However there were some similarities between Australian and British fans for the importance of parents and the importance of school for males. The importance of the local community was lower for British and American fans.

In each sample, family members, and in particular fathers, were most often listed as the primary socialization agent (see Table 3). Generally, the results from this British study were more comparable to those found in the Australian and United States samples than with the other European studies. The key differences between the British sample and those from Greece and Norway were that Greek fans reported more friends as most influential while Norway had higher percentages from friends and, especially, community.

Discussion

The current study reveals that British sport fans are highly identified, typically with a British association football team, and that they engage in fandom related behaviors more frequently than their North American, European, and Australian counterparts. The data reflect the strong sporting culture within the UK, particularly that in terms of the cultural dominance of association football as a spectator sport. The favorite sports selected by the UK sample is noteworthy as only one of the four most popular participation sports for adults (association football) provided the favourite sports team for the subjects. While swimming, cycling, and athletics are popular participation sports they did not feature in the list of favorite teams. However given the more individual nature of these sports it may not be surprising that they were not selected as a favorite.

It is interesting to note that, overall, British fans demonstrate greater proximity to Australian fans than the two, geographically closer, European nations. This may support the notion that a common European culture remains more theory than practice (Schneider & Barsoux, 2003). The proximity of behavior reflects Hofstede's (1980) suggestion that British and Australian cultures are similar while differing on a number of dimensions from both Norwegian and Greek cultures. This seems reasonable given the colonial history of Australia which, although Asian influences are becoming more dominant, is still influential in Australian sport behavior (Cashman, 2010; Schirato, 2007; Taylor et al. 2009). It has been documented that there are tangible benefits from identifying with a sport team and that this identification leads to greater levels of social interaction (Wann, 2006). As such, it might be expected that this social comiection could result in increased game attendance. However, sport fandom was not found to be a significant predictor of attendance. This evidence reinforces the findings of Wann, Brame, Clarkson, Brooks, and Waddill (2008) who, for the case of college sports, found no interaction between game attendance and identification.

Higher levels of team identification for male fans was found to result in increased indirect consumption through electronic media which suggested that the social connections for highly identified male fans may be formed through the communal watching of sport through electronic media. The consumption of sport through the media was found to be more common than attendance in person at games across all samples, reinforcing Crawford's (2004) view that indirect sport consumption is now the norm. The extent to which indirect consumption now dominates the fan experience can be see through a comparison of the frequencies of attending in person and watching sport on television. Across all samples at least 70 % of participants watched sport at least once a week via television, whereas, for attending a game this figure is below 50% and it is in the American and Australian samples that this distinction is most marked.

British fandom is also revealed to be somewhat male dominated. Specifically, male fans were more highly identified, engaged in fan related behaviors more often, and fathers were the most frequent socializing agent into sport fandom. This perhaps reflects the dominance of male football as a spectator sport within the UK (Jones, 2008). However, comparison to previous samples reveals that the socialization pattern is not confined to the UK. In all cultures, the father was the most frequently cited socialization agent, indicating that gender scripts [as discussed by Melnick and Wann (2004) in relation to their Norwegian study] may be present across cultures and are witnessed in Europe, North America, and Australia.

The apparent lack of importance placed upon community sport as a socializing agent by UK participants is worthy of future consideration. As was the case with American and Australian fans, community was ranked as the least important of the four socialization agents by both males and females (see Table 5). The lack of importance placed upon community sport may be reflected in the sport structures of these countries where large professional teams dominate the sporting landscape. The greater emphasis placed on community sport (or the lesser emphasis on professional sport) in Norway may be linked to their lower levels of obesity (WHO, 2010) and may offer consideration for the UK in light of their low sport participation figures (Sport England, 2011). The frequency of attending sports events in person and watching sport on television may also be of concern for the health of the UK population. If the frequency of consuming sport is at the expense of sport participation then, as this sample grow older (at present it is relatively young with a mean of just over 20 years of age) it is likely that their participation will decrease even further (Kokolakakis, Lera-Lopez, & Panagouleas, 2011).

Summary and Limitations

The data have demonstrated both similarities and differences across different cultures. Results revealed gender differences in behavior with males reporting greater levels of participation than females. Both team identification and fandom were significant predictors of fan behavior. The importance of the father as a socialization agent was highlighted throughout the sample, and levels of identification were generally high. Cross-cultural analysis indicated that socialization agents for the UK were more varied than other countries, UK fans were more likely to watch sport live, and UK males were more likely to watch and discuss sport daily. In general, it was demonstrated that UK fan behavior was closer to that of American and, in particular, Australian fans, rather than fans in European (Greek and Norwegian) samples.

