Team identification and sport fandom: gender differences in relationship-based and recognition-based perceived antecedents.

Koch, Katrina ; Wann, Daniel L.

In recent decades, sport scientists from a number of disciplines have shown an increased interest in the antecedents of sport team identification (the extent to which a fan feels a psychological connection to a team, Wann, Melnick, Russell, & Pease, 2001). Wann (2006a) developed a typology in which antecedents were classified as psychological, environmental, or team-related. Psychological antecedents involve basic psycho-social needs and include the need for belonging/affiliation (Donavan, Carlson, Zimmerman, 2005; Gwinner & Swanson, 2003; James, Kolbe, & Trail, 2002; Pritchard, Stinson, & Patton, 2010) and the need for distinctiveness (Dimmock & Gucciardi, 2008; Hyatt & Andrijiw, 2008). Environmental antecedents focus on the fan's surroundings. Thus, factors such as one's socialization into fandom (James, 2001; Kolbe & James, 2003), the presence of rivals (Luellen & Wann, 2010), living in close proximity to the team (Jones, 1997; Wann, Tucker, & Schrader, 1996), and the team's stadium (Swyers, 2005; Underwood, Bond, & Baer, 2001) are best classified as environmental causes. Team-related reasons focus on the team and players. These include perceived similarity with the team (e.g., fans' beliefs that they share commonalities with the players and have similar attitudes, see Aden & Titsworth, 2012; Fisher, 1998; Pritchard et al., 2010), player attributes (Nelson, 2004), and the team's history (Boyle & Magnusson, 2007; Kolbe & James, 2003).

Gender Differences in Antecedents to Team Identification

Although major advances have been established in our understanding of the antecedents of team identification, little research has been devoted to the examination of individual difference variables. Although a few studies have targeted specific groups such as fans of women's sports (Nelson, 2004), followers of newly formed teams (Lock, Darcy, & Taylor, 2009; Lock, Taylor, & Darcy, 2011), and fans varying in levels of self-esteem (Dimmock & Gucciardi, 2008), our understanding of how persons from different demographic groups varying in causes of identification is extremely limited. Yet demographics have been found to play a critical role in the sport fan and spectator experience. For instance, researchers have noted racial differences in fan motivation (Bilyeu & Wann, 2002) and reasons for sport attendance (Armstrong, 2008). Age has also been found to be important. For instance, older fans tend to be more "old school" than younger fans (Aiken, Campbell, & Sukhdial, 2010; Sukhdial, Aiken, & Kahle, 2002). That is, older ("old school") fans tend to be upset by athletes' materialism, believe athletes should serve as role models, and feel that how a sport is played (i.e., the process) is just as important as the outcome of an event (i.e., the product). In other research, younger fans were more likely to report disappointment in response to poor team outcomes (Rainey, Larsen, & Yost, 2009).

In addition to race and age, gender has been identified as an important predictor of sport fan and spectator reactions. For example, gender differences in fan motivation have often been found (but not always, e.g., Allen, Drane, & Byon, 2010; Armstrong, 2002). Female fans often report higher levels of family motivation while men report higher levels of eustress (i.e., positive stress) and self-esteem motives (Wann, 1995; Wann, Schrader, & Wilson, 1999). Men and women also report different behaviors as fans, as men more frequently discuss sport with friends and watch sport on television (Dietz-Uhler, Harrick, End, & Jacquemotte, 2000). Consistent with this finding, men have typically been found to report higher levels of fandom than women (Bahk, 2000; End, Meinert, Worthman, & Mauntel, 2009; Meier & Leinwathre, 2012; Wann, 2002).

The gender difference most relevant to Study 1 of the current investigation concerns differences in team identification. Researchers have found that men often report higher levels of identification than women. This finding has been replicated across a wide variety of settings including college athletic programs (Judson & Carpenter, 2005; Kwon & Armstrong, 2002), Arena Football (Greenwood, 2001), and professional baseball, basketball, and football teams (Wann, Dolan, McGeorge, & Allison, 1994). Interestingly, other researchers have failed to find gender differences in identification (Reding, Grieve, Derryberry, & Paquin, 2011; Robinson & Trail, 2005; Wann, Dimmock, & Grove, 2003) and, in certain circumstances (e.g., women's sports), women have reported higher levels of identification than men (Williamson, Zhang, Pease, & Gaa, 2003). Thus, although the precise pattern of effects is likely complicated, research suggests that, in many instances, men and women differ in their level of team identification.

The Current Investigation

The current investigation was designed to further our understanding of gender differences in team identification by investigating the origination of identification. Two antecedents were investigated: relationship-based origins of team identification and recognition-based origins of team identification. We operationally defined relationship-based origins as one's desire to originally follow a team to help establish and maintain connections with others. This antecedent is similar to affiliation and, as noted above, investigators have found that the desire to affiliate and establish relationships can serve as a powerful antecedent to team identification (Donovan et al., 2005; Gwinner & Swanson, 2003; James et al., 2002; Pritchard et al., 2010). On the other hand, recognition-based origins of team identification is defined as one's aspiration to be known, perceived, or recognized as a fan of a particular team. Individuals will typically choose to engage in sport (both as a fan and as a player) in an effort to be viewed by others in a particular manner (Leary, 1992; Wann, 1997; Wann, Royalty, & Roberts, 2000). As it relates to team identification, some individuals may choose to follow a team to be viewed by others as a fan of that team (Wann et al., 2001).

With respect to gender, although researchers had yet to examine differences between men and women in recognition-based origins of team identification, recent work suggests that women may be more likely than men to originally follow a sport team for relational reasons, that is, for the opportunity to increase connections with others. One such study (Koch & Wann, 2013) investigated differences in team identification among fans socialized to follow a team (i.e., "parents, siblings, and/or friends actively encouraged" the following of the team, p. 135) and those who self-selected the team (i.e., "because I chose to, without the encouragement of others", p. 135). These authors reported an unexpected finding in that female fans were more likely to report having been socialized to follow a team than men, suggesting that relationships may have been particularly important in the development of identification for women. There is prior work in social and developmental psychology to substantiate this finding. For instance, research indicates that women are particularly likely to value group memberships for the relationships they accrue (Seeley, Gardner, Pennington, & Gabriel, 2003). Others have argued that women tend to be more interdependent than men, placing greater importance on establishing and maintaining relationships (Cross & Madsen, 1997; McGuire & McGuire, 1982) and strong empirical support for this pattern of effects has been established (Gabriel & Gardner, 1999).

