Examining reactions to the Dale Earnhardt crash: the importance of identification with NASCAR drivers.

Wann, Daniel L. ; Waddill, Paula J.

Social scientists have been interested in spectators' evaluations of sporting events for many years. Although researchers have investigated evaluations of and reactions to a number of situations, such as reading about one's team in the sports section of a newspaper (Wann & Branscombe, 1992) and the closing of a team's stadium (Trujillo & Krizek, 1994), most examinations have focused on responses to a team's performance. That research indicates that fans tend to be biased in their analyses. For instance, researchers have found that fans' attributions tend to be self-serving as they internalize team successes (e.g., "we won because of our talent") and externalize team failures (e.g., "we lost because of poor officiating," see Wann & Dolan, 1994a; Wann & Schrader, 2000; Warm & Wilson, 2001). In addition, fans are often positively biased in their evaluations of the team's past and future performances, such as recalling more victories from past seasons than actually occurred (Wann, 1994; Wann & Dolan, 1994b). Researchers have found that spectators with a high degree of identification with the team (i.e., fans who feel a strong psychological connection to the team, see Warm & Branscombe, 1993) are particularly likely to report biased evaluations.

However, a limitation of past research on the evaluations and analyses of sport fans concerns the fact that, to date, investigators have focused on reactions to team sport events (e.g., wins and losses). Researchers have yet to sufficiently examine the reactions of fans following individual player sports (e.g., golf, gymnastics, tennis, etc.). Such was the focus of the current investigation. Rather than focusing on the reactions of fans after their favorite player had succeeded or failed, however, we were interested in reactions to a potentially more meaningful event: the crash and death of professional racecar driver Dale Earnhardt, Sr.. Earnhardt died on February 18,2001 after his car was hit by another and slid into the wall on the last turn of the last lap of the Daytona 500.

Because previous research had yet to examine spectator reactions to such events, and because a theory of such reactions was not available, we used the disposition theory of sport spectatorship to guide our predictions. Developed by Zillmann, Bryant, and Sapolsky (1989), disposition theory argues that fans gain enjoyment from witnessing two events: watching their team perform well and watching a rival team perform poorly. Thus, the greatest amount of enjoyment should occur when a favored team defeats a despised rival. Disposition theory argues further that a fan's disposition toward a favorite team and that team's rivals will impact the intensity of affect felt subsequent to a competitive event. Specifically, enjoyment from watching a favorite team perform well should increase with positive sentiments toward the team while enjoyment from watching a rival team lose is expected to increase as sentiments toward that team decrease. Researchers have found strong empirical support for disposition theory (Madrigal, 1995; Sapolsky, 1980; Zillmann et al., 1989; Zillmann & Paulas, 1993). For instance, Wann, Dolan, McGeorge, and Allison (1994) asked college students to report their level of identification (i.e., disposition, see Bryant & Raney, 2000) with their university's men's team prior to witnessing the team win or lose a regular season contest. Subsequent to the games, the participants were asked to complete a questionnaire assessing their affective state. Consistent with disposition theory, highly identified fans (i.e., those with positive sentiments toward the team) reported strong negative affect after watching their team lose and a rival succeed and positive affect after watching their team win and a rival lose.

With respect to evaluations of and reactions to the Earnhardt crash and death, the following hypotheses were generated using disposition theory. First, it was predicted that fans of Earnhardt would be highly sympathetic in their reactions to the crash and that their responses would indicate that the event was quite painful to them. Conversely, we expected racing fans whose favorite driver was someone other than Earnhardt to be rather unsympathetic in their evaluations of the event. Certainly, we did not expect fans of other drivers to feel happy in response to Earnhardt's death. Rather, we were interested in testing the possibility that these persons would trivialize the crash. For instance, we expected fans of other drivers to be more likely to blame Earnhardt for the crash or to indicate that auto racing is simply a dangerous sport and these things happen from time to time. With respect to Earnhardt fans, we expected to find more reactions that indicated emotions like shock and sadness, and more perceptions that the event was a tragedy.

