Examining the potential causal relationship between sport team identification and psychological well-being.

Wann, Daniel L.

A number of social scientists have proposed theoretical models suggesting that social support networks can serve as important buffers from anxiety, loneliness and the like, thereby assisting individuals in maintaining their psychological health (Cohen & Wills, 1985; Rowe & Kahn, 1998; Thoits, 1982; Wann& Hamlet, 1994, 1996). One of the more extensive explanations of the impact of social groups on self-esteem and psychological health is provided by Tajfel (1981) in his social identity theory. Social identity theory posits that individuals prefer and attempt to maintain positive self- and social-identities. By identifying with and maintaining memberships in valued social groups and networks, a person's social-identity can be enhanced, thereby benefiting their overall self-concept (Hogg & Abrams, 1990).

The benefits of social connections are found among many groups, including those who are stigmatized (Crocker & Major, 1989), and in a variety of settings, including sport and physical activity (Bianco & Eklund, 2001). In fact, recent work indicates that social connections are particularly important for stigmatized athletes (Krane, Barber, & McClung, 2002). Further, the benefits of social support are not limited to one's psychology but, rather, also predict both physical health (Cohen, Doyle, Skoner, Rabin, & Gwaltney, 1997) and longevity of life (Rowe & Kahn, 1998).

For many years, authors have suggested that sport fandom is a societal connection that may assist in the maintenance of psychological health (Curtis, Loy, & Karnilowicz, 1986; Eastman & Land, 1997; Melnick, 1993; Pan, Gabert, McGaugh, & Branvold, 1997; Smith, 1988, 1989; Zillmann, Bryant, & Sapolsky, 1989). Indeed, as early as 1929, Brill stated "are you a fan? It is altogether to be hoped, for your psychic health and well-being, that you are" (p. 429). However, until the last decade, claims of the potential psychological benefits of sport fandom had remained largely speculative. To remedy this, Wann and his colleagues have conducted a series of empirical investigations into the relationship between fandom and psychological health. Wann's research program has tested the hypothesis that high levels of identification with (i.e., psychological connection to) a local sport team are related to psychological well-being (see Wann, Melnick, Russell, & Pease, 2001). By identifying with a local sport team, an individual becomes attached to a larger social group. These associations to other fans form the basis for a valuable connection to society at large and serve as a buffer to loneliness, isolation, and so forth. Consequently, those with strong ties to the local team would likely exhibit a more positive psychological profile than those lacking in team identification.

Identification with a distant team and mere sport fandom per se are not expected to provide significant well-being benefits because these factors do not readily lead to social connections, the driving force behind the benefits of team identification (Wann et al., 2001). With respect to identification with a distant team, it is likely difficult for such fans to find and interact with fellow fans of their team. Rather, they are more likely to either be surrounded by fans of rival teams (who may openly dislike and perhaps ridicule the fan's choice) or find that other fans are disinterested in their preferred team. In either situation, it would be difficult to generate social connections through their association with their team. In the case of general fandom, again, the important social connections that accompany identification with a local team are lacking. Social support benefits result from active group participation in an activity, not simple interest (Rowe & Kahn, 1998). For instance, sport consumption can easily be done in the privacy of one's own home via television, radio, and the Internet. Certainly, fans can commune with one another, regardless of their level of interest in any one team (e.g., Monday morning water cooler conversations following Sunday football games). However, interacting with others who share in one's passion for a specific team or player adds an additional interpersonal link. One need only watch the interactions occurring at tail-gating parties to see this in action. Simply put, there is likely to be a greater bond between two die-hard fans of a specific team than between two die-hard fans of a specific sport (similar to the bond that fans of a specific musical group feel, versus the bond felt by fans of a specific type of music). Such logic is supported by decades of research indicating that interpersonal attraction, cohesion, and the like are strongly linked to the similarity of the parties involved (Hendrick & Hendrick, 1992; Kelley et al., 1983; Turner, 1999).

Wann's first attempt to empirically examine the team identification-psychological health relationship involved college students who completed the Sport Spectator Identification Scale (SSIS; Wann & Branscombe, 1993) for their university's men's basketball team and scales assessing personal self-esteem and depression (Branscombe & Wann, 1991). Consistent with expectations, the results revealed a positive correlation between level of identification and self-esteem and a negative correlation between identification and frequency of depression. Branscombe and Wann found that higher levels of identification were positively correlated with the frequency of positive emotions and negatively correlated with the frequency of negative emotions and alienation. Warm (1994) was able to replicate the relationship between identification and self-esteem using collective (i.e., social) self-esteem.

