An Examination of NFL Fans' Computer Mediated BIRGing.
End, Christian M.
This investigation examined how sport fans use the World Wide Web (WWW) to publicly display their associations with sport teams and manage these identities during their teams' victories and defeats. Two types of WWW usage were examined: WWW home pages and message boards. Fans did use their home pages to publicly associate with sport teams via linking, but the number of home pages that linked to the pages of successful teams did not differ significantly from the number of home pages linked to unsuccessful teams' pages. Analysis of the messages posted on the Sporting News' National Football League message boards during the last five weeks of the 1997-1998 regular season and playoffs suggest that fans use the WWW to bask in the reflected glory of their team as well as to "blast" their opponents and fans of their opponents. The implications of these findings are discussed in regards to sport fans' management of their social identities.
Fans often go to great lengths to publicly display their associations with sports teams. Fans spend large amounts of money on their favorite teams' or players' jerseys, and other team merchandise that literally enables them to show "their true colors". Some fans decorate their cars with their team's bumper stickers, window decals, and flags to proclaim their sport team affiliations even while traveling.
Despite the history of sport fanship in America, there has been relatively little research dedicated to understanding sport fan behavior (Dietz-Uhler, Harrick, End, & Jacquemotte, 2000; Wann & Hamlet, 1995). Although there exists a rich array of sport psychology and sport sociology research, Wann and Hamlet (1995) found that only 4% of this research examined sport fans and their behavior. Much of this research has examined the costs and benefits of identifying with sport teams. This research suggests that some of the benefits that sport fans enjoy include experiencing a sense of belonging or identity (Sloan, 1989), excitement and/or recreation (Sloan, 1989), and psychological well-being (Wann, lnman, Ensor, Gates, & Caldwell, 1999). Researchers have also identified costs associated with being a sport fan. These include lower estimates of personal performance following a loss (Hirt, Zillman, Erickson, & Kennedy, 1992) and increases in spectator hostility (Russell, 1981). Of particular interest to researchers has been the effects of team identification on college-aged fans' self-esteem and how fans use these associations as a way to manage their self-presentation (Cialdini et al., 1976; Cialdini & Richardson, 1980; End, Dietz-Uhler, Harrick, & Jacquemotte, 2001).
Previous research has demonstrated that college students may use associations with sport teams to manage their self-presentations (Cialdini et al., 1976; End, et al., 2001). One purpose of this study is to try to replicate these findings in a sample that includes college students as well as fans outside the college population. Specifically, I want to examine if and how sport fans use the World Wide Web (WWW) to publicly display their associations with sports teams and manage these identities during their teams' victories and defeats.
Self-Presentation Research
Although previous research has not examined how sport fans use the WWW to manage their association with sport teams, it has examined how college students manage their team identifications in other settings. For example, Cialdini et al. (1976) examined how the outcome of a sporting event influenced college football fans' self-presentational behavior. They found that college students were more likely to wear school-identifying apparel following their football teams' wins in comparison to losses. Cialdini et al. (1976) also found that students were more likely to use the pronoun "we" when describing their teams' victories than their teams' losses. The authors concluded that this behavior may reflect sport fans' motivations to publicly associate with successful others. This tendency to bask in reflected glory (BIRG) may be the result of fans perceiving that others will view them as being successful as a consequence of their association with a successful team.
Additional research has demonstrated a strong, positive relationship between the success of sport teams and fans' identification with teams. Wann, Tucker, and Schrader (1996) found that the success of the team was reported to contribute to individuals' past, present, and initial identification with teams. End et al.'s (2001) findings that college students were more likely to report identifying with successful teams than unsuccessful teams support the findings of Wann et al. (1996). End et al. (2001) found that 71% of the teams that participants identified with were categorized as being successful. These results suggest that college-age fans may be aware of the benefits of associating with a successful team.
Social Identity Theory
End et al. (2001) used social identity theory to explain how this tendency to identify with successful teams can be used by fans to manage their self-presentation. Social identity theory states that individuals are driven to maintain positive self-esteem. According to Tajfel and Turner (1986), one way that individuals can achieve positive self-esteem is through their associations with successful groups. These social identities are deemed positive and are valued when a comparison to a relevant out-group is favorable.
For sport fans, the most relevant dimension for social comparison is their team's performance relative to their opposing team's performance. Social identity theory would predict that a fan of a victorious team feels superior to the fan of the defeated team. Thus, a fan a victorious team would achieve a positive social identity relative to the fan of the losing team.
The potential for a less-than-positive social identity is one risk of being a sport fan. Although research has suggested that fans may identify with teams that they feel have a high potential for being extremely successful (End et al., 2001), very rarely is a team able to progress through a season without losing. Consistent with social identity theory, in situations when one's team loses, an individual who is a member of that group may feel and be perceived to be inferior to a relevant out-group (the rival fan) as a result of social comparison. The fan of the losing team may personally feel and be perceived as being inferior, resulting in a threatened self-esteem. According to Tajfel and Turner (1986), in these instances of negative social identity and threatened self-esteem, individuals will strive to improve their social identity. Briefly, Tajfel and Turner (1986) suggest that individuals can restore their positive social identities and self-esteem in one of three ways: They can leave the group (social mob ility). They can compare their group to the out-group on another dimension (social creativity). Or, they can directly compete with the relevant out-group in an attempt to reverse the relevant positions (social competition).