Finally, a few limitations of the current research warrant mention. First, as with several of the previous studies serving as comparisons for the current investigation (e.g., Melnick & Warm, 2004, 2011; Warm et al., 2001), the college student sample tested here was quite homogeneous with respect to age and, perhaps, experience. That is, although perhaps unlikely (Warm et al., 2001), it remains possible that younger and older fans as well as fans without college experience may respond in a different fashion (e.g., report a differential pattern of fan behaviors and/or a different set of primary socialization agents). Further, there were several important fan behaviors that were not examined here, such as likelihood of purchasing team merchandise and propensity to respond in an aggressive manner while watching sporting events. Future investigators should add to the current work by expanding the behaviors assessed. And finally, although the list of cross-cultural analyses of fan behaviors and socialization processes continues to grow (and now totals five countries), there are certainly many more locales from which samples are needed to acquire a more complete picture of the cultural differences and idiosyncrasies that impact sport fandom.

References

Bilyeu, J. K., & Wann, D. L. (2002). An investigation of racial differences in sport fan motivation. International Sports Journal, 6, 93-106.

Cashman, R. (2010). Paradise of Sport: a history of Australian sport. Petersham, N.S.W.: Walla Walla Press.

Cox, T., Coleman, L., & Roker, D. (2006). Understanding participation in sport: What determines sports participation among 15-19 year old women? London: Sport England.

Crawford, G. (2004). Consuming sport: Fans, sport and culture. London: Routledge.

Elias, N. (1986). An essay on sport and violence, In N. Elias & E. Dunning, Quest for excitement: sport and leisure in the civilizing process (pp. 150-174). Oxford: Blackwell.

ESPN, (2012). Barclays Premier League Stats - 2011-12. Available from: http://soccemet. espn.go.com/stats/_/league/eng. l/year/2011/barclays-premier-league?cc=3436

Hill, J. (2010). Sport in History: An Introduction, London: Palgrave MacMillan.

Hofstede, G. (1980). Culture's Consequences, London: Sage.

Johnes, M. (2005). United Kingdom. In D. Levinsen & K. Christensen (eds.), Berkshire Encyclopaedia of World Sport Vol 4 (pp. 1653-1660). Great Barrington, USA: Berkshire Publishing

Jones, K. W. (2008). Female fandom: Identity, sexism, and men's professional football in England. Sociology of Sport Journal, 25, 516-537.

Kokolakakis, T., Lera-Lopez, F. & Panagouleas, T. (2011). Analysis of the determinants of sports participation in Spain and England. Applied Economics, 44, 2785-98.

McPherson, B. (1975). Sport consumption and the economics of consumerism. In D. W. Ball & J. W. Loy (Eds.) Sport and social order: Contributions to the sociology of sport (pp. 243-275). Reading, MA: Addison-Wesley.

McPherson, B. (1976). Socialization into the role of sport consumer: A theory and causal model. Canadian Review of Sociology and Anthropology, 13, 165-177.

Melnick, M. J., & Wann, D. L., (2004). Sports fandom influences, interests, and behaviors among Norwegian university students. International Sports Journal. 8, 1-13.

Melnick M. J., & Wann, D. L. (2011). An examination of sport fandom in Australia: Socialization, team identification, and fan behaviour. International Review for the Sociology of Sport, 46, 456-470.

PwC, (2011). Attendances rise at UK's biggest annual sporting events Available from http:// www.ukmediacentre.pwc.com/News-Releases/Attendances-rise-at-UK-s-biggest-annualsporting-events-lOdc.aspx.

Schirato, T. (2007). Understanding sports culture. London: Sage.

Schneider, S. C., & Barsoux, J. L. (2003). Managing across cultures. Harlow: Pearson Education.

Spaaij, R., & Anderson, A. (in press). Parents or peers: Which is it? Sport socialization and team identification in Australia: A rejoinder to Melnick and Wann. International Review for the Sociology of Sport.

Sport England, (2011). Once a week sport participation factsheet: summary of results. Available from: http://www.sportengland.org/research/active_people_survey/aps5.aspx.

Taylor, T., Lock, D., & Darcy, S. (2009). The Janus face of diversity in Australian sport. Sport in Society 12, 861-875.

Theodorakis, N. D., & Wann, D. L. (2008). An examination of sport fandom in Greece: influences, interests and behaviours. International Journal of Sport Management and Marketing. 4, 356-374.

Wann, D. L. (2002). Preliminary validation of a measure for assessing identification as a sport fan: the Sport Fandom Questionnaire. International Journal of Sport Management, 3, 103-115.