Recall from our earlier discussion that disposition theory predicts that the intensity of a fan's reaction to a sporting event will be influenced by his or her identification with the teams involved. Applying this logic to reactions to the Earnhardt crash, we therefore expected that high levels of identification with Earnhardt or another driver would magnify the effects noted above. Specifically, it was hypothesized that, compared to Earnhardt fans with a low level of identification, those with a high level identification with this driver would be particularly likely to report painful evaluations of the crash and less likely to report unsympathetic responses (i.e., to trivialize the event). An opposite pattern of effects was expected for persons whose favorite driver was someone other than Earnhardt. Among those fans, we predicted that unsympathetic reactions would be especially prominent among highly identified fans while those same fans were expected to report fewer painful reactions.

Method

Participants

The sample contained 234 (84 male; 150 female) college students earning extra course credit in exchange for their participation (most students were enrolled in introductory classes). While this may seem to be a large percentage of female participants (64%), this mirrors the percentages of male and female students at the university in which testing took place. Further, research indicates that approximately 40% of NASCAR television viewers are female and that this number is rising (McGinnis, Chun, & McQuillan, 2003; Wiessman, 1999). The participants had a mean age of 20.75 years (SD = 3.74, range = 18 to 48).

Materials and Procedure

Upon providing their consent, participants (tested in groups) completed a questionnaire packet containing five sections. Section one asked respondents to complete demographic items assessing age and gender. Section two contained the Sport Fandom Questionnaire (SFQ), a reliable and valid measure assessing one's level of interest in sport as a fan (Wann, in press). The SFQ contains five Likert-scale items with anchors ranging from 1 (strongly disagree) to 8 (strongly agree). Thus, higher numbers indicate greater levels of fandom. Although the SFQ was developed to assess general sport fandom, the items were altered slightly for use in the current study to allow for an assessment of auto racing fandom. A sample item from this adapted scale read, "Being an auto racing fan is very important to me." Section three asked participants to provide their open-ended reactions to the fatal crash involving Dale Earnhardt, which occurred February 18,2001 during the Daytona 500 (testing occurred between six and twelve months subsequent to the crash). Specifically, subjects were asked, "What are your thoughts and emotions about the death of racecar driver Date Earnhardt? That is, how do you feel about and how would you analyze his death?" Respondents were also told that there were no right or wrong answers and that they were to be complete in their responses.

The fourth section contained two forced-choice items assessing responses to the crash. First, subjects were asked, "In your view, how tragic was the death of racecar driver Dale Earnhardt?" Responses to this item ranged from 1 (not at all tragic) to 9 (extremely tragic). For the second item, participants read, "In your view, which of the following statements best summarizes your reaction and analysis of Dale Earnhardt's crash and death?" The five choices to this item were:

a) Any way you look at it, Earnhardt's death was tragic. His death is a huge loss for the sport of auto racing and sport in general.

b) Although Earnhardt's death was certainly tragic, as a professional racecar driver, he certainly knows the risks involved. Although unfortunate, accidents and injury are simply a part of this sport.

c) Although Earnhardt's death was tragic, the fact is the crash was his fault. Thus, he is generally responsible for his own death.

d) Individuals who choose to become professional racecar drivers must be "nuts." Why else would someone voluntarily do something so dangerous?

e) None of the above--I have no reaction and/or analysis of the Earnhardt crash. Respondents were to circle which of the aforementioned responses best matched theft feelings.