Impression Management in the Face of Defeat
Although sport fans are incapable of scheduling direct competition between their team and a superior opponent (social competition), fans may be able to utilize other techniques to manage and improve their threatened self-esteem. Research has demonstrated that fans often attempt to distance themselves from an unsuccessful sport team by reducing their association with the team (Cialdini et al., 1976; Wann & Branscombe, 1990). This occurs when fans cease or reduce the frequency in displaying public associations (e.g., wearing team apparel, publicly displaying team merchandise, referring to the team as "we", etc.) with their team after an unsuccessful performance or a loss. This reduction in one's association with or distancing from an unsuccessful team is similar to the concept of social mobility (Tajfel & Turner, 1986). Fans may hope that by reducing or ceasing their association with an unsuccessful team, they will be able to cut off the reflected failure (CORF) associated with their team's unsuccessful perfor mance (Snyder, Higgens, & Stucky, 1983).
Although research has indicated that fans do indeed CORF to maintain a positive social identity (Cialdini et al., 1976), Wann and Branscombe (1990) found that individuals who identified highly with a sport team were less likely to CORF than individuals who reported lower levels of identification. The authors concluded that individuals who exhibited high levels of identification with sport teams may be forced to use alternative techniques to maintain their positive social identity (Hirt, et al., 1992). Research has indicated that high-identifying fans might utilize a strategy (blasting) to help maintain a positive social identity while reducing the threat to their self-esteem on occasions when their team loses (Cialdini & Richardson, 1980). Blasting occurs when a fan attempts to make another person perceive her/his team in a positive manner by derogating the opposition. Fans who use this technique hope that they can convince an observer that even though their team lost, they should be perceived in a positive light as a result of their team being presented more positively than their "blasted" opponent.
In addition to blasting, high-identifying fans may utilize other techniques as a means of maintaining a positive social identity when threatened by their team's loss. Some additional techniques identified by research include creating a new standard for comparison between teams (social creativity; Lalonde, 1992), attempting to decrease identification prior to an athletic competition in order to cut off the potential of future failure (COFFing; Wann, Hamlet, Wilson, & Hodges, 1993), and attributing defeat to uncontrollable external factors (Mann, 1974; Wann & Dolan, 1994).
Self-Presentation on the WWW
The purpose of this study is to examine whether sport fans use tactics such as BIRGing, CORFing, and blasting to manage their self-presentation through their WWW identities. Although most research on BIRGing and CORFing behaviors of sport fans has been limited to a college setting, Joinson and Harris (1995) examined the behavior of soccer fans on the WWW. Specifically, Joinson and Harris were interested in determining whether or not fans' BIRGing and CORFing behaviors would be evident in individuals' WWW usage. Whereas in most studies BIRGing has been operationally defined as wearing team related apparel or referring to the team as "we" and CORFing has been operationally defined as not wearing team apparel or referring to the team as "they," Joinson and Harris defined BIRGing and CORFing in terms of information seeking. For their purposes, BIRGing was defined as increases in the number of accesses of team's WWW pages following a victory, and CORFing as decreases in the number of accesses of teams' WWW pages following a defeat.
Joinson and Harris' (1995) examination of soccer fans' WWW usage prior to and following 77 soccer matches played by three teams found partial support for parallels between soccer fans' WWW behavior and the impression management found among college students. Specifically, they found that soccer fans were more likely to access teams' WWW pages following a victory than prior to a game (BIRGing), but did not find evidence that WWW-using fans CORF. The number of WWW page accesses prior to a game did not differ from the number of accesses following a loss. Joinson and Harris (1995) concluded that these findings might have been the result of how they operationalized BIRGing and CORFing behaviors. Accessing a team's web page is not a social activity. Unless others are physically present when a fan accesses a team's web page, the only person capable of recognizing the association between a fan and the team is the individual herself/himself. Joinson and Harris concluded that frequency of home page access could at best be conceptualized as a measure of internal BIRGing and CORFing, and not as a form of self-presentation. They suggested that this lack of social interaction or deindividuation may account for their CORFing results. This deindividualized medium allows a fan to access their losing team's web page without the fear of being evaluated negatively by an outside observer.
Although accessing a team's web page is void of social interaction, there are other Internet and Web activities that possess social aspects. As Kiesler (1997) writes, "What makes the Internet special is the social interaction it is inspiring." (p.x). One impression management activity of interest is the indirect interaction between web page designers and the individuals who access their pages. Turkle (1995) suggests that individuals construct home pages that reflect personal "home" identities. Web page designers construct pages by composing content, including features such as graphics and images, as well as determining which other web pages or Internet sites they should link their page to. These features and associations to other sites on the Internet and Web combine to create one's "Web page" identity. According to Walther (1996), previously unacquainted users develop impressions based on the content of the material presented to them. Therefore, individuals who access WWW pages form impressions about the cr eator based on the WWW pages' content, associations, and design (Turkle, 1995). Links as well as other components of Web pages can be used to manage one's self-presentation by stressing one's associations with various groups. According to Tajfel and Turner (1986), individuals should be motivated to present themselves in a positive light by stressing their membership with groups that are perceived in a favorable manner. Unlike Joinson and Harris who examined internal BIRGing, this study intends to examine the social aspects of WWW usage, specifically if sport fans use their web pages for impression management by identifying with successful sports teams. I hypothesize that more people will link their web pages to the official NFL home pages of teams categorized as being successful in comparison to unsuccessful teams.