Wann, D. L. (2006). Examining the Potential Causal Relationship Between Sport Team Identification and Psychological Well-being. Journal of Sport Behavior, 29, 79-95.

Wann, D. L., Brame, E., Clarkson, M., Brooks, D. & Waddill, P. J. (2008). College Student Attendance at Sporting Events and the Relationship Between Sport Team Identification and Social Psychological Health. Journal of Intercollegiate Sports, 1, 242-254.

Wann, D. L., & Branscombe, N. R. (1993). Sports fans: Measuring degree of identification with the team. International Journal of Sport Psychology, 24, 1-17.

Wann, D. L., Bilyeu, J. K., Breenan, K., Osborn, H. & Gambouras, A. F. (1999). An exploratory investigation of the relationship between sport fan motivation and race. Perceptual and Motor Skills, 88, 1081-1084.

Warm, D. L., & Hamlet, M. A. (1995). Author and subject gender in sports research. International Journal of Sport Psychology, 26, 225-232.

Wann, D. L., Melnick, M. J., Russell, G. W. & Pease, D. G. (2001). Sport Fans: The psychology and social impact of spectators. New York: Routledge.

WHO, (2010). WHO Global Infobase. Available from: https://apps.who.int/infobase/Comparisons.aspx.

Keith D. Parry

University of Western Sydney

Ian Jones

Bournemouth University

Daniel L. Wann

Murray State University

Address correspondence to: Keith Parry, School of Business, University of Western Sydney, Locked Bag 1797, Penrith South DC, NSW 1797. Email: k.parry@uws.edu.au

Table 1
Regression Statistics for the Analyses Incorporating Sport
Fandom Behaviors as the Dependent Variables and Team
Identification (SSIS Scores) and Sport Fandom (SFQ Scores)
as the Predictor Variables.

                                 Team
                            Identification    Sport Fandom

                           Beta      t      Beta      t

Attendance at events       .19    2.4 *     .07    0.8
Sport television viewing   .31    4.5 ***   .33    4.7 ***
Sport radio listening      .27    3.5 ***   .18    2.2 *
Sport on the Internet      .24    3.6 ***   .41    5.8 ***
Sport discussions          .25    3.6 ***   .43    6.3 ***

Note: * p < .05, ** p < .01., *** p < .001.

Table 2
Means and Standard Deviations for the Impact of the
Four Socialization Agents.

Socialization                                       Sample
Agent                   British        Greek      Australian

Parents
  All participants    5.67 (2.03)   4.36 (2.93)   5.72 (2.23)
  Males               5.72 (2.00)   4.85 (2.94)   5.76 (2.18)
  Females             5.54 (2.14)   3.16 (2.57)   5.66 (2.34)
Friends
  All participants    5.67 (1.76)   4.47 (2.71)   6.01 (1.80)
  Males               5.91 (1.69)   4.74 (2.74)   6.24 (1.58)
  Females             5.01 (1.80)   3.81 (2.52)   5.65 (2.07)
Schools
  All participants    5.33 (1.77)   4.26 (2.47)   5.52 (1.86)
  Males               5.36 (1.72)   4.56 (2.46)   5.45 (1.79)
  Females             5.27 (1.92)   3.55 (2.36)   5.63 (1.99)
Community
  All participants    3.97 (1.89)   4.28 (2.57)   5.42 (2.00)
  Males               4.09 (1.86)   4.53 (2.44)   5.65 (1.84)
  Females             3.66 (1.94)   3.65 (2.78)   5.03 (2.19)

Socialization                         United
Agent                  Norwegian     States *

Parents
  All participants    4.67 (2.28)      4.69
  Males               4.43 (2.30)      4.88
  Females             4.84 (2.26)      4.52
Friends
  All participants    4.97 (1.93)      5.47
  Males               5.47 (1.89)      6.02
  Females             4.62 (1.89)      4.99
Schools
  All participants    3.82 (2.12)      5.14
  Males               3.59 (1.98)      5.25
  Females             3.99 (2.21)      5.04
Community
  All participants    5.33 (2.01)      4.08
  Males               5.45 (1.94)      4.30
  Females             5.25 (2.15)      3.88

Notes: Response options ranged from 1 (the agent had no
influence) to 8 (the agent had a great deal of influence).
Standard deviations appear in parentheses next to each mean.
* Wann et al. (2001) did not report standard deviations for
the US sample.

Table 3
Percentages of Sport Socialization Agents Listed as Most Influential.