The fifth and final section of the questionnaire packet contained two versions of the Sport Spectator Identification Scale (SSIS; Wann & Branscombe, 1993). The SSIS contains 7 Likert-scale items with response options ranging from 1 (low) to 8 (high). Thus, higher numbers represented greater levels of identification. The SSIS has been used in a number of studies involving sport fans and has strong reliability and validity (Wann & Branscombe, 1993; Wann, Melnick, Russell, & Pease, 2001). Subjects were to complete only one version of the SSIS. The first version specifically targeted Dale Earnhardt (e.g., "How important was being a fan of Dale Earnhardt to you?"). Subjects read that if Dale Earnhardt was theft favorite professional racecar driver, or if he had been theft favorite driver prior to his death, they were to complete this version of the SSIS. Those participants whose favorite driver was someone other than Earnhardt were to complete the second version of the SSIS. Here, subjects were asked to listed theft favorite driver (open-ended) and to target this person when completing the scale [e.g., "How important is being a fan of your favorite professional racecar driver (the person listed above) to you?"]. Subjects who did not have a favorite racecar driver were instructed to leave both versions blank. After completing the questionnaire packet (20-30 minutes), the participants were debriefed and excused from the testing session.

Results

Preliminary analyses

The seven items comprising the SSIS targeting Earnhardt and the seven item comprising the SSIS targeting a different favorite driver were combined to form two separate indices of driver identification (Cronbach's alpha = .95 and .94, respectively). The five items comprising the SFQ were combined to form a single index of racing fandom (alpha = .93). Based on theft responses to the two versions of the SSIS, participants were classified into three groups: nonfans (i.e., those who left both forms of the SSIS blank, thus indicating that they did not have a favorite driver, n = 155; 41 male, 114 female), (1) Earnhardt fans (i.e., those who completed the SSIS targeting Earnhardt, thus indicating that he was theft favorite racecar driver, n = 36; 23 male, 13 female), and other fans (i.e., those who completed the SSIS targeting a driver other than Earnhardt, thus indicating that they had a different favorite driver, n = 43; 20 male, 23 female).

A comparison of SSIS scores for those whose favorite driver was Earnhardt (M = 30.58, SD = 12.90) and those whose favorite driver was someone else (M = 25.74, SD = 13.93) revealed that the two groups did not differ significantly in their level of identification with the target drivers, F(1,77) = 2.53, p >. 10. Scores on the SFQ were also examined by group. A one-way ANOVA revealed a significant difference in auto racing fandom, F(1,231) = 38.68, p <.001. Post hoc analyses (Newman-Keuls) indicated that persons in the Earnhardt fan (M = 15.33, SD = 9.00) and other driver fan (M = 14.58, SD = 9.13) groups did not differ in the level of racing fandom. However, each of these groups scored significantly higher on the SFQ than persons in the nonfan group (M = 7.21, SD = 4.41).

A final series of preliminary analyses involved using scores on the two versions of the SSIS to construct groups of fans high or low in identification with either Earnhardt or a different driver. Frequencies of SSIS scores for each version were calculated and a median split was then use to develop the groups. With respect to Earnhardt fans, 18 persons were placed into the low identification group (MSSIS score = 19.50, SD = 6.10, range = 7 to 29) and 18 were placed into the high identification group (MSSIS score = 41.67, SD = 6.73, range = 30 to 53). An ANOVA confirmed that the high identification group was indeed higher in driver identification than the low identification group, F(1, 34) = 107.21, p < .001. In addition, a separate ANOVA indicated that the high identification group reported higher levels of general racecar fandom (M = 19.28, SD = 10.45) than those in the low identification group (M = 11.39, SD = 4.95), F(1,34) = 8.38, p < .01. As for participants whose favorite driver was someone other than Earnhardt, 20 subjects were placed into the low identification group (MSSIS score = 13.35, SD = 5.38, range = 7 to 23) and 23 were placed into the high identification group (MSSIS score = 36.48, SD = 8.66, range = 24 to 49). An ANOVA confirmed that the high identification group was indeed higher in driver identification than the low identification group, F(1,41) = 106.63, p <.001. In addition, a separate ANOVA indicated that the high identification group reported higher levels of general racecar fandom (M = 19.70, SD = 9.03) than those in the low identification group (M = 8.70, SD = 4.68), F(1,41) = 24.01, p < .001.