A second social interaction of interest is the use of bulletin boards or message boards on the Internet. Rickard (1993) reports that there are more than 45,000 public electronic bulletin boards in the United States alone that discuss topics that range from self-help to sports. Bulletin boards simply act as a forum for asynchronous communication, delayed in real time, between individuals under the heading of a specific topic (Gackenbach & Ellerman, 1998). In addition to information sharing, these message boards can be used to communicate one's membership in social groups or other on-line communities to others visiting the message board. As mentioned above, information posted on message boards can be used by other users to form an impression of the individual who posted the message (Walther, 1996). In accordance with Tajfel and Turner's (1986) social identity theory that states individuals should be motivated to stress their membership with groups that are perceived in a favorable manner, it is hypothesized th at NFL fans will use their teams' message boards to stress their associations with successful teams (BIRG). Specifically, NFL fans will post more messages on the message boards of successful teams than unsuccessful teams. Fans' BIRGing should be evident in terms of increases in the number of postings as a function of a team's overall seasonal success (End et al., 2001) as well as event-specific (game) success (Cialdini, et al., 1976).
In addition to stressing one's association with a successful team, message boards may afford sport fans with additional options for enhancing their self-esteem that may not be available in face-to-face interactions. Joinson (1998) suggests that individuals who use asynchronous communication such as message boards may be vulnerable to disinhibition, which can be defined as the absence or reversal of factors such as self-consciousness, anxiety of social situations, worries about public evaluation and other factors that constrain or restrain behavior. Disinhibition, which may be the result of the perception of relative anonymity while using the Internet as well as the physical safety of electronic communication, can encourage aggressive and even abusive behavior called flaming (Reid, 1998). Flaming, which is usually defined as being communication that is hostile (Joinson, 1998), can be conceptualized as being similar to blasting behavior exhibited by sport fans (Cialdini & Richardson, 1980). As mentioned earlie r, fans whose social identity is threatened by a failure can attempt to reduce the threat by derogating their opposition. Additionally, fans' BIRGing of teams that are successful can attempt to heighten their positive social identities by blasting others who associate with unsuccessful teams. Therefore, flaming or blasting others via postings on message boards may be a way for fans to enhance their self-esteem without the physical risk or accountability associated with the same behavior when it occurs face-to-face. It is hypothesized that sport fans will use teams' message boards to post a greater percentage of messages that blast a team following a failure than following a success.
To summarize, my hypotheses include the following:
1. NFL fans will have more links from their home pages to the official home pages of successful teams than unsuccessful teams.
2. NFL fans will post more messages on the message boards of teams that have experienced overall success (playoff bound) and event-specific success (victory) during the regular season.
3. NFL fans will post more messages on the message boards of teams that are successful (victory) during the playoffs than teams that are unsuccessful.
4. A greater proportion of messages posted on a team's message board will indicate identification with a team following a victory than a loss.
5. Fans will BIRG within the content of messages more often after successes (victories) in comparison to instances when they are unsuccessful.
6. Teams' message boards will be blasted more following teams' failures (loss) than following teams' successes.
Method
Overview
In order to examine the various aspects of sport fans' WWW usage, data were collected on several occasions from two sources. Alta Vista was used to determine the number of web pages that provide links to each of the NFL teams' official Web pages at a single point in time. A second source of data was the message boards maintained by the Sporting News. This web site provided three distinct samples of data. Data were collected from 10 of the 30 NFL teams' message boards during the last four weeks of the regular season to examine the influence of seasonal success and weekly success on sport fans' message posting during the regular season. A sample of this data was coded to examine the impact of game outcome on the content of the messages posted. Data from all 30 NFL teams' message boards was collected on a single occasion to provide an additional examination of sport fans' message board use during the regular season. Finally, data were collected from the message boards of each team that qualified for that season 's playoffs to examine sport fans' use of message boards during the playoffs, specifically their tendency to BIRG.
BIRGing has been defined in a variety of ways. Typically, BIRGing has been defined as a tendency by individuals to publicly associated with teams that have been victorious (Cialdini et al., 1976; Cialdini & Richardson, 1980; End, et al., 2001). Cialdini et al. (1976) operationalized BIRGing as increases in students' apparel wearing and use of the pronoun "we" while Joinson and Harris (1995) conceptualized BIRGing as increases in the number of hits on a team's web page. In the context of social identity theory, BIRGing can be viewed as an attempt by individuals to publicly stress their membership with groups that are perceived in a favorable manner. The behaviors that will be considered to be representative of BIRGing, CORFing, and blasting on the WWW include uses that enable a sport fan to publicly associate with a sport team. These behaviors include: web page identification, message board identification, and differences in the frequency of message board posts.