                British         Greek

Agent           All   M    F    All   M    F

Father          49    55   33   39    49   34
Mother           4    2    10    0    0    0
Both parents     2    2    5     0    0    0
Brother          6    5    8     6    5    6
Sister           2    1    5     0    0    0
Siblings         0    0    0     0    0    0
Grandfather     <1    1    0     3    4    0
Grandmother      0    0    0     0    0    0
Husband          0    0    0     0    0    0
Uncle            2    3    0     5    6    4
Cousin          <1    1    0     2    2    2

Family total    66    70   61   55    66   46
Friend           8    10   3    23    22   37
Boyfriend       <1    0    2     0    0    0

Friend total     9    10   5    23    22   37
School           7    4    15    0    0    0
Teacher          2    2    2     0    0    0

School total    11    6    17    0    0    0
Community       <1    1    0     0    0    0
Mass media       2    3    2     1    <1   1
Coach            2    1    3     2    1    4
Sport team       1    1    0     1    2    0
Famous player    6    8    4     0    0    0
Myself           0    0    0     4    3    9

                Sample          Norwegian
                Australian

Agent           All   M    F    All   M    F

Father          45    53   34   30    28   33
Mother           4    1    10    3    0    5
Both parents     0    0    0     2    0    3
Brother          8    6    11    5    10   0
Sister           1    0    2     3    0    5
Siblings         0    0    0     1    0    2
Grandfather      3    2    5     0    0    0
Grandmother      1    1    0     0    0    0
Husband          0    0    0     1    0    2
Uncle            2    3    0     0    0    0
Cousin           1    0    2     1    2    0

Family total    65    66   64   46    50   50
Friend           7    6    8    13    14   12
Boyfriend        1    0    2     2    0    3

Friend total     8    6    10   15    14   15
School           7    5    10    9    6    10
Teacher          1    0    3     0    0    0

School total     8    5    13    9    6    10
Community        1    2    0    22    26   19
Mass media       3    3    2     7    12   3
Coach            1    1    0     1    2    0
Sport team       0    0    0     0    0    0
Famous player   10    14   5     0    0    0
Myself           1    0    2     0    0    0

                United States

Agent           All   M    F

Father          35    39   31
Mother           5    3    7
Both parents     1    0    4
Brother          8    11   5
Sister           2    1    2
Siblings        <1    0    <1
Grandfather      1    2    1
Grandmother     <1    <1   <1
Husband          1    0    2
Uncle            2    2    2
Cousin           1    <1   1

Family total    57    59   56
Friend           9    10   7
Boyfriend        4    0    7

Friend total    13    10   14
School          12    8    15
Teacher         <1    0    <1

School total    12    8    16
Community        2    2    2
Mass media       5    7    4
Coach            2    3    1
Sport team       0    0    0
Famous player    1    1    <1
Myself           1    1    1

Notes: Numbers are percentages rounded to the nearest whole number.
<1 = .1% to .4%. Columns do not total 100% due to responses
classified as "other" and persons failing to provide a response.

Table 4
Means and Standard Deviations for Sport Fandom (SFQ Scores)
and Team Identification (SSIS Scores).

                                          Sample
                         British          Greek
Sport Fandom
  All participants    32.53 (6.89)     27.66 (3.96)
  Males               34.74 (4.80)     30.29 (6.77)
  Females             26.40 (8.04)     21.75 (9.41)

Team Identification
  All participants    41.31 (10.50)    36.91 (11.82)
  Males               44.01 (8.91)     40.203 (10.18)
  Females             33.90 (11.00)    29.91 (12.21)

                       Australian       Norwegian
Sport Fandom
  All participants    32.35 (7.82)     26.32 (N/A)
  Males               33.58 (6.83)     28.50 (7.40)
  Females             30.34 (8.90)     24.80 (8.90)

Team Identification
  All participants    40.31 (8.52)     30.50 (N/A)
  Males               41.19 (8.05)     33.80 (9.80)
  Females             38.74 (9.17)     27.20 (10.20)

Notes: Wann et al. (2001) did not report fandom and
identification scores for their United States
sample. N/A = data not available.

Table 5
Rank Order of the Impact of the Four Socialization Agents.

Socialization              Sample
Agent
                    British           Greek

                All    M     F    All    M     F

Parents          1     2     1     2     1     4
Friends          2     1     3     1     2     1
Schools          3     3     2     4     3     3
Community        4     4     4     3     4     2

Socialization              Sample
Agent
                  Australian        Norwegian

                All    M     F    All    M     F

Parents          2     2     1     3     3     2
Friends          1     1     2     2     1     3
Schools          3     3     3     4     4     4
Community        4     4     4     1     2     1

Socialization      Sample
Agent
                United States

                All    M     F

Parents          3     3     3
Friends          1     1     2
Schools          2     2     1
Community        4     4     4