Analysis of open-ended item

The first set of analyses designed to test the predictions generated by disposition theory involved coding the participants' responses to the open-ended item requesting their thoughts and emotions about the death of racecar driver Dale Earnhardt (coders were blind to the participants' driver group). The number of independent coded responses listed by the respondents ranged from a low of one to a high of four. A total of 296 reactions were listed by 155 subjects in the nonfan group (M = 1.91 reactions per participant). The 36 participants in the Earnhardt fan group listed a total of 67 reactions (M = 1.86) while those in the other fan group listed a total of 74 reactions (M = 1.72).

The two independent raters judged and classified the responses, agreeing on approximately 98 percent of the cases. All disagreements were settled through discussion. Nine different categories of responses were established. Four of these categories reflected negative (i.e., painful and sympathetic) reactions to the crash. They were: tragedy (i.e., "it was a tragedy," "it was horrible"), sadness (i.e., "it was a sad thing that happened," "I felt sad"), shock (i.e., "I was in shock," "I refused to believe it"), and loss (i.e., "it was a great loss to racing," "a great athlete died"). Four other categories reflected unsympathetic (i.e., trivializing) reactions to the crash. These reactions were: blame (i.e., "it was his fault," "he should have driven better"), thought of someone else (i.e., "I thought about his family," "I thought about my friends who are fans"), indifference (i.e., "I have no reaction," "none"), and dangerous sport (i.e., "it is a dangerous sport," "he knew the risks"). The final category was an other category containing idiosyncratic and unusual responses that did not fit into the other eight categories.

The percentage of fans in each group listing each type of response was then calculated. These percentages appear in Table 1. For instance, with respect to the tragedy category, 56 of the total (i.e., 296) responses from those in the nonfan group were classified into this category. Thus, 18.9 percent of the responses listed by these persons reflected the belief that the event was tragic. Similarly, 14 of the total responses listed by the Earnhardt fan group were placed into the tragedy category, resulting in a percentage of 20.9. Finally, 8 of the 74 responses listed by fans in the other fan condition were categorized as reflecting the tragedy group. Consequently, this group received a score of 10.8 percent for that category.

Overall, 46.6 percent of the responses offered by subjects in the nonfan group reflected a painful/sympathetic response to the crash (see Table 1). Similarly, 50.1 percent of the responses by persons with a different favorite driver reflected pain. However, for participants in the Earnhardt fan group, 70.2 percent of the responses were in one of the four pain categories. With respect to unsympathetic, trivializing reactions, 45.0 of the responses of nonfans and 43.4 percent of the responses by persons in the other fan group were classified into this category. Only 24.0 percent of the responses by Earnhardt fans reflected a lack of concern for Earnhardt and the crash. As an initial test of the disposition theory-generated hypothesis that fans of other racecar drivers would be relatively unsympathetic about the Earnhardt crash, percentages of unsympathetic responses for the three groups were compared using tests of proportions. The first series of analyses revealed that, as expected, responses of fans of other drivers were indeed less sympathetic than were the responses of Earnhardt fans, z = 2.43, p < .05. Similarly, responses of nonfans were more often categorized as unsympathetic than were the reactions of Earnhardt fans, z = 3.18, p < .01. The lack of sympathy expressed by fans of other drivers and nonfans did not differ significantly, z = 0.25, p > .50. A second series of analyses examined frequencies of painful responses. (2) This analysis indicated that, as expected Earnhardt fans were more sympathetic toward the crash that fans of other drivers, z = 2.42, p < .05. Responses of nonfans were also less likely to be sympathetic than reactions of Earnhardt fans, z = 3.47, p < .001. Similar to the analysis on unsympathetic responses, nonfans and fans of other drivers did not differ in the frequency of reporting painful reactions, z = 0.54, p > .50.