Web Page Identification
One way sport fans can publicly stress their association with a sport team (BIRGing) is by linking their home page to a sport team's official web page. In order to examine if NFL fans are more likely to associate with successful teams than unsuccessful teams via Web pages (seasonal BIRGing), it was necessary to determine the number of Web pages that contained links to each NFL teams' official Web pages. The number of pages that contained NFL team links was found using the search engine AltaVista. By entering each teams' official URL (each team has its own official web page supported by the NFL) into Alta Vista's link:URLtext command (which provides a list of all links to the specific URL), I found the total number of home pages that had links to the NFL teams' official Web pages. For example, in determining the number of Web pages that had links to the Green Bay Packers' official Web site, I simply entered 'link:www.nfl.com/packers' into AltaVista's search field.
Although there are numerous criteria that can be used to determine if a team is successful (End et al., 2001), a team was coded as being successful if they qualified for that season's playoff.
Message Board Identification
In terms of message board use, fans can publicly associated with successful teams by posting a message on a successful team's message board or by identifying with a team in the content of a message that one posts. In order to examine if fans use message boards as a means to associate more with successful NFL teams than unsuccessful teams (seasonal BIRGing), the number of postings made on each team's message boards were examined in relation to the team's seasonal success. I monitored the 30 NFL message boards maintained by the Sporting News (http://tsn.sportingnews.com/messageboards/). The Sporting News site was used because it is one of the better known sport sites (as a result of its association with the magazine) and because the web site maintains one message board for each of the 30 NFL teams. Data collected on December 31 produced a list of all the postings made on each teams' message boards during the period of December 21 to December 31. Data were collected during this time because this was when it bec ame clear which teams had qualified for the playoffs, allowing for teams to be coded as being successful and unsuccessful. The number of posts made on each team's message board was determined and teams were coded as being successful if they qualified for that season's playoff.
Message Board Regular Season Event-specific BIRGing
Data were collected during the regular season for 5 weeks, from December 7 to January 3, from 10 of the 30 NFL teams' message boards maintained by the Sporting News. In order to control for seasonal BIRGing, the message boards of the 5 teams that had the worst record in the NFL were monitored as were the message boards of the 5 teams whose regular season performance indicated that they would make the playoffs. In addition to copying the content of each message posted during this time span, the date the message was posted and the identity provided by the individual who posted the message was recorded. The specific time of day that the posts were made was not available. Therefore, for messages posted on the day of the game, two coders blind to the hypotheses categorized messages as being written prior to the game, written after the game, or as ambiguous. If the messages were written in future tense (e.g., We will beat you today!), they were coded as being posted prior to the game. If messages were written in p ast tense (e.g., We beat you today!), they were coded as being posted after the game. Finally if the message provided no cues to when it was written, the message was coded as being ambiguous and excluded from the analysis.
Each team's weekly performance (win or loss) was recorded during this time period with wins representing successes and losses representing failures.
In order to determine whether event-specific BIRGing occurred, I recorded the number of messages posted the week prior to each of the 10 teams' games for the last four weeks of the NFL season as well as the outcome of the game. BiRGing was defined to have occurred if the number of posts following victories were greater than the number of posts following a loss.
Content Analysis of Regular Season Message Boards' Postings
Each week, 100 messages posted on each of the 10 NFL teams' message boards were coded as either being representative of BIRGing, blasting, or information sharing. These 100 messages were randomly selected from each teams' message board. If a message board had less than 100 messages posted during the week, all messages were coded. The final sample consisted of a total of 2,928 messages. Messages were coded as BIRGing if they contained any mention of association between the individual who posted the message and the team whose message board the message was posted (e.g., PackerSteveO, we, etc.) and the content stressed a positive team accomplishment, (e.g., Great win today!). Blasting messages were coded into two different types based on the affiliation of the individual who posted the blasting comments. A message was coded as being a "Blasting Opponent" message if the individual who posted the message was associated with the team whose message board the message was posted and claimed that their team was superio r to oilier teams on any dimensions (e.g., "Our pass rush is so much better than your team's line." Posted by PackerSteveO on the Green Bay Packers' message board). If these individuals insulted a team other than the one associated with the message board, those messages were also coded as Blasting Opponent. Messages were coded as being a "Blasting by Opponent" message if an individual posted a message with content similar to the Blasting Opponent message and the individual affiliated with a team other than the one associated with the message board. Any other messages posted by individuals who did not mention an affiliation with the team where they posted their message and who did not make any comparison between teams were coded as being information sharers (e.g., "I thought both teams left it on the field today. I hope next week's game is just exciting". Posted by Jeff Kretschmar on the Oakland Raiders' message board). In addition to team affiliation and content of the message, the 25 messages that appeared d irectly above and below the post were examined to determine whether or not the individual had posted a message in the past (49% of the messages posted were made by individual who had posted messages in the past). The researcher coded the sample of 2,928 messages. In order to test for reliability, a research assistant coded three teams' data (625 messages). The coding of content, whether or not an individual had posted a message in the past, and whether or not the message indicated identification was found to be reliable (Cohen's Kappa .89, .89 and .87 respectively). In addition, each team's weekly performance (win or loss) was recorded.
Message Board Post-Season Event-specific BIRGing
Data were collected during the NFL playoffs from the 12 NFL teams' message boards that made the playoffs. The number of messages posted on each teams' Sporting News message boards was recorded the week following the II playoff games. Each team's weekly performance (win or loss) during this time period was recorded. Similar to the examination of regular season event-specific BIRGing, BIRGing was defined to have occurred if the number of posts following victories were greater than the number of posts following a loss.