Koch and Wann (2013) concluded that women may be more likely to become fans of a team to bond with their socialization agents (e.g., peers, family, friends) while men become fans of the team both to bond with others and to be a member of the fan group. There is support for Koch and Wann as research suggests that the importance of the relationship antecedent is greater for female fans than male fans. Aiken and Koch (2009) examined gender differences in importance rankings for a variety of antecedents to team identification. Women reported that social affiliation was the number one antecedent while men rated this factor near the bottom. Similarly, Yoh, Pai, and Pedersen (2009) found that female adolescents were more likely than men to report that friends and family were influential in the development of their loyalty to a team. In the current line of investigation, we expanded on these prior investigations by specifically assessing and examining gender differences in relationship-based and recognition-based origins for originally identifying with a sport team.

Based on the aforementioned literature, we examined the following hypotheses and research questions:

Hypothesis 1: It was hypothesized that female fans would be more likely than male fans to report that their team identification had originated for relationship-based reasons.

Research Question 1: Will women and men significantly differ in their level of recognition-based team identification? This issue was addressed as a research question rather than a hypothesis because the lack of prior research on this topic rendered the development of a hypothesis inappropriate.

Research Question 2: Will women and/or men be more likely to report having originally identified with a team due to relationship-based or recognition-based origins? This issue was also addressed as a research question rather than a hypothesis because the lack of prior research on this topic rendered the development of a specific hypothesis inappropriate.

Study 1

Method

Participants. The original sample contained 174 university students. Twenty-seven questionnaires were discarded because the participant did not have a favorite team. This resulted in a final sample of n = 147. Of these participants, 57 were male and 90 were female. Their ages ranged from 18 years old to 51 years old, with a mean age of 20.66 (SD = 4.25). Men's varsity sports offered at the participants' university were baseball, basketball, cross country, football, golf, rifle, and tennis. Women's sports offered were basketball, cross country, golf, rifle, soccer, softball, tennis, track and field, and volleyball.

Materials and procedure. Subsequent to receiving IRB approval and gaining participant consent, participants then received a questionnaire packet containing three questionnaires. The first questionnaire consisted of two demographic items (age and gender) and one contingency question that assessed whether the participant had a favorite team. A positive response to the third item necessitated the completion of the remainder of the packet. A negative response to the third question terminated the session for those participants. These individuals were then debriefed and excused from the testing session.

The second questionnaire began with an open-ended item asking participants to indicate their favorite sport team. They then completed the Sport Spectator Identification Scale (SSIS; Wann & Branscombe, 1993) targeting that team. The SSIS is comprised of seven Likert-scale items with response options varying from 1 to 8; a lower score indicating low identification and a higher score indicating high identification. The scale has been successfully translated into multiple languages (Wann et al., 2001) and is psychometrically sound. For instance, Wann and Branscombe (1993) report Cronbach's alpha greater than .90, a significant one-year test-retest reliability (r = .60), and demonstrated the validity of the measure as scores on the SSIS were logically related to external assessments such as involvement, biased attributions, and consumption. A sample item from this scale reads, "How important to you is it that (Target Team) wins?"

The third questionnaire was developed specifically for this project. This scale was labeled the Relationship and Recognition Fan Origin Scale--Team (RRFOS-T) and contained 9 Likert-format items. Response options ranged from 1 (Strongly Disagree) to 8 (Strongly Agree) and, similar to the SSIS, participants were instructed to target their favorite team when completing the RRFOS-T. A higher score on each item reflected a greater belief that the origin of one's fandom for the favorite team was due to either relationship-based or recognition-based reasons. That is, questions on the RRFOS-T were designed to assess the extent to which a participant originally began following his or her favorite team for relationship-based reasons and the degree to which he or she began following the team for recognition-based reasons. As noted previously, relationship-based origins involve following a given team to connect with friends, family, and peers who are also fans of that team. Recognition-based origins involve following a team to be recognized as a fan of that team. A sample item designed to assess relationship-based origins of team identification read: "I became a fan of my favorite team to bond with my family, friends, and/or peers." A sample item designed to assess recognition-based origins of team identification read: "I became a fan of my favorite team to be known as a fan of that team."

After completing the questionnaire packet, participants received a debriefing statement explaining the purpose of the study. This form provided contact information if the participant wished to obtain a copy of the results. The entire session lasted approximately 15 minutes.

Results

Preliminary analyses. The Relationship and Recognition Fan Origin Scale-Team (RRFOS-T) was first examined via exploratory factor analysis with varimax rotation. This analysis revealed two subscales containing a total of nine items. The first subscale was labeled the Relationship-based Origin subscale and contained six items (Eigenvalue = 4.95, percent of variance = 55.02). The second subscale, Recognition-based Origin, contained three items (Eigenvalue = 2.08, percent of variance = 23.12). (1)

Items contained in each scale/subscale were summed and then divided by the number of items in each to establish scale scores for each measure. Cronbach's reliability alphas revealed that each was internally consistent (SSIS alpha = .91; Relationship-based Origin = .95; Recognition-based Origin = .84). The sample reported a mean SSIS score of 5.91 (SD = 1.51) on a scale ranging from 1 to 8. This indicates that the sample, as a whole, had a moderately high level of identification with their favorite team (Wann et al., 2001). There were no significant gender differences found on team identification (male M= 6.12, SD = 1.56; female M = 5.77, SD = 1.47; F< 2.00, p > .15). Therefore, team identification was not included in subsequent analyses. The correlation between the Relationship-based and Recognition-based subscales was not significant (r = .030, p = .722).

Tests of hypotheses. To test the hypothesized pattern of effects, a 2 (Gender: male or female) x 2 (RRFOS-T Subscale: Relationship-based Origin and Recognition-based Origin) multivariate analysis of variance (MANOVA) was used (see Table 1 for means and standard deviations). A gender multivariate main effect was found, Wilks' Lambda F(2, 144) = 5.03, p < .01. Follow up univariate analyses found that, consistent with Hypothesis 1, women were significantly more likely to become fans of their favorite team for relationship-based reasons {M = 4.82; SD = 1.90) than were men (M = 3.89; SD = 1.87), F(1, 145) = 8.46, p < .01. With respect to Research Question 1, there were no significant gender differences for recognition-based fandom, F(1, 145) = 1.28, p > .05 (male M = 5.83, SD = 1.61; female M = 5.53; SD = 1.50). As for Research Question 2, follow-up paired samples t-tests indicated that both men and women were more likely to report originally following a team for recognition-based reasons than for relationship-based reasons (women t = 2.95, p < .005; men t = 5.90, p < .001).