Although the above analyses provided initial support for disposition theory, a more complete test needed to consider the level of identification the Earnhardt and other driver fans felt for their favorite competitors. That is, not only were fans of other drivers expected to be less sympathetic about the crash, we also predicted that those fans with a strong disposition (i.e., degree of identification) toward other drivers would be particularly unsympathetic. Similarly, not only were Earnhardt fans expected to report a particularly high level of painful responses, we predicted that highly identified Earnhardt fans would report the greatest proportion of these reactions. To test these predictions, comparisons of open-ended responses were made between fans with high and low levels of identification with their favorite driver. (3) With respect to persons in the Earnhardt fan group, as expected, persons with a high level of identification with Earnhardt were especially likely to report painful reactions and unlikely to report unsympathetic responses. An examination of Table 2 reveals that the proportions were consistent with expectations. However, tests of proportions also revealed that, although the difference in unsympathetic responses between those high (.121) and low (.354) in identification was significant, z = 2.24, p < .05, the difference in painful responses between highly identified persons (.788) and fans with a low level of identification (.618) was only marginally different, z = 1.55, p = .12.

Similar comparisons were computed for persons with a different favorite driver. An examination of Table 2 reveals that not only was the hypothesis not supported, the pattern of proportions was directly opposite of predictions. That is, rather than reporting a particularly high level of unsympathetic reactions, fans with a high level of identification with a different driver (.304) were actually less likely to report such reactions than persons with low levels of identification (.536), z = 2.00, p < .05. Also contrary to expectations, the highly identified fans of other drivers were more likely to list painful responses (.607) than low identified persons in this group (.414), although this difference was only approached statistical significance, z = 1.65, p =.10.

Analysis of forced-choice items

Tragedy of event. The next series of analyses involved examinations of the two forced-choice items. First, responses to the item assessing perceptions of tragedy were examined using a one-way analysis of variance (ANOVA). This ANOVA revealed a main effect for group, F(2, 231) = 7.37, p < .001. Post hoc tests (Newman-Keuls) revealed that, as expected, persons in the Earnhardt fan group (M = 7.89, SD = 1.04) viewed the event as more tragic than persons in the other driver group (M = 6.58, SD = 2.18). Earnhardt fans also viewed the crash as more tragic than nonfans (M = 6.34, SD = 2.36). The nonfan and other fan groups did not differ significantly in their perceptions of the event as a tragedy.

To test the hypothesis that the perceptions of the crash would be related to one's identification with the drivers, responses to the tragedy item were analyzed by comparing fans in the high and low identification groups. For fans in the Earnhardt group, such a comparison revealed that although highly identified fans (M = 8.06, SD = 0.94) tended to view the event as more tragic than low identified fans (M = 7.72, SD = 1.13), this difference was not statistically significant, F(1, 34) = 0.93, p > .30. For fans whose favorite driver was someone other than Earnhardt, the results again contradicted expectations. Highly identified persons in this group (M = 7.30, SD = 1.69) viewed the crash as significantly more tragic than persons with low levels of identification (M = 5.75, SD = 2.43), F(1,41) = 6.07, p < .05.

Summary of event. The next series of analyses examined participants' responses to the item requesting their summary of the crash. Although participants could choose from one of five reactions (see above), very few subjects (< 8 percent) chose the items that placed the blame on Earnhardt (option c), that said Earnhardt was "nuts" (d), or that reported no reaction (e). Because the number of these "other" reactions was so small (18 out of 234), participants choosing one of these three categories were not included in the analyses. Frequencies of the remaining responses to the summary item were examined using a 2 x 3 chi-square test of independence. This analysis indicated significant frequency differences in the matrix, [chi square](2, N = 216) = 18.10, p < .001. As shown in Table 3 and consistent with expectations, participants in the other driver group were less likely to view Earnhardt's death as "tragic" and a "loss for the sport of auto racing and sport in general" than were those in the Earnhardt fan group. Rather, those in the other fan group tended to believe that Earnhardt "knew the risks" and that "accidents and injury are simply a part of this sport." Responses of those in the nonfan condition were similar to those in the other driver condition.