A somewhat more extensive examination of the psychological health of highly identified sport fans was recently published by Wann, Inman, Ensor, Gates, and Caldwell (1999). These authors used the SSIS to classify college students into two groups: those with high levels of identification with their university's men's basketball team and those with low levels of identification with the team. Using the Profile of Mood State (POMS; Grove & Prapavessis, 1992), Wann et al. compared the mood profiles of the two identification groups. As hypothesized, the highly identified groups reported a more positive profile. Specifically, compared to those in the low identification group, the highly identified group reported higher levels of self-esteem and vigor, and lower levels of depression, anger, confusion, tension, and fatigue. However, when the participants were classified into high and low general fandom groups, no group differences in psychological health emerged. In a second study, participants were classified into groups high and low in identification with a distant (but equally popular) team and asked to complete the POMS. There were no differences in psychological health between the two groups. Thus, Wann et al. found that it was neither mere sport fandom nor identification with a distant team that predicted psychological health. Rather, psychological health was only related to a high level of team identification with a local team.

Wann, Dunham, Byrd, and Keenan (2004) conducted another recent study on the psychological well-being of highly identified sport fans. These authors attempted to extend previous research by using the NEO PI-R, an instrument assessing the five-factor model of personality (John, 1990). Previous research (Costa & McCrae, 1992; McCrae & Costa, 1991) had found that several of the personality domains assessed by the NEO PI-R were related to psychological well-being and that the instrument was a valid measure of this construct (i.e., psychological health corresponded with higher levels of extraversion, openness, and conscientiousness, and lower levels of neuroticism). Consequently, Wann and his colleagues expected positive relationships between identification with a local team and extraversion, openness, and conscientiousness while a negative relationship was expected with neuroticism. The results confirmed each hypothesis, with the exception of the expected negative association between identification and neuroticism where no relationship was found. There was no relationship between NEO PI-R domains scores and identification with a distant team.

By combining the aforementioned empirical investigations with the theoretical suggestions offered by Wann and his colleagues, a theoretical model can be developed. This theory, referred to here as the Team Identification--Psychological Health Model, contains three predictions:

1) High levels of identification with a local sport team will lead to positive psychological health.

2) High levels of identification with a distant sport team will not be sufficient to garner the psychological well-being effects.

3) High levels of mere sport fandom will not be sufficient to garner the psychological well-being effects.

The previously described literature strongly supports both Prediction 2 and Prediction 3. Further, tentative support has been found for Prediction 1. That is, the literature to date supports the notion that strong psychological ties to a local sport team are positively related to psychological health. However, this body of evidence is only tentative support for the Prediction 1 because, as noted by Warm and his associates (Wann, 1994; Wann et al., 1999), the research to date has been correlational in nature. Information regarding the causal direction of the relationship is lacking and it is not possible to determine if higher levels of identification cause better psychological health, vice versa, or if the relationship tends to be bidirectional (i.e., circular). This is an important point, and a vital piece missing in the research to date, because the model being proposed here explicitly hypothesizes a causal pattern in which identification with a local team has a direct and positive effect on one's psychological health. While a number of persons have proposed such a pattern of effects (e.g., Melnick, 1993; Smith, 1989; Wann et al., 2004; Warm et al., 2001; Zillmann et al., 1989), to date empirical support for the causality component of the relationship between identification with a local team and wellbeing was lacking.

Using longitudinal research and structural equation modeling, the focus of the current investigation was to examine the potential causal relationship between sport team identification and psychological well-being. Specifically, Prediction 1 of the model was tested using a cross-lagged longitudinal design methodology. It was hypothesized that there would be a significant relationship between identification with a local sport team and psychological health at Time 1 (i.e., the synchronous correlation). This prediction simply reflects the aforementioned research indicating a positive relationship between these variables (i.e., the tentative support for Prediction 1 noted above). Also, the Time 1 to Time 2 autocorrelation paths for both team identification and psychological health were expected to be significant, which simply reflects the consistency found in one's level of identification and psychological wellbeing. The key predictions, .of course, involved the cross-lagged paths from team identification at Time 1 to psychological health at Time 2, and from psychological health at Time 1 to team identification at Time 2. It was hypothesized that the identification to psychological health chronological pattern would be significant while the psychological health to identification path would not.