Discussion

Although a number of authors have examined antecedents of sport team identification, the relationship between gender and antecedents had yet to be fully investigated. In Study! we tested for gender differences in relationship-based origins (one's desire to originally follow a team to help establish and maintain connections to others) and recognition-based origins (aspirations to be perceived as a fan of a particular team). Based on previous literature (Aiken & Koch, 2009; Koch & Wann, 2013), we hypothesized that women would be more likely than men to report that their team identification had originated for relationship-based reasons. This hypothesis was supported. As noted previously, women tend to highly value relationships they receive from their group members (Cross & Madsen, 1997; Gabriel & Gardner, 1999; Seeley et al., 2003). The current data suggest that this pattern of effects extends to the realm of sport team identification and the process of becoming allegiant to a specific team. Women appear to be more likely than men to originally follow a sport team to establish and/or maintain social relationships via their team fandom.

However, the findings of Research Questions 1 and 2 further frame the aforementioned finding. Although women were more likely than men to endorse relationship-based origins for identifying with a team (as expected), both genders were more likely to endorse recognition-based origins than relationship-based. Men and women did not report differential levels of recognition-based team identification. Thus, although persons do report originally following a team for relationship-based reasons (especially women), they are significantly more likely to originally support a team in an attempt to be perceived as a fan of that particular team. Potential explanations for the processes underlying these findings are examined in the General Discussion section.

Study 2

Study 1 focused on the antecedents of team identification and gender differences in relationship-based and recognition-based origins. Study 2 extended Study 1 by investigating relationship-based and recognition-based origins of sport fandom. Although similar concepts, team identification and sport fandom are distinct (Wann et al., 2001). Sport fandom involves having an interest in a sport, team, and/or specific player(s). On the other hand, as noted above, team identification concerns one's specific association with a team and its players. Sport fans may be found at all levels of identification. Some persons may enjoy following sport (i.e., have high levels of fandom), yet have low levels of identification for any particular team. Likewise, it is conceivable for an individual to develop a strong attachment to a particular team yet remain less enthusiastic about the sport played by the team. Empirical support for the distinctiveness of these concepts is found in research finding that the positive correlation between fandom and identification is moderate in magnitude (Hoogland et al., in press; Wann, 2002).

Study 2 was designed to test for gender differences in relationship-based and recognition-based origins of fandom. We examined two forms of fandom: general fandom (i.e., one's level of overall interest in sport) and fandom for one's favorite sport. Similar to research on team identification, authors have identified important predictors of sport fandom. For instance, the socialization process is a key factor in one's original decision to follow sport (both generally and for a specific team). In fact, the powerful influence of socialization has been documented in a number of countries including the United States (Wann et al., 2001), Norway (Melnick & Wann, 2004), Australia (Melnick & Wann, 2011), Greece (Theodorakis & Wann, 2008), and the United Kingdom (Parry, Jones, & Wann, 2014). Essentially, individuals are far more likely to originally become sport fans if others in their environment (i.e., socialization agents) encourage participation in the pastime. Trait curiosity is another antecedent to fandom (Park, Andrew, & Mahony, 2008). These authors found that individuals high in trait curiosity (interest in novel stimuli) were more likely to be interested in consuming new sports than those with lower levels of trait curiosity. Other potential antecedents include socioeconomic status (Liebennan, 1991; Mashiach, 1980) and one's own prior involvement in sport as an athlete (Grove, Pickett, & Dodder, 1982; Shank & Beasley, 1998).

As noted previously, research has consistently (but not exclusively) identified a gender difference in fandom in which men report higher levels than women (Bahk, 2000; End et al., 2009; Meier & Leinwathre, 2012; Wann, 2002). More importantly for the current study, studies suggest there may be gender differences in antecedents of fandom. That is, consistent with past work on team identification and the results of Study 1 above, male and female fans may be differentially influenced by various antecedents. Although patterns vary across culture (Parry et al., 2014), research suggests that socialization agents may differentially influence men and women. For instance, women often report that mothers were a more influential agent than do men (Melnick & Wann, 2004; Melnick & Wann, 2011; Wann et al., 2001). In addition, researchers have found that relationships (e.g., enjoyment of sport consumption with others) are a greater antecedent for women than men (Dietz-Uhler et al., 2000). This mirrors research in developmental psychology indicating that young girls tend to emphasize relationships to a greater extent than young boys (Nicoloppulou, 2008) and these patterns can be prominent in the socialization process of young female sport fans (Markovits & Albertson, 2012).

Thus, Study 2 was designed to extend Study 1 by investigating gender differences in relationship-based and recognition-based fandom origins. We investigated the following hypotheses and research questions:

Hypothesis 2: Consistent with Study 1 and past work (Dietz-Uhler et al., 2000; Markovits & Albertson, 2012), it was hypothesized that women would be more likely than men to report that their fandom (both general fandom and fandom for their favorite sport) had originated for relationship-based reasons.

Research Question 3: Will women and men significantly differ in their level of recognition-based origins of fandom? This issue was again addressed as a research question rather than a hypothesis because of the lack of prior research on this topic. Although Study 1 results suggested there should not be differences, this single study (and one with a focus on identification rather than fandom) was not deemed sufficient to warrant the development of a hypothesis.

Research Question 4: Will women and/or men be more likely to report having originally followed sport for relationship-based or recognition-based origins? Once again, this topic was examined as a research question rather than a hypothesis due to the lack of prior research. Again, although the findings of Study 1 suggest that persons will be more likely to endorse recognition-based origins than relationship-based origins, this single study (targeting identification) was not viewed as sufficient to warrant the development of a hypothesis.

Method

Participants. The original sample contained 204 university students. However, 56 participants were removed from the sample because they were not a sport fan. This resulted in a final sample of n = 148. Of these participants, 70 were male and 78 were female. Their ages ranged from 18 years old to 38 years old, with a mean age of 20.22 (SD = 3.07). Participants comprising the Study 2 sample were acquired from a new participant pool and, therefore, had not participated in Study 1.