Finally, responses to the summary item were examined by identification groups using separate 2 x 2 chi-square tests (see Table 4). The analysis on persons in the Earnhardt fan group was marginally significant, [chi square](1, N = 34) = 2.98, p = .08 (with Yates' correction, [chi square] = 1.91, p = .17). An examination of the responses indicates that they were in the predicted direction: highly identified Earnhardt fans were, in comparison to those low in identification, more likely to view Earnhardt's crash and death as "tragic" and a "loss" but less likely to conclude that "he knew risks" and that it was a "part of the sport". As for persons in the other driver group, the chi-square analysis revealed a statistically significant finding that was opposite of expectations, [chi square](1, N = 41) = 4.04, p < .05 (with Yates' correction, [chi square] = 2.74, p <.10). That is, although we had expected highly identified fans of another driver to be unsympathetic in their summary of the crash and death (i.e., to indicate that Earnhardt knew what he was doing and that it is a part of the sport), in fact these persons were more sympathetic than persons with low levels of identification with another driver.

Discussion

The current study examined sport fans' evaluations of the crash and death of professional racecar driver Dale Earnhardt. Using the disposition theory of sport spectatorship (Zillmann et al., 1989) as a framework for the development of specific hypotheses, it was predicted that fans of Earnhardt would be more likely to express painful evaluations (e.g., shock) and less likely to report unsympathetic reactions (e.g., blame). Fans of drivers other than Earnhardt were expected to report the opposite pattern. Further, these findings were hypothesized to be most prominent among persons with high levels of identification with their favorite driver. The results described above presented a fairly clear pattern of effects, some of which were consistent with predictions and some of which were quite unexpected. It warrants mention that the unexpected finding involving identification with rival drivers and evaluations of the crash should not be considered as contrary to disposition theory as outlined by Zillmann and his associates (1989). Disposition theory was not developed to explain non-emotional reactions such as evaluations of non-outcome events. Rather, we simply used disposition theory as a background from which to generate testable hypotheses. Because this was merely a first step in understanding such evaluations, future research is still needed on other events (athlete injury, suspension, retirement, etc.), as is theory development specifically related to events such as those examined here.

One finding that was consistent with predictions involved the tendency for Earnhardt fans to report a great deal of sympathy and pain in their evaluations of the event while fans of other drivers were more likely to trivialize the event. This finding was noted on each of the dependent measures (e.g., persons in the Earnhardt fan group were more likely to list sympathetic evaluations, report that the event was tragic, and summarize the event as a loss) and was found despite the fact that Earnhardt fans and rival driver fans reported similar levels of interest in auto racing and identification with their favorite driver. Responses of persons in the other driver group were quite similar to those in the nonfan group. It is important to note that although these group differences did exist, there was still a substantial amount of sympathy reported by fans of other drivers (particularly among those with high levels of identification, see below). For instance, consider the item assessing perceptions of the degree to which the event was tragic. Although responses to this item did indicate significant group differences in the predicted direction, the mean score for fans of other drivers was well above the midpoint on the scale. This illustrates an important point about the evaluations of those persons analyzing the crash and death of a rival. As noted earlier, we did not expect fans of different drivers to report joy in their evaluations of the crash. Experiencing and reporting a positive reaction to an event such as this would be clearly inappropriate and violate many norms operating within sport and our culture at large. Rather, we expected rival fans to downplay the event (e.g., concluding that auto racing is a dangerous sport and Earnhardt knew the risks involved). An example of this downplaying can be seen in the participants' responses to the tragedy item. Fans of other drivers were much more comfortable concluding that the crash was somewhat tragic than highly tragic. Similarly, on the open-ended item requesting a general analysis, no fans of another driver expressed happiness or delight at the death of Earnhardt, nor were they expected to do so. However, they were comfortable in reporting less sympathetic reactions such as blaming Earnhardt or feeling general indifference.