Method

Time 1

Participants. The initial, Time 1 sample consisted of 214 (90 male; 124 female) university students earning extra course credit in exchange for participation. They had a mean age of 21.64 years (SD = 4.63, range = 18 to 53).

Materials and Procedure. Upon entering the testing room and providing their consent to participate, the participants (tested in small groups) were asked to complete a questionnaire packet containing three sections. The first section contained demographic items assessing age, gender, name, and telephone number. The participants were informed that their name and telephone number were requested because this was a two-part study and some of the participants would be contacted at a later date to, if they so chose, participate in the second phase of the research. They were informed that once the participants had been contacted, the portion of the packet containing identifying information would be removed and destroyed, thus insuring anonymity.

The second section of the questionnaire packet contained the Sport Spectator Identification Scale (SSIS, Wann & Branscombe, 1993). The SSIS contains 7 Likert-scale items with response options ranging from 1 (low identification) to 8 (high identification). The participants targeted their university's men's basketball team when completing the SSIS. Wann and Branscombe report strong internal consistency, test-retest reliability, and criterion validity for the SSIS, and the questionnaire has been successfully used in a number of countries (see Wann et al., 2001). A sample item contained in the SSIS reads "How important to you is it that (name of target team) wins?"

The third section contained four measures designed to assess psychological health. First, participants completed the 10-item Rosenberg (1979) Personal Self-esteem Scale. This highly reliable and valid instrument is a standard measure of an individual's personal self-evaluation. Next, they completed Luhtanen and Crocker's (1991) 16-item Collective Self-esteem Scale. This reliable and valid scale was selected because it is currently the most psychometrically sound assessment of social-level (i.e., group) self-esteem. Third, participants completed the 20-item UCLA Loneliness Scale (Russell, Peplau, & Cutrona, 1980), a reliable and valid measure of loneliness. And finally, respondents completed the 14-item Perceived Stress Scale developed by Cohen, Kamarck, and Mermelstein (1983). This measure is a standard assessment of stress. In addition to being selected based on sound psychometric qualities, these inventories were used because past work had successfully incorporated then into research assessing the mental well-being of sport fans (e.g, Branscombe &Wann, 1991; Warm, 1994; Wann, Dimmock, & Grove, in press). These measures were scored so that higher numbers reflected better psychological health (i.e., higher personal and collective self-esteem, lower levels of loneliness and perceived stress).

After the participants had completed their packet, they returned it to a researcher who handed them a debriefing statement. This statement reminded the participants that they might be contacted at a later date to complete the second phase of the research (it was explicitly noted that any further involvement was strictly voluntary). The statement also contained information on contacting the researcher for a final report of the project. Once the participants had received the debriefing statement, they were excused from the testing session. The sessions lasted approximately 30 minutes.

Time 2

Participants. The Time 2 sample consisted of 45 (10 male; 35 female) university students earning a five-dollar girl certificate to the university's bookstore in exchange for participation (21 percent of the original sample participated in the second phase of the project). They had a mean age of 20.18 years (SD = 3.60, range = 18 to 39).

Materials and Procedure. Approximately three months after the initial testing was complete, the participants were contacted by phone and asked if they would be willing to participate in the second phase of the study. They were informed that the second phase would involve completing a questionnaire packet similar to that which they had completed earlier. Those agreeing to participate were given information on where and when the second phase would occur.

Upon entering the testing room for Phase 2 and providing their consent to participate, participants (again tested in small groups) were asked to complete a questionnaire packet nearly identical to the packet used in Time 1. The only difference between the packets was that the demographic section was omitted from the packet used in Time 2. After the participants had completed their Time 2 packet, they returned it to a researcher who handed them a debriefing statement. This statement described the nature and hypotheses of the study and conmined information on contacting the researcher for a final report of the project. Once the participants had received the debriefing statement, they were excused from the testing session. The sessions again lasted approximately 30 minutes.

Results

Preliminary Analyses

Prior to testing the hypotheses, a series of preliminary analyses were conducted. First, to simplify the analyses, scores on the four assessments of psychological well-being (i.e., personal and collective self-esteem, loneliness, and perceived stress) were transformed into z scores and combined to form a single index of psychological health. Such a procedure was appropriate because each of the four measures of psychological health were intercorrelated, with coefficients ranging from .55 to .81 (all ps < .01).