Materials and procedure. Subsequent to receiving IRB approval and gaining participant consent, participants then received a questionnaire packet containing five questionnaires. The first questionnaire consisted of two demographic items (age and gender) and one contingency question that assessed whether the participant was a sport fan. A positive response to the third item necessitated the completion of the remainder of the packet. A negative response to the third question terminated the session for those participants. These individuals then debriefed and excused from the testing session.

The second scale was the Sport Fandom Questionnaire (SFQ; Wann, 2002) used to measure participants' degree of general sport fandom. The SFQ is a reliable and valid unidimensional scale contains five Likert-scale items. Response options ranged from 1 (strongly disagree) to 8 (strongly agree). Therefore, higher numbers represented greater levels of general sport fandom. A sample item read: "Being a sport fan is very important to me."

The third questionnaire was based on the previously discussed 9-item RRFOS-T. However, rather than targeting origin of identification for a team, the current version (labeled the Relationship and Recognition Fan Origin Scale--General Fandom; RRFOS-GF) investigated reasons for originally becoming interested in sport as a fan. Once again, response options ranged from 1 (Strongly Disagree) to 8 (Strongly Agree) and, thus, a higher score on each item reflected a greater belief that the origin of one's general sport fandom was either relationship-based or recognition-based. A sample item designed to assess relationship-based origins of general fandom read: "I became a sport fan, in general, to bond with my family, friends, and/or peers." A sample item designed to assess recognition-based origins of general fandom read: "I became a sport fan, in general, to be known as a sport fan."

The fourth scale was a second version of the SFQ. Rather than targeting general fandom, items on this version targeted fandom for a favorite sport (SFQ-FS). Each participant listed his/her favorite sport to follow and then answered the five SFQ-FS items for that sport. Once again, response options ranged from 1 (strongly disagree) to 8 (strongly agree) resulting in higher numbers reflecting greater levels of fandom for the participant's favorite sport.

The final questionnaire was a second version of the RRFOS. Rather than assessing origin for general fandom, this form focused on fandom for one's favorite sport. This version, labeled the Relationship and Recognition Fan Origin Scale--Favorite Sport (RRFOS-FS), measured one's reasons for originally becoming interested in his or her favorite sport. Participants listed the same favorite sport they targeted when completing the SFQ-FS and then answered the nine RRFOS-FS items for that sport. Response options again ranged from 1 (Strongly Disagree) to 8 (Strongly Agree). Consequently, a higher score reflected a greater belief that the origin of one's fandom for a favorite sport was due to either relationship-based or recognition-based reasons. A sample item assessing relationship-based origins for a favorite sport read: "I became a sport fan of the sport listed above to bond with my family, friends, and/or peers." A sample item assessing recognition-based origins for a favorite sport read: "I became a sport fan of the sport listed above to be known as a fan of that sport."

After completing the questionnaire packet, participants received a debriefing statement explaining the purpose of the study. This form also provided contact information should the participant wish to obtain a copy of the results. The session lasted approximately 15 minutes.

Results

Preliminary analyses. The Relationship and Recognition Fan Origin Scale--General Fandom and the Relationship and Recognition Fan Origin Scale--Favorite Sport were analyzed via exploratory factor analysis with varimax rotation. Consistent with Study 1, each analysis revealed two subscales containing nine items. Once again, the first subscale was labeled Relationship-based Origin and contained six items (RRFOS-GF Eigenvalue = 5.77, percent of variance = 64.06; RRFOS-FS Eigenvalue = 5.99, percent of variance = 66.59). The second subscale, again called Recognition-based Origin, contained three items (RRFOS-GF Eigenvalue = 1.91, percent of variance = 21.25; RRFOS-FS Eigenvalue = 1.85, percent of variance = 20.54). (1)

Items contained in each scale and subscale were summed and divided by the number of items in each to establish scores for each measure. Cronbach's analyses revealed that each was internally consistent (SFQ alpha = .92; SFQ-FS = .91; Relationship-based General Fandom Origin = .96; Recognition-based General Fandom Origin = .92; Relationship-based Favorite Sport Origin = .97; Recognition-based Favorite Sport Origin = .93). The sample reported a mean SFQ score of 5.52 (SD = 1.63) and a mean SFQ-FS score of 6.15 (SD = 1.56) on scales ranging from 1 to 8. This indicates that the sample, as a whole, had a moderately high level of fandom for both general fandom and fandom for a favorite sport (Wann, 2002). Significant gender differences were not found for either general fandom (SFQ male M= 5.65, SD = 1.89; female M = 5.40, SD = 1.35; F < 1.00, p > .30) or fandom for one's favorite sport (SFQ-FS male M = 6.23, SD = 1.75; female M = 6.07, SD = 1.39; F < 1.00, p > .50). Therefore, neither general nor favorite sport fandom was included in subsequent analyses. Correlations among the four origin subscales are listed in Table 2.

Tests of hypotheses. To test the hypothesized pattern of effects, a pair of 2 (Gender: male or female) x 2 (RRFOS Subscale: Relationship-based Origin and Recognition-based Origin) multivariate analyses of variance (MANOVA) were utilized, one each for participants' general fandom (RRFOS-GF) and their favorite sport (RRFOS-FS). With respect to the origin of participants' general fandom, a gender multivariate main effect was found, Wilks' Lambda F(2, 145) = 8.41, p < .001 (see Table 1 for means and standard deviations). Follow up univariate analyses found that, consistent with Hypothesis 2, women were significantly more likely to originally become sport fans for relationship-based reasons (M = 5.64; SD = 1.34) than were men (M = 4.74; SD = 1.88), F(1, 146) = 11.55, p < .001. There were no significant gender differences for recognition-based origins (Research Question 3), F(1, 146) = 0.37, p > .50 (male M= 3.51, SD = 1.91; female M= 3.35; SD = 1.33). With respect to Research Question 4, follow-up paired samples t-tests indicated that both women and men were more likely to report originally becoming a sport fan, in general, for relationship-based reasons than for recognition-based reasons (female t = 11.91, p < .001; male t = 5.87, p < .001).