Results
Web Page Identification
In order to test the hypothesis that NFL fans are more likely to stress their association with successful teams than unsuccessful teams via home page links (Hypothesis I), I compared the average number of links to the 12 successful teams' official home pages to the average number of links to the 18 unsuccessful teams official home pages. There was no difference between the number of home pages linked to the official home pages of the successful teams (M= 858.23, SD = 275.23) and unsuccessful teams (M 878.11, SD = 215.19), t (28) = .22, p[greater than].83, ES = .08. This measure was obtained at a single point in time.
Message Board Identification
In order to test the hypothesis that NFL fans will post more messages on successful teams' message boards than unsuccessful teams' message boards (Hypothesis 2), a t-test was used to compare the average number of postings on the 12 successful teams' message boards and the average number of postings on the 18 unsuccessful teams' message boards during the period of December 21st to December 31st. People posted significantly more messages on successful teams' message boards (M= 572.75, SD = 346.33) than unsuccessful teams' message boards (M= 298.11, SD = 191.85), t (28) = 2.80 p [less than].01, ES= 1.08. The comparison of the effect sizes suggests that the difference between the number of messages posted on a successful teams' message boards and unsuccessful teams' message boards is a large difference (Cohen cited in Thomas, Salazar, & Landers, 1991).
A second way to examine if fans post more messages on successful teams' message boards than unsuccessful teams' message boards (Hypothesis 2) is to examine the average number of messages posted weekly on the five successful teams' boards and the five unsuccessful teams' boards during the last four weeks of the season. The results show that there were more messages posted weekly on successful teams' message boards (80.1% of the total messages) than unsuccessful teams' message boards (19.9% of the total messages), [X.sup.2] (1) = 2639.72p[less than].001
Finally, chi-square tests were used to determine if a greater proportion of messages would indicate identification with a successful team than unsuccessful team (Hypothesis 4). The results showed that proportion of messages that indicated identification differed as a result of the teams' levels of success, [X.sup.2] (I) = 4.45, p[less than].05. Inspection of Table I reveals that more than half of the posts indicated identification with the unsuccessful teams (50.9%), while slightly less than half of the messages indicated identification with successful teams (46.9%).
Message Board Regular Season Event-specific BlRGing
In order to examine if event-specific BIRGing occurred during the regular season (Hypothesis 2), a t- test was performed to compare the mean number of weekly messages posted for teams that had been victorious the prior week to the mean number of weekly messages for teams who had been defeated. The comparison was made using the number of messages and the success or failure of 10 of the 30 NFL teams over the last four weeks of the NFL season. People posted more messages on the message boards of teams who had been victorious the week before (M = 218.00, SD = 201.86) than teams that had been defeated the preceding week 41= 104.79, SD = 108.95), t(38) = 4.13, p [less than].05, ES = .72. The comparison of the effect sizes suggests that the difference between the number of messages posted on a victorious teams' message boards and the message boards of teams that had been defeated is a moderate difference (Cohen cited in Thomas, et al., 1991).
A second way to examine if event-specific BIRGing occurred during the regular season was to determine if a greater proportion of messages indicated identification with a team following a victory than following a loss (Hypothesis 4). A chi-square analysis showed that the proportion of messages that indicated identification did not differ as a result of the teams' weekly performance, [X.sup.2] (I) = .36, p[greater than].85. Table 2 reveals that slightly more than half of the messages lack identification indicators following victories (51.7%) than defeats (51.4%).
Content Analysis of Regular Season Message Boards' Postings
Testing the hypothesis that there would be a greater percentage of blasting messages following teams' failures than teams' successes (Hypothesis 6) and a greater percentage of BIRGing messages following a teams' successes in comparison to teams' failures (Hypothesis 5), a chi-square analysis was used to examine if the proportions of the types of messages posted on 10 teams' message boards over the last four weeks of the regular season differed as a result of outcome. Message content did differ as a function of outcome, [X.sup.2](3) = 36.57, p[less than].001. Message board users were five times as likely to post a BIRGing message on a victorious teams' message board (84.0%) than a losing teams' message board (16.0%). Users were also twice as likely to post messages that blasted their opponent following a win (68.6%) in comparison to a loss (31.4%). Users were nearly twice as likely to be blasted by an opponent and share information following a victory (61.0% and 65% respectively) than following a loss (39% an d 35% respectively). Table 3 summarizes the percentage of the total messages by content as a function of outcome.
Message Board Post-Season Event-Specific BIRGing
In order to examine if event-specific BIRGing occurred during the post-season (Hypothesis 3), a t-test was performed to compare the mean number of weekly messages posted for teams' who had won their previous playoff game and those teams that had lost their previous playoff game. Evidence of event-specific BIRGing would be indicated if more messages were posted on victorious teams' message boards than losing teams' message boards. Although there was a tendency for more messages to be posted on the message boards of teams that won their playoff game (M= 588.82, SD = 318.22) than teams that did not win their playoff game (M= 435.45, SD = 304.93), the difference was not significant, t (20) = 1.15, p = .26, ES = .49
Discussion
The purpose of this investigation was to examine if fans, including those outside the college population, use the WWW to manage their sport fan identities. Although sport fans who have internet access and who participate in message board are probably not representative of sport fans in general, the results from this selective sample suggest that sport fans' use of self-presentational behaviors vary depending on the specific WWW activities.