Next, to insure that the Time 2 sample was representative of the Time 1 sample, a series of one-way analyses of variance (ANOVAs) were conducted to examine scores on the measures of identification and psychological health for those participants who did and did not participate in the second phase. This series of analyses revealed that the two groups did not differ on either team identification, F(1,210) = 0.08,p >.77, or psychological health, F(1,210) = 0.96,p > .32 (Time 1 scores). Thus, it was reasonable to conclude that the original sample of 214 persons was accurately represented by the Time 2 sample.

The final set of preliminary analyses involved testing for gender differences in the measures of identification and psychological health for those participating in the second phase of the research. A series of one-way ANOVAs revealed that males and females did not significantly differ in psychological health at Time 1, F(1, 43) = 0.67,p > .41, or Time 2, F(1,43) = 1.60,p > .21. Further, males and females did not significantly differ in level of identification at Time 1, F(1,43) = 0.04,p > .83, or Time 2, F(1,43) = 0.82,p > .36.

Structural Equation Modeling Analyses

The temporal relationships between team identification and psychological health were tested using structural equation modeling analyzed with Bentler's (1995) EQS program. The correlation matrix for the variables in the equations appears in Table 1. This analysis indicated that the model was highly consistent with the data, [x.sup.2] (1) = .09, p = .77, comparative fit index = 1.00. An examination of the pattern of coefficients shown in Figure 1 (in particular the crosslagged coefficients) indicates strong support for the hypothesized pattern of effects. Specifically, the path from team identification at Time 1 to psychological health at Time 2 was significant (and positive) while the inverse chronological pattern was not.

Discussion

The results presented above, in combination with previous research, provide strong support for each of the three predictions comprising the Team Identification--Psychological Health Model. Specifically, level of identification With a local sport team is positively correlated with psychological well-being (Branscombe &Wann, 1991; Wann, 1994; Wann et al., 1999; Warm et al., 2004), the relationship appears to involve an identification to psychological health causal pattern (the current study), and identification with distant teams and general sport fandom are not related to psychological health (Wann et al., 1999; Wann et al., 2004). However, it does warrant mention that while structural equation modeling and cross-lagged methodologies provide important clues with respect to causal patterns among variables, they do not provide the definitive answers gained through true experimental methodologies. Thus, while the data and analyses reported above provide strong support for Prediction 1 of the Team Identification--Psychological Health Model, additional experimental research is still needed. In all likelihood, this research would involve randomly assigning individuals to various groups, manipulating level of identification with a local sport team, and some time later, assessing and comparing the psychological health of the two groups. Although such an undertaking would be cumbersome, it would be feasible given current information on the antecedents of team identification. For instance, Wann, Tucker, and Schrader (1996) found that a number of factors influenced fans' decisions to originally identify with a particular sport team. Some of these factors, such as the socialization fans received from their parents would be difficult if not impossible to manipulate. The experimental manipulation of other factors, however, such as the importance of geography (i.e., supporting a local team), the desire to support one's own school's team, and importance of group affiliations gained through one's identification with the team, would be plausible. For instance, individuals randomly assigned to the "high identification" condition could be invited to social gatherings with other fans, receive special seating in the arena, and be reminded that the team is a representation of their university. Because there appear to be cultural differences in identification decisions (Jones, 1997; see Wann et al., 2001), one would need to be particularly careful in selecting the factors to be manipulated.

The finding that team identification has direct positive effects on psychological wellbeing is particularly important in light of the fact that many traditional connections to society (e.g., religion and extended family) appear to be declining (Denney, 1979; Guttmann, 1978; Putnam, 1995). Affiliations with others stemming from religious institutions, work-related organizations, and relationships with extended family members have all have shown reduced numbers in recent years. It has been argued that identification with sport teams may serve to replace the traditional but declining social ties as members of society attempt to reestablish and maintain their social connectedness (Melnick, 1993). Consequently, these identifications become more vital to the maintenance of psychological well-being. That is, because social support opportunities are becoming sparse, identifications with local sport teams may become increasingly valuable as "effective venues for social capital formation" (Wann et al., 2001, p. 187).