As for the origin of fandom for a favorite sport, a gender multivariate main effect was again found, Wilks' Lambda F(2, 145) = 8.01, p < .001 (see Table 1 for means and standard deviations). Consistent with Hypothesis 2 and the previous analysis on one's general fandom, follow up univariate analyses found that women were significantly more likely to become fans of their favorite sport for relationship-based reasons (M= 5.88; SD = 1.37) than were men (M= 4.94; SD = 1.90), F(1, 146) = 12.14, p < .001. As for Research Question 3, consistent with general fandom, there were no significant gender differences for recognition-based origins, F(1, 146) = 0.21, p > .80 (male M = 3.94, SD = 2.10; female M= 3.89; SD = 1.70). Also consistent with the previous analysis (Research Question 4), follow-up paired samples t-tests indicated that both women and men were more likely to report originally becoming a fan of their favorite sport for relationship-based reasons than for recognition-based reasons (female t =8.96, p < .001; male t = 4.90, p < .001).

Discussion

The results of Study 2 indicated that, consistent with expectations, women were more likely than men to report originally became sport fans for relationship-based reasons, a pattern of effects found both for one's favorite sport and general sport fandom. Men and women did not differ in their levels of recognition-based origins. Overall, both men and women were more likely to report becoming a fan (both in general and of a favored sport) for relationship-based reasons than for recognition-based reasons.

General Discussion

Combining the findings of Study 2 with those reported in Study 1 leads to several interesting patterns. First, in both studies women reported higher scores for relationship-based origins than did men. Specifically, women more frequently indicated that the origin of their identification with a favorite team, fandom for a sport favorite sport, and fandom in general were a function of their desire to establish and maintain connections to others. These findings are not only consistent within this research, but also substantiate prior work as well. That is, the data reported above mirror previous studies suggesting that the establishment of relationships is a particularly key factor for women in the development of both their sport fandom (Dietz-Uhler et al., 2000; Markovits & Albertson, 2012) and team identification (Aiken & Koch; 2009; Koch & Wann, 2013; Yoh et al., 2009). This pattern of effects is also consistent with gender research in both social psychology (Cross & Madsen, 1997; Gabriel & Gabriel, 1999; Seeley et al., 2003) and developmental psychology (Nicoloppulou, 2008).

A second consistent result indicated that men and women did not report differential levels of recognition-based origins for any of the three indices of fandom (i.e., team identification, general fandom, and fandom for a favorite sport). Due to a lack of work relevant to this issue, the relationships between gender and levels of recognition-based origins were examined within the framework of research questions rather than specific hypotheses. Thus, the current findings are quite novel. However, because the same pattern was found in each of the three analyses, the effect appears to be robust. Evidently, the importance of being recognized as either a sport fan or a supporter of a specific team is equally important to men and women.

However, other results were inconsistent across the two studies. For origins of team identification, men and women reported higher levels of recognition-based than relationship-based origins. Thus, for the development of identification, being recognized as a follower of a team was of greater importance than the establishment of relationships. However, this pattern was reversed for the origin of fandom. For both general fandom and fandom for one's favorite sport, men and women were more likely to endorse relationship-based than recognition-based origins. Therefore, in terms of development of fandom, establishing relationships was of greater prominence than was gaining recognition as a fan. There appear to be several potential explanations for this pattern of effects.

With respect to sport fandom, it is likely that most (but not all) persons are exposed to sport fandom prior to being socialized to support a specific team (Wann et al., 2001). Because exposure to sport itself likely occurs very early in one's journey into fandom, at this stage, individuals may be investigating those types of people with which they would like to associate. As a result, these persons are interested in becoming fans in part to assist in the formation of social relationships. Essentially, they see fandom as a way to increase connections with others. These persons could choose from any number of activities (e.g., various forms of pop culture, academics, etc.) but they select sport fandom because they wish to establish relationships with others who have a similar general interest in sport. This process continues with interest in a specific sport. Once an individual has determined that he or she would like to establish relationships with others who are sport fans, they then become more particular, establishing relationships with fans of specific sports. That is, they likely become aware that sport fan groups often differ (e.g., fans of golf differ from fans of auto racing) and they choose a specific sport to follow in an attempt to establish relationships with those specific fans. In essence, the decision to become a fan in general and of a specific sport is impacted by an individual's decision to establish relationships with others engaging in that activity.

Having chosen a specific sport to follow, fans now turn to choosing a favorite team. Our data indicate that the choice of one's favorite team is more likely to be a function of a desire to be recognized (i.e., perceived) as a fan of that team. The selection of a favorite team is strongly connected to self-presentational concerns (Cialdini et al., 1976; Wann & Branscombe, 1990; Wann et al., 2000), and persons appear to be more interested in announcing teams they support than sports they follow. For instance, fans purchase a great deal of sports apparel that publicizes their team allegiances. However, it is less common to see apparel describing a fan's favorite sport (e.g., there are fewer persons wearing shirts simply saying "football" or "I'm a basketball fan"). In fact, theory and research suggests that team identification has greater ingroup and well-being implications than does fandom (Wann, 2006b; Wann, Dunham, Byrd, & Keenan, 2004; Wann & Weaver, 2009), a pattern of effects that is consistent with the logic presented here.

Limitations and Suggestions for Future Research

Certain limitations of the current work and suggestions for future research are warranted. First, the current data were collected at a single collegiate institution, thus resulting in a rather homogeneous sample. Subsequent research is needed from other geographic locations and with a wider age range. Second, we limited our investigation to gender differences in antecedents. However, it may be that other demographic variables also relate to relationship-based and recognition-based origins. Thus, additional research is needed that examines additional demographics such as age and race (such an investigation was not appropriate in the current study due to the homogeneous composition of the sample). Given that past work has found that both of these variables are related to fandom (Bilyeu & Wann, 2002; Murrell & Deitz, 1992; Wann et al., 2001), such investigations are justified. Third, the chronological pattern suggested above (i.e., general sport fan [right arrow] fan of a specific sport [right arrow] fan of a specific team) is speculative at this point. Additional research is needed to provide empirical support for its validity.

In addition, it is important to note that although the current work examined gender differences in fandom and identification, past research has indicated that gender role orientation is also critical. Specifically, previous investigations have found that gender role orientation is important in predicting both sport fandom (Wann, Waddill, & Dunham, 2004) and sport fan motivation (Wann & Waddill, 2003). With respect to fandom, Wann et al. (2004) found that masculinity was a better predictor of fandom than was anatomical sex. As for motivation, Wann and Waddill (2003) reported that masculinity was a good predictor of motivation but that femininity best predicted family motivation (i.e., the desire to use fandom to spend time with one's family). Given these findings, future researchers should attempt to examine the relationships among gender role orientation (masculinity, femininity, and androgyny) and relationship-based and recognition based origins for team identification and sport fandom.