In terms of WWW home pages, I found that sport fans were no more likely to provide links from their home page to the official NFL pages of successful teams than the official NFL pages of unsuccessful teams. This finding is contrary to the prediction that NFL fans would use these links to publicly identify with successful others to stress their positive sport fan identities. One possible explanation for these findings is Alta Vista's link:URLtext command does not distinguish between fan and commercial home pages. Therefore, the proportion of commercial home pages (e.g., team related products, area newspapers, etc,) to personal home pages that provide links may vary from team to team and this disproportion may account for the lack of difference. A second explanation for this finding is that Alta Vista's link:URLtext command reports every WWW page that contains the specific link, but does not eliminate duplicate results. For example, if a home page creator supplies the link to the Green Bay Packers' official pa ge from her/his WWW page 15 times, Alta Vista would report each instance individually. In this study, one page that contains 15 identical links would be counted as being 15 separate WWW pages, skewing the data. These explanations could be examined in the future by specifically examining each WWW page that contains an official NFL link and eliminating commercial pages and duplicate results from the data set.
Examination of message board identification suggests mixed results in terms of fans' self-presentational behaviors. Fans were more likely to post messages on the message boards of successful teams (playoff bound) than unsuccessful teams (not playoff bound) based on the ten-day sample of all 30 NFL teams, and four-week sample of five successful and five unsuccessful teams. These results are consistent with the finding that college-aged fans are more likely to identify with successful teams than unsuccessful teams (End et al., 2001), suggesting an attempt to publicly BIRG with an overall successful team.
On the other hand, fans' BIRGing was not found in terms of a difference between successful and unsuccessful teams in the proportion of total posts that indicated identification to the proportion of posts that lack identification indicators or indicated identification with an opposing team. Contrary to the predictions, a greater proportion of the total messages indicated identification with unsuccessful teams than successful teams. The predicted increase in the proportion of individuals who post a message indicating identification in hopes of BIRGing may be negated by fans of opposing teams posting messages attempting to block one's reflected glory. Thus, the increases in both identifying and non-identifying messages negate each other.
The tendency for fans to use successful teams' message boards more than unsuccessful teams' message boards was also evident in the data for game-specific success, suggesting that fans are more likely to publicly stress their identification with teams following victories, BIRGing (Cialdini et al., 1976). More messages were posted on the message boards of teams that had won compared to teams that had lost. As with overall team success, increases in the number of messages posted by both fans who identify with the team and those fans who do not result in no difference in the proportion of identifying messages to total messages for teams that won in comparison to team that lost.
The content analysis of regular season messages indicates that fans may use message boards to post BIRGing, Blasting, and Information Sharing messages. Of particular interest is the finding that almost one-third of the total messages (32.7%) arc Blasting messages. These results suggest that this deindividuated medium may provide an outlet for fans to blast their opponents in order to protect or enhance their self-esteem. Fans would probably not use this tactic to this extent in a face-to-face interaction with other fans due to the potential of physical risk or psychological harm. In a computer medium where physical distance is unknown and the physical risk is alleviated (Reid, 1998), fans blast each other.
Looking at the impact of outcome on message content, I found there is an increase in the proportion of BIRGing messages following wins, supporting the hypothesis. The increase in the percentage of BIRGing messages following victories compared to losses can be explained as victories providing fans an accomplishment with which to BIRG. These victories also afford fans with the opportunity to make a favorable comparison to an opponent. As a result a greater proportion of Blasting Opponent messages were made following wins than following losses.
The playoff data suggest that during the regular season, there is tendency (although not significant) for greater use of successful teams' message boards in comparison to unsuccessful teams' message boards as a function of game outcome, suggesting that fans are more likely to publicly stress their identification with teams following victories; BIRGing. More messages were posted on the message boards of the playoff teams that had won in comparison to the teams that had lost. The lack of statistical significance may be a result of the size of the data set. With only 11 games in the NFL playoffs, the comparison was made between the 11 teams that won the playoff games and the 11 teams that lost. A second explanation for this lack of statistical significance that can help explain the lack of statistical significance of earlier tests as well, is the large variance found within my samples. Because this is archival research utilizing "real world" data, there is large variability within the samples, which is evident by the large standard deviations reported. As a result, many of the mean differences that appear to be quite large, like the difference found using the playoff data, are not statistically different.
Conclusion
As evident from the data collected, sport fans use the World Wide Web to publicly associate with NFL teams. Whether or not this public social identity is stressed by fans is influenced by a team's success. Consistent with other forms of public affiliation (apparel wearing, use of identifying pronouns, Cialdini et al., 1976), I found there is greater public affiliation with teams that demonstrate event-specific success (weekly win) as well as overall success (playoff bound). For individuals whose sport fan identity is an important aspect of their social identity, this use of the WWW may be a technique for individuals to manage their sport fan identities. Following victories or during seasons when a team is successful, fans may be using the WWW to bask in the reflected glory of a successful team and improve their feelings of self-worth, while during times of mediocrity or following losses, fans may avoid using these messages boards in attempt to protect their self-worth (CORF, Snyder, Higgens, & Stucky, 1983).