Although the data detailed above support the notion that high levels of identification with a local sport team can have measurable psychological benefits, this does not imply that such fans are immune from the negative affect fans experience subsequent to a favorite team's defeat. Rather, highly identified fans are particularly likely to report negative emotion (Bernhardt, Dabbs, Fielden, & Lutter, 1998; Wann, Dolan, McGeorge, & Allison, 1994) and depression (Schwarz, Strack, Kommer, & Wagner, 1987; Schweitzer, Zillmann, Weaver, & Luttrell, 1992) upon witnessing a poor performance by their team. Thus, we are left with a bit of a paradox. How can highly identified persons possess a particularly positive mental health profile if these same persons experience such intense negative responses to their team's losses? According to Warm and his colleagues (2001), the solution to this paradox lies in the defense mechanisms highly identified fans develop to help them cope with their team's poor play and, ultimately, return to a positive mental state. Warm et al. noted several such defense mechanisms including derogation of outgroup fans (Branscombe & Wann, 1994), biased attributions that place the blame for the team's failures on outside forces (Hastorf & Cantril, 1954; Wann & Dolan, 1994a; Warm & Schrader, 2000), and biased evaluations of the team's past and future (Dietz-Uhler & Murrell, 1999; Warm & Dolan, 1994b). By employing these and other defense strategies, highly identified fans are able (with time) to handle their team's failures and once again benefit from the social connections stemming from their allegiance to a local team.

One portion of the Team Identification--Psychological Health Model that could use further development concerns the notion of local and distant teams. As mentioned previously, research indicates that significant psychological health benefits are associated with local team attachments but not associations with distant teams (Warm et al., 2004; Warm et al., 1999). However, precise operational definitions of "local" and "distant" teams have yet to be articulated. Should classifying teams as local or distant be a function of the number of miles separating the team from the fan, the amount of media exposure the teams receive, or the number of other supporters for the team found in the general vicinity (to name but a few of the potential operationalizations of these terms)? Precise definitions of the team types are important because they would imply which fans would benefit from their association with the team. For instance, Wann and his associates (2004) have suggested that when highly identified fans of distant teams are in the company of other fans of this team, they may gain temporary increases in their psychological health as a consequence of their identification. Their argument was that when individuals are in the presence of fellow fans of the distant team, these temporary social connections should provide these persons with a valuable link to society at large and, consequently, result in psychological benefits. Such a suggestion seems logical and has support from research indicating that the salience of a social category can have a powerful impact on intergroup behavior (Smith, Spears, & Hamstra, 1999; Spears, Doosje, & Ellemers, 1999; Turner, 1999; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). Perhaps salience impacts the Team Identification--Psychological Health Model as well. That is, it is certainly logical to view local teams as more salient than distant teams--they receive greater levels of media coverage, the fans of the team are more visible, consumption of the team is more convenient (e.g., attending games and buying team paraphernalia), and so forth. However, in certain instances distant teams become quite salient, such as when one is surrounded by other fans of the distant team. Likewise, there may be situations in which the local team is not salient, for instance, when one is alone or during the team's off-season. In the former situation, the "distant" social category of team follower would be salient, even though the team originated some distance from the fan. In the latter instance, the social category would not be salient, even though the team was "local". These situations may reveal patterns that are reversed from past work; for fans of distant teams (because the category of team follower is salient) identification may be positively related to psychological health, but for fans of local but not salient teams, identification might not be associated with well-being. Future work that manipulates salience is needed to determine the impact of this variable on the psychological health benefits reported by highly identified fans.

Although examining potential causality was the logical next step in validating the Team Identification--Psychological Health Model, there are many aspects to the relationship between identification and well-being that remain unclear and untested. For example, the impact of multiple ties to local (or salient) sport teams is yet unknown. If high levels of identification with one local sport team are associated with psychological well-being, would identifying strongly with multiple teams lead to an even better state of mental health? Or, rather, is identification with one team sufficient to garner the maximum benefits to be accrued through sport team identification? The impact of identification with "lesser" sports is also unknown. In the current and past research (e.g., Branscombe & Wann, 1991; Wann, 1994), the target of the identification was a locally popular sport team. Although the Team Identification--Psychological Health Model would predict that identification with lesser teams would also be related to psychological health (as long as societal connections could be obtained through associations with the team), such a prediction is as yet untested. And finally, because most of the past research has examined identification with university basketball teams (e.g., Branscombe & Wann, 1991; Wann et al., 2004), future endeavors should focus on additional sports and levels of competition.