In addition, it was noted that the current work is consistent with the notion that one's identification with a sport team is more central to his or her identity than one's sport fandom. However, given that this pattern of effects is speculative, future researchers may want to investigate this issue. Three such studies using slightly different methodologies would be beneficial. First, experimenters may want to adopt a strategy similar to that employed by Wann et al. (2000) in which participants described themselves to another student (they were given a response sheet asking them to list up to 20 self-descriptors). The authors were able to use subjects' mention of their fandom for a team as well as the position of listing as measures of centrality (e.g., listing "I am a fan of the Kentucky Wildcats" as the first descriptor relative to mentioning it later in the list). A similar procedure could be adopted by examining the extent to which individuals mention following a team versus a sport and the relative position of each.

Second, researchers could modify existing scales to compare the relative strength of one's identity as a fan of a sport and a fan of a team. For example, Deaux (2001) suggested that relative strength of varying identities could be assessed by utilizing a slightly modified version of the Collective Self-esteem Scale (Luhtanen & Crocker, 1991, 1992). This psychometrically sound instrument contains items such as "Being a member of a social group is an important reflection of who 1 am" and "In general, belonging to social groups is an important part of my self-image." These and similar items comprise the identity subscale, which was designed to assess "the importance of one's social group memberships to one's self-concept" (p. 305). Deaux suggests that by replacing "a social group" with a specific group, one can assess the strength of the identity for that particular domain. Extending this logic to the current work, researchers could have participants complete the scale twice, once for their fandom for a certain sport (e.g., "Being a fan of baseball is an important reflection of who I am" and once for their fandom for a specific team (e.g., "Being a fan of the Chicago Cubs is an important reflection of who 1 am"). A similar stagey was used by Smith, Grieve, Zapalac, Derryberry, and Pope (2012). These investigators altered the identification scale developed by Crisp, Stone, and Hall (2006) so that it specifically assessed degree of identification with specific groups (e.g., religious, sport, school). By modifying scales as suggested by Deaux (2001) and Smith et al. (2012), researchers should be able to assess the relative strength of social categories for fan of a specific sport and fan of a certain team.

And finally, investigators may wish to assess the loyalty of individuals toward their favorite team and sport using a methodology similar to that employed by Sebastian and Bristow (2000). These authors asked participants to rate their loyalty toward several entities via Likert-scale items such as "Once I get used to a brand of--, I hate to switch." They found that brand loyalty to sport teams was much greater than loyalty to other products such as clothing and food/drink companies. By slightly modifying their scale, researcher should be able to establish relative loyalty of persons to teams versus sports.

Conclusion

The current investigation examined gender differences in the origin of sport fandom and team identification for two antecedents: relationship-based and recognition-based. Women reported higher scores for relationship-based origins than did men and this effect was found for both team identification and fandom. Men and women did not report differential levels of recognition-based origins for any of the three indices of fandom/ identification. For the development of team identification, recognition-based origins were more prominent than were relationship-based origins. This pattern was reversed for the origin of fandom as for both general fandom and fandom a favorite sport, men and women were more likely to endorse relationship-based than recognition-based origins. Subsequent research is needed to extend these findings to other populations and to determine if fandom for a team is typically a more central component of a fan's social identity than is following a specific sport.

Footnote

(1) Factor loadings are available from the second author upon request. Scale items are available from the second author upon request.

References

Aden, R. C., & Titsworth, S. (2012). Remaining rooted in a sea of red: Agrarianism, place attachment, and Nebraska Comhusker football fans. In A. C. Eamheardt, P. M. Haridakis, & B. S. Hugenberg (Eds.), Sports fans, identity and socialization: Exploring the fandemonium (pp. 9-23). Lanham, MD: Lexington.

Aiken, K. D., Campbell, R. M., & Sukhdial, A. (2010). An investigation of Old School values in the Arena Football League. Sport Marketing Quarterly, 19, 125-131.

Aiken, K. D., & Koch, E. C. (2009). A conjoint approach investigating factors in initial team preference formation. Sport Marketing Quarterly, 18, 81-91.

Allen, J. T., Drane, D., & Byon, K. K. (2010). Gender differences in sport spectatorship among college baseball fans. International Journal of Sport Management, 11, 418-439.

Armstrong, K. (2002). Race and sport consumption motivations: A preliminary investigation of a black consumers' sport motivation scale. Journal of Sport Behavior, 25, 309-330.

Armstrong, K. L. (2008). Consumers of color and the "culture" of sport attendance: Exploratory insights. Sport Marketing Quarterly, 17, 218-231.

Bahk, C. M. (2000). Sex differences in sport spectator involvement. Perceptual and Motor Skills, 91, 79-83.

Bilyeu, J. K., & Warm, D. L. (2002). An investigation of racial differences in sport fan motivation. International Sports Journal, 6(2), 93-106.

Boyle, B. A., & Magnusson, P. (2007). Social identity and brand equity formation: A comparative study of collegiate sport fans. Journal of Sport Management, 21, 497-520.

Cialdini, R. B., Borden, R. J., Thome, A., Walker, M. R., Freeman, S., & Sloan, L. R. (1976). Basking in reflected glory: Three (football) field studies. Journal of Personality and Social Psychology, 34, 366-375.

Crisp, R. J., Stone, C. H., & Hall, N. R. (2006). Recategorization and subgroup identification: Predicting and preventing threats from common ingroups. Personality and Social Psychology Bulletin, 32, 230-243.

Cross, S. E., & Madsen, L. (1997). Models of the self: Self-construals and gender. Psychological Bulletin, 122, 5-37.

Deaux, K. (2001). Social identity. In J. Worell (Ed.) Encyclopedia of women and gender: Sex similarities and differences and the impact of society on gender (pp. 1-9). Waltham, MA: Academic Press.

Dietz-Uhler, B., Harrick, E. A., End, C., & Jacquemotte, L. (2000). Sex differences in sport fan behavior and reasons for being a sport fan. Journal of Sport Behavior, 23, 219-231.

Dimmock, J. A., & Gucciardi, D. F. (2008). The utility of modern theories of intergroup bias for research on antecedents to team identification. Psychology of Sport and Exercise, 9, 284-300.

Donavan, D. T., Carlson, B. D., & Zimmerman, M. (2005). The influence of personality traits on sports fan identification. Sport Marketing Quarterly, 14, 31-42.