The data suggest that fans may be using blasting tactics to enhance or protect their sport fan identities at the expense of other fans. Fans of successful teams may be using the WWW to blast the fans of their unsuccessful opponent presumably, to enhance the benefits afforded to their team following victory (Cialdini & Richardson, 1980). While fans of the losing teams, whose sport fan identities and self-esteem may be threatened and who are unable to bask in the glory of their unsuccessful team, blast fans of other teams as a way to protect their threatened self-esteem at the cost of others (e.g., We may have lost, but at least we are not as bad as you are.). Fans' use of blasting seems to be enhanced by the deindividuated state afforded them in the WWW context. In face-to-face interactions, blasting of another individual or an individual's team carries the risk of physical violence or psychological harm. This risk may inhibit a fan from blasting another as a means of protecting or enhancing their sport fan i dentity and force them to cope with the threatened identity in another manner, e.g., social mobility, or have a lowered self-worth until their team's next victory. On the other hand, fans using message board do not fear this physical retaliation and as result, blasting may be seen as plausible option for protecting or enhancing one's sport fan identity. For fans of an unsuccessful team, protection from a threatened identity may occur immediately after a loss as a result of posting a message blasting their opponent. Fans of an unsuccessful team who lack access to the WWW may go weeks before their team's next win and as a result have to come up with other alternatives to deal with their threatened sport fan identities.
A shortcoming of this study and archival research in general is that the above-mentioned conclusions have been formed via inference and have not been experimentally tested. As a result there are various alternative explanations and limitations to the study. One possible explanation for the mixed results is that people may perceive that the self-presentational benefits and costs that are associated with being a sport fan are nonexistent in a deindividuated medium such as the WWW. Fans may perceive that there are no self-presentational benefits associated with BIRGing when their "self" is hidden in this deindividuated medium. Conversely, the deindividuated medium itself may protect a fan's identity following losses. A fan may think, 'Others are not perceiving me as a loser because they do not know who I really am'. Future research should examine these possibilities.
A second explanation for the finding that fans post more messages on the message boards of successful teams than unsuccessful teams could be that successful teams have more issues to discuss. Unlike their unsuccessful competitors, fans of successful teams have potential playoff match ups, the homefield advantage, how the impact of other games affects their team's playoff chances, and other post-season topics to discuss via the message boards.
Another explanation is that the use of teams' message boards may unfold as the season itself unfolds and may be influenced by perceptions of the probability that one's message will receive a response. At the beginning of the season, fans may post a similar amount of messages on each team's message board. As games begin, the number of posts on successful teams' board may stay constant or increase as the result of fans who post to BIRG or even a result of engaging in "fair weathered fan" behavior. Fans of teams that are losing may attempt to distance themselves from the team and CORF while "fair weather fans" may jump off the bandwagon, causing the number of messages posted to dwindle. Users of the message boards who desire that others respond to their messages may realize the best way to illicit a response is to post a message on a successful team's page.
There are several other limitations of this study. One limitation is that several assumptions had to be made about the message board users. One assumption is that the people who use the message boards are sport fans. A second assumption is that the pool of fans who used the message boards over the period of time that data were collected was relatively stable. If this was the case, discrepancies in the number of posts could be interpreted as BIRGing and not the byproduct of "fair weathered fans". A second limitation of the study is the sample itself, specifically the success level of teams in the sample and the point during the season when data were collected. As mentioned above, fans' message-posting behaviors may have changed over the course of the season and research examining the first four weeks of the season may provide data that contradict my findings. By choosing teams based on postseason potential or lack of potential with four weeks remaining in the season, the teams selected and thus their fans usi ng their message boards may represent extremes, the best and the worst, The problem arises in the generalizability of these results to the fans of the teams that fall within these two extremes
Despite these shortcomings, this study contributes to the current literature in a number of ways. One contribution is that it demonstrates that sport fans use the WWW to manage their sport fan identity. In addition to the posting messages that BIRG, blast opponents, and/or share information, fans provide direct links from their home pages to the home pages of their favorite teams. The WWW has become another tool to enable sport fans to reap the benefits of associating with a successful sport team (BIRG). Finally, I have demonstrated that the WWW enables sport fans to blast other fans, a method that may be unavailable to them in face to face interaction.
In addition to examining sport fans' use of the WWW over the course of an entire season, there are other issues that should be examined in future research. An attempt should be made to study the users themselves. Are sport fans who provide links to their favorite teams' official home pages and who post messages on their teams' message boards representative of "average" sport fans? Probably not, but this sample is probably representative of the subsample of NFL fans who have internet access. Information such as their sport fan identification level, demographics, and self-report of motivations may help us to better understand and predict sport fans' WWW usage. This information may also shed light on the importance of one's "cyber" sport fan identity. How does this identity compare to their face-to-face identity? Is one's "cyber" sport fan identity more resilient to a team's failure than their sport fan identity? Finally, research should attempt to examine the effects of blasting on message boards on rival fans face to face behavior. Do these posts effectively act as release for frustrated and defeated fans or do they just prime them for physical confrontation at the ballpark?
References
Cialdini, R.B., Borden, R.J., Thorne, A., Walker, M.R., Freeman, S., & Sloan. L.R. (1976). Basking in reflected glory: Three (football) studies. Journal of Personality and Social Psychology 34, 366-375.
Cialdini, R.B., & Richardson, K.D. (1980). Two indirect tactics of image management: Basking and blasting. Journal of Personality and Social Psychology, 39, 406-415.