Acknowledgements

This project was partially supported by funds from the Murray State University Provost's Office. The author thanks Brian Keenan for his assistance with the data collection and entry.

References

Bentler, P. M. (1995). EQS structural equations program manual. Encino, CA: Multivariate Software.

Bernhardt, P. C., Dabbs, J. M., Fielden, J. A., & Lutter, C. D. (1998). Testosterone changes during vicarious experiences of winning and losing among fans at sporting events. Physiology and Behavior, 65, 59-62.

Bianco, T., & Eklund, R. C. (2001). Conceptual considerations for social support research in sport and exercise settings: The case of sport injury. Journal of Sport & Exercise Psychology, 23, 85-107.

Branscombe, N. R., & Wann, D. L. (1991). The positive social and self-concept conse-quences of sport team identification. Journal of Sport & Social Issues, 15, 115-127.

Branscombe, N. R., & Wann, D. L. (1994). Collective self-esteem consequences ofoutgroup derogation when a valued social identity is on trial. European Journal of Social Psychology, 24, 641-657.

Brill, A. A. (1929). The why of the fan. North American Review, 228, 429-434.

Cohen, S., Doyle, W. J., Skoner, D. P., Rabin, B. S., & Gwaltney, J. M. Jr. (1997). Social ties and susceptibility to the common cold. Journal of the American Medical Association, 277, 1940-1944.

Cohen, S., Kamarck, D. R., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385-395.

Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310-357.

Costa, P. T. Jr., & McCrae, R. R. (1992). NEO Personality Inventory Revised manual. Odessa, FL: Psychological Assessment Resources.

Crocker, J., & Major, B. (1989). Social stigma and self-esteem: The self-protective properties of stigma. Psychological Review, 96, 608-630.

Curtis, J., Loy, J., & Karnilowicz, W. (1986). A comparison of suicide-dip effects of major sports events and civil holidays. Sociology of Sport Journal, 3, 1-14.

Dietz-Uhler, B., & Murrell, A. (1999). Examining fan reactions to game outcomes: A longitudinal study of social identity. Journal of Sport Behavior, 22, 15-27.

Denney, R. (1979). Feast of strangers: Varieties of social experience in America. In H. J.

Gans, N. Glazer, J. R. Gusfield, & C. Jencks (Eds.), On the making of Americans: Essays in honor of David Riesman (pp. 251-269). Philadelphia, PA: University of Pennsylvania Press.

Eastman, S. T., & Land, A. M. (1997). The best of both worlds: Sports fans find good seats at the bar. Journal of Sport & Social Issues, 21, 156-178.

Grove, J. R., & Prapavessis, H. (1992). Preliminary evidence for the reliability and validity of an abbreviated profile of mood states. International Journal of Sport Psychology, 23, 93-109.

Guttmann, A. (1978). From ritual to record: The nature of modern sports. New York: Columbia University Press.

Hastorf, A. H., & Cantril, H. (1954). They saw a game: A case study. Journal of Abnormal and Social Psychology, 49, 129-134.

Hendrick, S. S., & Hendrick, C. (1992). Liking, loving, and relating (2nd ed.). Pacific Grove, CA: Brooks/Cole.

Hogg, M. A., & Abrams, D. (1990). Social motivation, self-esteem, and social identity. In D. Abrams & M. Hogg (Eds.), Social identity theory: Constructive and critical advances (pp. 28-47). New York: Springer-Verlag.

John, O. P. (1990). The "Big Five" factor taxonomy: Dimensions of personality in the natural language and in questionnaires. In L. Pervin (Ed.), Handbook of personality theory and research (pp. 66-100). New York: Guilford.

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.

Kelley, H. H., Berscheid, E., Christensen, A., Harvey, J. H., Hutson, T. L., Levinger, G., McClintock. E., Peplau, L. A., & Peterson, D. R. (1983). Close relationships. New York: Freeman.

Krane, V., Barber, H., & McClung, L. R. (2002). Social psychological benefits of Gay Games participation: A social identity theory explanation. Journal of Applied Sport Psychology, 14, 27-42.

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.

McCrae, R. R., & Costa, P. T. Jr. (1991). Adding Liebe und arbeit: The full five-factor model and well-being. Personality and Social Psychology Bulletin, 17, 227-232.

Melnick, M. J. (1993). Searching for sociability in the stands: A theory of sports spectating. Journal of Sport Management, 7, 44-60.