The hypothesis that Earnhardt fans with a high level of identification would be more likely than low identified fans to report painful evaluations of the crash and less likely to report unsympathetic responses garnered only modest support. Although the patterns of means and proportions were unanimously in the expected direction, more often than not statistical analyses revealed that the difference were only marginally significant, at best. Thus, we are left to conclude that if highly identified Earnhardt fans are more sympathetic in their evaluations, the differences are small. Perhaps an explanation for this finding can be found in previous research on the tendency for fans with varying degrees of identification to bask in the glow of their team's victories (Cialdini et al., 1976). Wann and Branscombe (1990) have found that, among fans of sport teams, even fans with low to moderate levels of identification will jump on the bandwagon and increase their association with the team when they are performing well. Even if the trait of team follower is only peripheral to their overall social-identity, fans with low levels of identification will use the team's successes as an opportunity to boost their self-image. Perhaps somewhat of an opposite effect has been shown here. That is, if only a modest link between the self and a sport entity is needed to experience joy when a team (or player) performs well, it stands to reason that a similarly modest connection is all that is required to evaluate an event such as the Earnhardt crash as tragic and painful.

Perhaps the most intriguing finding, and one that was contrary to expectations, concerns the tendency for highly identified fans of other drivers to be more likely than less identified fans to offer sympathetic evaluations and less likely to express trivializing summaries. In fact, a close inspection of the data reveals a striking similarity in the responses of those high in identification with a different driver and persons with low levels of identification with Earnhardt. For instance, as shown in Table 2, roughly 61.8 percent of the open-ended responses of fans with a high degree of identification with other drivers were classified as sympathetic, a total that is quite similar to the 60.7 percent of such responses reported by low identified Earnhardt fans. Aplausible explanation for this finding is that the highly identified other driver fans were able to empathize with the pain and loss felt by the Earnhardt fans. Because they, too, had a favorite driver, they may have understood how they would have felt had the crash happened to their favorite. Consequently, they were more sympathetic in their evaluations of the crash. Support for these assertions can be found in social psychological research indicating that feelings of empathy toward another often moderate reactions involving pain and suffering (see Batson, 1991, 1995; Batson, Duncan, Ackerman, Buckley, & Birch, 1981). Because this is a post hoc explanation, future work is needed to test its validity. Such research could be completed subsequent to a similar event by assessing level of identification with the athletes in question (i.e., the person involved and that athlete's rivals) as well as levels of empathy for the victim.

References

Batson, C. D. (1991). The altruistic question." Toward a social-psychological answer. Hillsdale, NJ: Erlbaum.

Batson, C. D. (1995). Prosocial motivation: Why do we help others? In A. Tesser (Ed.), Advanced social psychology (pp. 333-382). New York: McGraw-Hill.

Batson, C. D., Duncan, B., Ackerman, P., Buckley, T., & Birch, K. (1981). Is empathetic emotion a source of altruistic motivation? Journal of Personality and Social Psychology, 40, 290-302.

Bryant, J., & Raney, A. A. (2000). Sport on the screen. In D. Zillmann & P. Vorderer (Eds.), Media entertainment: The psychology of its appeal (pp. 153-174). New York: Erlbaum.

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.

Madrigal, R. (1995). Cognitive and affective determinants of fan satisfaction with sporting event attendance. Journal of Leisure Research, 27, 205-227.

MvGinnis, L., Chun, S., & McQuillan, J. (2003). A review of gendered consumption in sport and leisure. Academy of Marketing Science Review, 03 (5). Retrieved January 5, 2006, from http://www.amsreview.org/articles/mcginnis05-2003.pdf

Sapolsky, B. S. (1980). The effect of spectator disposition and suspense on the enjoyment of sport contests. International Journal of Sport Psychology, 11, 1-10.

Trujillo, N., & Krizek, B. (1994). Emotionality in the stands and on the field: Expressing self through baseball. Journal of Sport & Social Issues, 18, 303-325.

Wann, D. L. (in press). Preliminary validation of a measure for assessing identification as a sport fan: The Sport Fandom Questionnaire. International Journal of Sport Management.

Warm, D. L. (1994). Biased evaluations of highly identified sport spectators: A response to Hirt and Ryalls. Perceptual and Motor Skills, 79, 105-106.

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

Wann, D. L., & Branscombe, N. R. (1992). Emotional responses to the sports page. Journal of Sport & Social Issues, 16, 49-64.

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

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

Wann, D. L., & Dolan, T. J. (1994b). The influence of spectator identification on the evaluation of the past, present, and future performance of a sports team. Perceptual and Motor Skills, 78, 547-552.