Dietz-Uhler, B., Harrick E.A., End, C.E. & Jacquemotte, L. (2000). Sex differences in sport fans. Journal of Sport Behavior, 23, 219-231.
End, C.E., Dietz-Uhler, B., Harrik, E.A., & Jacquemotte, L. (2001, in press). Identifying with winners: A reexamination of sport fans' tendency to BIRG Journal of Applied Social Psychology.
Gackenbach, J., & Ellerman, E. (1998). Introduction to psychological aspects of internet use. In J. Gackenbach (Ed.), Psychology and the Internet: Intrapersonal. Interpersonal, and Transpersonal Implications (pp. 1-26). San Diego: Academic Press.
Hirt, E.R., Zillman, D., Erickson, G.A., & Kennedy, C. (1992). Costs and benefits of allegiance: Changes in fans' self-ascribed competencies after team victory versus team defeat. Journal of Personality and Social Psychology, 63, 724-738.
Joinson, A.N. (1998). Causes and implications of disinhibited behavior on the internet. In J. Gackenbach (Ed.), Psychology and the Internet: Intrapersonal, Interpersonal, and Transpersonal Implications (pp. 43-60). San Diego: Academic Press.
Joinson, A. N. & Harris, P.R. (1995). Self-enhancement and self protection on the Internet: A study of football fans on the WWW. Paper presented at the BPS London Conference, Institute of Education, London.
Kiesler, S. (1997). Culture of the Internet (pp. ix-xvi). Mahwah, NJ: Lawrence Earlbaum Associates.
Lalonde, R.N. (1992). The dynamics of group differentiation in the face of defeat. Personality and Social Psychology Bulletin, 18, 336-342.
Mann, L. (1974). On being a sore loser: How fans react to their team's failure. A Australian Journal of Psychology, 26, 37-47.
Reid, E. (1998). The self and the internet: Variations on the illusion of one self. In J. Gackenbach (Ed.), Psychology and the Internet: Intrapersonal. Interpersonal, and Transpersonal Implications (pp. 29-42). San Diego: Academic Press.
Rickard, J. (1993, September/October). Home-grown BB$. Wired, pp.42-45
Russell, G. W. (1981). Spectator moods at an aggressive sports event. Journal of Sport Psychology, 3, 217-227.
Sloan, L.R. (1989). The motives of sports fans. In J.H. Goldstein (Ed.), Sports, games, and play: Social and psychological viewpoints (2nd ed., pp.175-240). Hillsdale, NJ: Erlbaum.
Snyder, C.R., Higgens, R.L., & Stucky, R.J. (1983). Excuses: Masquerades in search of grace. New York: Wiley-Interscience.
Tajfel, H. & Turner, J.C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. Austin (Eds.), Psychology of intergroup relations (2nd ed., pp. 7-24). Chicago: Nelson-Hall.
Thomas, J.R., Salazar, W., & Landers, D.M. (1991). What is missing in p [less than].05? Effect size. Research Quarterly for Exercise and Sport, 62(3), 344-348.
Turkle, S. (1995). Identity crisis. In Life on the screen: Identity in the age of the internet (pp. 255-269). New York: Simon & Shuster.
Walther, J.B. (1996). Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23, 3-43.
Wann, D.L. & Branscombe, N.R. (1990). Die-hard and fair-weather fans: Effects of identification on BIRGing and CORFing tendencies. Journal of Sport and Social Issues, 14, 103-117.
Wann, D.L. & Dolan, T.J. (1994). Attributions of highly identified sports spectators. Journal of Social Psychology, 134, 783-792.
Wann, D.L. & Hamlet, M.A. (1995). Author and subject gender in sports research. International Journal of Sport Psychology, 26, 225-232.
Wann, D.L., Hamlet, M.A., Wilson, T.M., & Hodges, J.A. (1993). Basking in reflected glory, cutting off reflected failure, and cutting off future failure: The importance of group identification. Social Behavior and Personality 25, 377-388.
Wann, D.L., Inman, S., Ensor, C.L., Gates, R.D., & Caldwell, D.S. (1999). Assessing the psychological well-being of sport fans using the profile of mood states: The importance of team identification. International Sports Journal, Winter, 81-90.
Wann, D.L., Tucker, K.B., & Schrader, M.P. (l996). An exploratory examination of the factors influencing the origination, continuation, and cessation of identification with sports teams. Perceptual and Motor Skills, 82, 995-1001.
Author's Acknowledgments
The author would like to thank Dr. Beth Dietz-Uhler, Dr. Richard C. Sherman, and Dr. Robert S. Weinberg for their feedback and contributions. Percentage of Messages Indicating Identification with a Team as a Functi Overall Success Identification No Identification Overall Success Successful Teams 46.9% 53.1% Unsuccessful Teams 50.9% 49.1% Percentage of Messages Indicating with a Team as a Function of Outcome Identification No Identification Outcome Victory 51.7% 48.3% Defeat 51.4% 48.6% Percentage of Total Messages by Content as a Function of Outcome BIRGing Blasting Opponent Blast by Opponent Outcome Victory 6.7% 12.4% 21.5% Defeat 2.3% 10.0% 20.5% Information Sharing Outcome Victory 61.0% Defeat 67.2%