Pan, D. W., Gabert, T. E., McGaugh, E. C., & Branvold, S. E. (1997). Factors and differential demographic effects on purchases of season tickets for intercollegiate basketball games. Journal of Sport Behavior, 20, 447-464.

Putnam, R. D. (1995). Bowling alone: America's declining social capital. Journal of Democracy, 6, 65-78.

Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.

Rowe, J. W., & Kahn, R. L. (1998). Successful aging. New York: Dell.

Russell, D., Peplau, L. A., & Cutrona, C. E. (1980). The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39, 472-480.

Schwarz, N., Strack, E, Kommer, D., & Wagner, D. (1987). Soccer, rooms, and the quality of your life: Mood effects on judgments of satisfaction with life in general and with specific domains. European Journal of Social Psychology, 17, 69-79.

Schweitzer, K., Zillmann, D., Weaver, J. B., & Luttrell, E. S. (1992). Perception of threatening events in the emotional aftermath of a televised college football game. Journal of Broadcasting & Electronic Media, 36, 75-82.

Smith, G. J. (1988). The noble sports fan. Journal of Sport & Social Issues, 12, 54-65.

Smith, G. J. (1989). The noble sports redux. Journal of Sport & Social Issues, 13, 121-130.

Smith, H. J., Spears, R., & Hamstra, I. J. (1999). Social identity and the context of relative deprivation. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social Identity (pp. 205229). Malden, MA: Blackwell.

Spears, R., Doosje, B., & Ellemers, N. (1999). Commitment and the context of social perception. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social Identity (pp. 59-83).Malden, MA: Blackwell.

Tajfel, H. (1981). Human groups and social categories. Cambridge, UK: Cambridge University Press.

Thoits, P. A. (1982). Conceptual, methodological, and theoretical problems in studying social support as a buffer against life stress. Journal of Health and Social Behavior. 23, 145-159.

Turner, J. C., Hoog, M., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford, UK: Basil Blackwell.

Warm, D. L. (1994). The "noble" sports fan: The relationships between team identification, self-esteem, and aggression. Perceptual and Motor Skills, 78, 864-866.

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

Warm, D. L., Dimmock, J. A., & Grove, J. R. (in press). Generalizing the Team IdentificationPsychological Health Model to a Different Sport and Culture: The Case of Australian Rules Football. Group Dynamics: Theory, Research, and Practice.

Warm, D. L., & Dolan, T. J. (1994a). Attributions of highly identified sport spectators. The Journal of Social Psychology, 134, 783-792.

Warm, D. L., & Dolan, T. J. (1994b). Influence of spectators' identification on evaluation of the past present and future performance of a sports team. Perceptual and Motor Skills, 78, 547-552.

Warm, 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 & Exercise Psychology, 16, 347-364.

Warm, 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., & Hamlet, M. A. (1994). The Joiners Scale: Validation of a measure of social-complexity. Psychological Reports, 74, 1027-1034.

Warm, D. L., & Hamlet, M. A. (1996). Being a "joiner" and psychological well-being. Psychological Reports, 79, 1186.

Warm, 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, 3, 81-90.

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

Warm, D. L., & Schrader, M. P. (2000). Controllability and stability in the self-serving attributions of sport spectators. Journal of Social Psychology, 140, 160-168.

Warm, 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, 995-1101.

Zillmann, D., Bryant, J., & Sapolsky, B. S. (1989). Enjoyment from sports spectatorship. In J. H. Goldstein (Ed.), Sports, games, and play: Social and psychological viewpoints (2nd ed., pp. 241-278). Hillsdale, NJ: Erlbaum.

Daniel L. Wann

Murray State University

Address Correspondence To: Daniel L. Wann, Department of Psychology, Murray State University, Murray, KY 42071, email:dan.wann@murraystate.edu.

Table 1
Correlation Matrix for the Variables Included in the Structural
Equation Modeling Analyses.

 Time 1 Time 2
 Identification Identification

Time 1 Identification --
Time 2 Identification .77 ** --
Time 1 Well-being .34 * .18
Time 2 Well-being .43 ** .28

 Time 1 Time 2
 Well-being Well-being

Time 1 Identification
Time 2 Identification
Time 1 Well-being --
Time 2 Well-being .85 ** --

Notes: * indicates p <.05, ** indicates p <.01.