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

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

Wann, D. L., & Wilson, A. M. (2001). The relationship between the sport team identification of basketball spectators and the number of attributions generated to explain a team's performance. International Sports Journal, 5, 43-50.

Weissman, R. X. (1999). The green flag is up: High-tech companies are burning rubber to reach NASCAR fans. American Demographics, 21, 33-36.

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, N J: Erlbaum.

Zillmann, D., & Paulas, P. B. (1993). Spectators: Reactions to sports events and effects on athletic performance. In R. N. Singer, M. Murphey, & L. K. Tennant (Eds.), Handbook of research on sport psychology (pp. 600-619). New York: Macmillan.

Daniel L. Wann and Paula J. Waddill

Murray State University

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

(1) Although reactions of persons in the nonfan group were not involved in the hypotheses, they are included in the basic analyses for comparison purposes.

(2) Although the analyses of unsympathetic and painful responses may seem redundant, the inclusion of an "other" category in the classification scheme rendered separate analyses of proportions necessary (i.e., the sum of the sympathetic and painful responses did equal 100 percent due to the "other" responses).

(3) High and low identified fans in both groups reported a similar number of total reactions. Specifically, among persons in the Earnhardt group, highly identified fans listed a total of 33 responses while low identified fans listed 34. As for persons in the other driver group, highly identified fans listed 33 responses while low identified fans listed 41.

Table 1
Percent of Responses Listed for Each Category by Group.

 Earnhardt Other
Category Nonfans Fans Driver Fans

Painful reactions
 Tragedy 18.9 20.9 10.8
 Sadness 21.3 22.4 28.4
 Shock 2.0 16.4 4.1
 Loss 4.4 10.5 6.8
 Total painful 46.6 70.2 50.1

Unsympathetic reactions
 Blame 5.1 6.0 6.8
 Thoughts 13.9 9.0 9.5
 Indifference 11.8 1.5 9.5
 Dangerous sport 14.2 7.5 17.6
 Total unsympathetic 45.0 24.0 43.4
Other 8.5 6.0 6.8

Note: Percentages may not sum to 100 due to rounding.

Table 2
Percent of Responses Listed for Each Category by Persons with High
and Low Levels of Identification with Their Favorite Driver.

 Earnhardt Fans Other Driver Fans

Category High ID Low ID High ID Low ID

Painful reactions
 Tragedy 21.2 20.6 6.1 14.6
 Sadness 24.2 20.6 36.4 22.0
 Shock 18.2 14.7 6.1 2.4
 Loss 15.2 5.9 12.1 2.4
 Total painful 78.8 61.8 60.7 41.4
Unsympathetic reactions
 Blame 6.1 5.9 6.1 7.3
 Thoughts 3.0 14.7 3.0 14.6
 Indifference 0.0 3.0 6.1 12.2
 Dangerous sport 3.0 11.8 15.2 19.5
 Total unsympathetic 12.1 35.4 30.4 53.6
 Other 9.1 2.9 9.1 4.9

Note: Percentages may not sum to 100 due to rounding.

Table 3
Frequencies (and Percentages) of Responses to Summary Item by Group

 Group

 Earnhardt Other
Reaction Nonfans Fans Driver Fans

Death was "tragic" 28 19 11
 and "loss" (19.9) (55.9) (26.8)
"He knew risks" and 113 15 30
 "part of sport" (80.1) (44.1) (73.2)

Note: Percentages (appearing in parentheses)
may not sum to 100 due to rounding.

Table 4
Frequencies (and Percentages) of Responses to Summary Item by Group

 Earnhardt Fans Other Driver Fans

Reaction High ID Low ID High ID Low ID

Death was "tragic" 12 7 9 2
 and "loss" (70.5) (41.2) (39.1) (11.1)
"He knew risks" and 5 10 14 16
 "part of sport" (29.4) (58.8) (60.9) (88.9)

Note: Percentages (appearing in parentheses)
may not sum to 100 due to rounding.