End, C. M., Meinert, J. L., Jr., Worthman, S. S., & Mauntel, G. J. (2009). Sport fan identification in obituaries. Perceptual and Motor Skills, 109, 551-554.

Fisher, R. J. (1998). Group-derived consumption: The role of similarity and attractiveness in identification with a favorite sports team. Advances in Consumer Research, 25, 283-288.

Gabriel, S., & Gardner, W. L. (1999). Are there "his" and "hers" types of interdependence? The implications of gender differences in collective versus relational interdependence for affect, behavior, and cognition. Journal of Personality and Social Psychology, 77, 642-655.

Greenwood, P. B. (2001). Sport fan team identification in a professional expansion setting. Unpublished Masters Thesis, North Carolina State University.

Grove, S. J., Pickett, G. M., & Dodder, R. A. (1982). Spectatorship among a collegiate sample: An exploratory investigation. In M. Etzel & J. Gaski (Eds.,) Applying marketing technology' to spectator sports (pp. 26-40). South Bend, IN: University of Notre Dame Press.

Gwinner, K., & Swanson, S. R. (2003). A model of fan identification: Antecedents and sponsorship outcomes. Journal of Services Marketing, 17, 275.294.

Hoogland, C. E., Schurtz, D. R., Cooper, C. M., Combs, D. Y. J., Brown, E. G., Smith, R. H. (in press). The joy of pain and the pain of joy: In-group identification predicts schadenfreude and gluckschmerz following rival groups' fortunes. Motivation and Emotion.

Hyatt, C. G., & Andrijiw, A. K. (2008). How people raised and living in Ontario became fans of non-local National Hockey League teams. International Journal of Sport Management and Marketing, 4, 338-355.

James, J. D. (2001). The role of cognitive development of socialization in the initial development of team loyalty. Leisure Sciences, 23, 233-261.

James, J. D., Kolbe, R. H., & Trail, G. T. (2002). Psychological connection to a new sport team: Building or maintaining the consumer base? Sport Marketing Quarterly, 11, 215-225.

James, J. D., & Ridinger, L. L. (2002). Female and male sport fans: A comparison of sport consumption motives. Journal of Sport Behavior, 25, 260-278.

Jones, I. (1997). A further examination of the factors influencing current identification with a sports team, a response to Wann et al. (1996). Perceptual and Motor Skills, 85, 257-258.

Judson, K. M., & Carpenter, P. (2005). Assessing a university community's identification to sport in a changing climate. Sport Marketing Quarterly, 4, 217-226.

Koch, K., & Wann, D. L. (2013). Fans' identification and commitment to a sport team: The impact of self-selection versus socialization processes. Athletic Insight, 5, 129-143.

Kolbe, R. H., & James, J. D. (2003). The internalization process among team followers: Implications for team loyalty. International Journal of Sport Management, 4, 25-43.

Kwon, H. H., & Armstrong, K. L. (2002). Factors influencing impulse buying of sport team licensed merchandise. Sport Marketing Quarterly 3, 151-163.

Leary, M. R. (1992). Self-presentational processes in exercise and sport. Journal of Sport & Exercise Psychology, 14, 339-351.

Lieberman, S. (1991, September/October). The popular culture: Sport in America-a look at the avid sports fan. The Public Perspective: A Roper Center Review of Public Opinion and Polling, 2(6), 28-29.

Lock, D., Darcy, S., & Taylor, T. (2009). Starting with a clean slate: An analysis of member identification with a new sports team. Sport Management Review, 12, 15-25.

Lock, D., Taylor, T., & Darcy, S. (2011). In the absence of achievement: The formation of new team identification. European Sport Management Quarterly, 11, 171-192.

Luellen, T. B., & Wann, D. L. (2010). Rival salience and sport team identification. Sport Marketing Quarterly, 19, 97-106.

Luhtanen, R., & Crocker, J. (1991). Self-esteem and intergroup comparison: Toward a theory of collective self-esteem. In J. Suls & T. A. Wills (Eds.), Social comparisons: Contemporary theory and research (pp. 211-234). Hillsdale, NJ: Erlbaum.

Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self-evaluation of one's social identity. Personality and Social Psychology Bulletin, 18, 302-318.

Markovits, A. S., & Albertson, E. K. (2012). Sportista: Female fandom in the United States. Philadelphia, Temple University Press.

Mashiach, A. (1980). A study to determine the factors which influence American spectators to go see the Olympics in Montreal, 1976. Journal of Sport Behavior, 3, 17-26.

McGuire, W. J., & McGuire, C. V. (1982). Significant others on self-space: Sex differences and developmental trends in social self. In J. Suls (Ed.) Psychological perspectives on the self(Vol. 1, pp. 71-96). Hillsdale, NJ: Erlbaum.

Meier, H. E., & Leinwathre, M. (2012). Women as "armchair audience"? Evidence from German National Team Football. Sociology of Sport Journal, 29, 365-384.

Melnick, M. J., & Wann, D. L. (2004). Sport fandom influences, interests, and behaviors-among Norwegian university students. International Sports Journal, 5(1), 1-13.

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

Murrell, A. J., & Dietz, B. (1992). Fan support of sports teams: The effect of a common group identity. Journal of Sport and Exercise Psychology, 14, 28-39.

Nelson, K. (2004). Identification with sports teams by fans of women's sports. Perceptual and Motor Skills, 99, 575-576.

Nicolopoulou, A. (2008). The elementary forms of narrative in young children's story-telling. Narrative Inquiry, 18, 299-325.

Park, S.-H., Andrew, D. P. S., & Mahony, D. F. (2008). Exploring the relationship between trait curiosity and initial interest in sport spectatorship. International Journal of Sport Management, 9, 286-302.

Parry, K. D., Jones, I., & Wann, D. L. (2014). An examination of sport fandom in the United Kingdom: A comparative analysis of fan behaviors, socialization processes, and team identification. Journal of Sport Behavior, 37, 251-267.

Pritchard, M. P, Stinson, J., & Patton, E. (2010). Affinity and affiliation: The dual carriage way to team identification. Sport Marketing Quarterly, 19, 67-77.

Rainey, D. W., Larsen, J., & Yost, J. H. (2009). Disappointment theory and disappointment among baseball fans. Journal of Sport Behavior, 32, 339-356.

Reding, F. N., Grieve, F. G., Derryberry, W. P., & Paquin, A. R. (2011). Examining the team identification of football fans at the high school level. Journal of Sport Behavior, 34, 378-391.

Robinson, M. J., & Trail, G. T. (2005). Relationships among spectator gender, motives, points of attachment, and sport preference. Journal of Sport Management, 19, 58-80.

Sebastian, R. J., & Bristow, D. N. (2000). Win or lose, take me out to the ball game! An empirical investigation of loyalty proneness among college students. Sport Marketing Quarterly, 9, 211-220.

Seeley, E. A., Gardner, W. L., Pennington, G., & Gabriel, S. (2003). Circle of friends or members of a group? Sex differences in relational and collective attachment to groups. Group Processes and Intergroup Relations, 6, 251-263.

Shank, M. D., & Beasley, F. M. (1998). Fan or fanatic: Refining a measure of sports involvement. Journal of Sport Behavior, 21, 435-443.

Smith, S. E., Grieve, F. G., Zapalac, R, K., Derryberry, W. P., & Pope, J. (2012). How does sport team identification compare to identification with other social institutions? Journal of Contemporary Athletics, 6, 69-82.

Sukhdial, A., Aiken, D., & Kahle, L. (2002). Are you Old School? A scale for measuring sport fan's old-school orientation. Journal of Advertising Research, 42(A), 71-81.

Swyers, H. (2005). Community America: Who owns Wrigley Field? The International Journal of the History of Sport, 22, 1086-1105.

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

Underwood, R., Bond, E., & Baer, R. (2001). Building service brands via social identity: Lessons from the sports marketplace. Journal of Marketing Theory and Practice, 9, 1-13.

Wann, D. L. (1995). Preliminary validation of the Sport Fan Motivation Scale. Journal of Sport & Social Issues, 19, 377-396.

Wann, D. L. (1997). Sport psychology. Upper Saddle River, NJ: Prentice Hall.

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. (2006a). The causes and consequences of sport team identification. In A. A. Raney & J. Bryant (Eds.), Handbook of Sports and Media (pp. 331-352). Erlbaum.

Wann, D. L. (2006b). Understanding the positive social psychological benefits of sport team identification: The Team Identification--Social Psychological Health Model. Group Dynamics: Theory, Research, and Practice, 10, 272-296.

Wann, D. L., & Branscombe, N. R. (1990). Die-hard and fair-weather fans: Effects of identification on BIRGing and CORFing tendencies. Journal of Sport & Social Issues, 14, 103-117.

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

Wann, D. L., Dimmock, J. A., & Grove, J. R. (2003). Generalizing the Team Identification --Psychological Health Model to a Different Sport and Culture: The Case of Australian Rules Football. Group Dynamics: Theory, Research, and Practice, 7, 289-296.

Wann, D. L., Dolan, T. J., McGeorge, K. K., & Allison, J. A. (1994). Relationships between spectator identification and spectators' perceptions of influence, spectators' emotions, and competition outcome. Journal of Sport and Exercise Psychology', 16, 347-364.

Wann, D. L., Dunham, M. D., Byrd, M. L., & Keenan, B. L. (2004). The five-factor model of personality and the psychological health of highly identified sport fans. International Sports Journal, 8(2), 28-36.

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

Wann, D. L., Royalty, J., & Roberts, A. (2000). The self-presentation of sport fans: Investigating the importance of team identification and self-esteem. Journal of Sport Behavior, 23, 198-206.

Wann, D. L., Schrader, M. R, & Wilson, A. M. (1999). Sport fan motivation: Questionnaire validation, comparisons by sport, and relationship to athletic motivation. Journal of Sport Behavior, 22, 114-139.

Wann, D. L., Tucker, K. B., & Schrader, M. P. (1996). An exploratory examination of the factors influencing the origination, continuation, and cessation of identification with sports teams. Perceptual and Motor Skills, 82, 995-1001.

Wann, D. L., & Waddill, P. J. (2003). Predicting sport fan motivation using anatomical sex and gender role orientation. North American Journal of Psychology, 5, 485-498.

Wann, D. L., Waddill, P. J., & Dunham, M. D. (2004). Using sex and gender role orientation to predict level of sport fandom. Journal of Sport Behavior, 27, 367-377.

Wann, D. L., & Weaver, S. (2009). Understanding the relationship between sport team identification and dimensions of social well-being. North American Journal of Psychology, 11, 219-230.

Williamson, D. P, Zhang, J. J., Pease, D. G., & Gaa, J. P. (2003). Dimensions of spectator identification associated with women's professional basketball game attendance. International Journal of Sport Management, 4, 59-91.

Yoh, T., Pai, H.-T., & Pedersen, P. M. (2009). The influence of socialisation agents on fan loyalty of Korean teens. International Journal of Sport Management and Marketing, 6, 404-416.

Katrina Koch and Daniel L. Wann

Murray State University

Address correspondence to: Daniel L. Wann, Department of Psychology, Murray State University, Murray, KY 42071. Email: dwann@murraystate.edu

Table 1
Means and Standard Deviations for Study 1 and Study 2

                                                   Female

                                                M       SD

Study 1 (Team Identification)
  Relationship-based Team Identification *    4.82    (1.90)
  Recognition-based Team Identification       5.53    (1.50)

Study 2 (Fandom)
  Relationship-based General Fandom **        5.64    (1.34)
  Recognition-based General Fandom            3.35    (1.33)
  Relationship-based Favorite Sport **        5.88    (1.37)
  Recognition-based Favorite Sport            3.89    (1.70)

                                                   Male

                                                M       SD

Study 1 (Team Identification)
  Relationship-based Team Identification *    3.89    (1.87)
  Recognition-based Team Identification       5.83    (1.61)

Study 2 (Fandom)
  Relationship-based General Fandom **        4.74    (1.88)
  Recognition-based General Fandom            3.51    (1.91)
  Relationship-based Favorite Sport **        4.94    (1.90)
  Recognition-based Favorite Sport            3.94    (2.10)

Notes: Significant gender differences indicated
by * = p < .01; ** = p < .001.

Table 2
Pearson Correlations among the RRFOS Subscales
for General Fandom and Favorite Sport

                                         1       2       3       4

Relationship-based General Fandom (1)    --
Recognition-based General Fandom (2)     .41 *   --
Relationship-based Favorite Sport (3)    .88 *   .38 *   --
Recognition-based Favorite Sport (4)     .35 *   .78 *   .44 *   --

Note. * = p < .01.