Mood Effects of Televised Sports Events

Abstract

Television viewers attend to sports programs primarily to gain emotional rewards. As not only wins but also defeats are inherently rooted in sport competitions, television viewers can be positively as well as negatively affected in their feelings when watching sports on television. Interestingly, some studies were able to show that the feelings evoked by watching sport television also influence viewers’ judgments, following feeling-as-information theory. The present study builds on these results by investigating the mood effect of viewing televised football Fédération Internationale de Football Association World Cup games on personal as well as economic estimations of viewers. A quasi-experimental design was employed, assessing the moods and estimations of viewers before and after a win and a defeat of the German national team. The results support feeling-as-information theory, as viewers reported enhanced mood and estimations after watching the victory. Results of previous studies are extended, as longer term effects are included and the mediating role of mood was explicitly tested and supported.

Keywords

football soccer mood-as-information television media effects judgments

Undeniably, sports represent one of the most important television contents today. Whenever you turn on television, you are likely to encounter sports on one of the channels available (Brown & Bryant, 2006). Within televised sport, mega events such as the Super Bowl and the Fédération Internationale de Football Association (FIFA) World Cup hold a special position. Being broadcast live and worldwide, there is little other media content that attracts as many viewers simultaneously. For example, the 2012 Super Bowl appeared to be the most watched television show in U.S. history, with an average audience of 111 million viewers in the United States alone (Seidman, 2012). In Europe, the FIFA World Cup is a similar case. Indeed, the semifinal of the 2010 FIFA World Cup attracted more than 31 million television viewers in Germany alone, representing the largest television audience in history in that country (Gerhard, Kessler, & Gscheidle, 2010). These figures clearly indicate the importance of such media events in people’s lives, but also point to the potential of televised sports events in pervasively affecting vast audiences, including entire nations.

Past research has shown that viewers’ emotions and moods constitute an area particularly susceptible to the influence of televised sports. Since most spectators primarily attend to sports television in order to gain emotional rewards in terms of enjoyment, sports programs are highly likely to affect viewers’ feelings throughout and after the perception (Raney, 2006). While this factor alone indicates that the topic is worthy of research, especially considering that large proportions of a national population may be influenced, it seems even more worthy if we consider the pervasive influence of feelings on issues unrelated to the sporting event being televised. According to feelings-as-information theory (Schwarz, 2011), people often draw on current mood states when making decisions and judgments. Hence, previous research has shown that televised sports events might not only change viewers’ feelings but also subsequent judgments with no apparent direct connection to the sports event (e.g., Schwarz, Strack, Kommer, & Wagner, 1987). The present article takes up this point by explicitly modeling and testing the indirect influence of sport outcomes on viewers’ perception of their personal situation as well as the current economic situation via mood states as the underlying mechanism. Of particular importance are influences on one’s estimation of the current economic situation, as a country’s economic strength can depend on the economic decisions of individuals and stock market traders in particular, especially during times of drastic change such as the ongoing financial crisis. As such decisions are often accompanied by a large degree of uncertainty in recent times (Bloom, 2009), influences appear to be highly likely.

Mood and Judgments—Feelings as Information

Generally, there are two typical cases in which people draw on affective experiences when making judgments. Obviously, people refer to their feelings if a particular affective reaction was originally directed toward the object being judged. In this case, the referent of a person’s feelings equals the object being assessed. For instance, if you feel a liking for somebody, this feeling will naturally guide your judgment toward the person. According to the feeling-as-information approach, people also draw on their feelings if an evaluative task is especially complex as well as demanding. Feelings then function as a kind of simplification or mental shortcut. Instead of investing a lot of effort and time to solve the task, people often refer to their temporary feelings by asking themselves how they feel about the object being judged (Schwarz & Clore, 1988). Since current feelings may be evoked by various causes that may be difficult to distinguish, people may also be influenced by feelings that were not originally evoked by the object being assessed:

Because we have only one window on our experience, it is difficult to distinguish integral feelings, elicited by the target, from incidental feelings that happen to be present at the time. Hence, we may mistake incidental feelings, like a preexisting mood, as part of our reaction to the target. This results in judgments that are congruent with the implications of our feelings (Schwarz & Clore, 2007, p. 386),

even if the feelings were not originally directed toward the judged object. It is important to note that people do not have to attribute their current feelings consciously to the target during a judging process. Instead, the process can be better understood as an automatic one; that is, people usually perceive their feelings as being about whatever is currently in the focus of their attention and thereby draw on them as the automatic default option (Schwarz, 2011). Following this reasoning, feelings evoked by viewing sports and by the outcome of the contest are highly likely to influence various judgments.

Looking at different kinds of feelings, this process is particularly likely to occur if people refer to temporary moods rather than specific emotions. While moods and emotions are both affective experiences, with both indicating a positive or negative valence of something, they differ in terms of their intensity, duration, and referent. Emotions are understood as responses to ongoing and implicit appraisals of a situation in terms of positive or negative consequences of one’s goals, and have an identifiable referent indicating the cause of the emotion (Schwarz & Clore, 2007). In contrast, moods lack a specific referent, meaning that responses influenced by moods are often unrelated to the mood-precipitating event. Compared with emotions, moods can influence a greater breadth of responses. In fact, researchers agree that precisely this lack of specificity, as well as primarily transmitting valence information, explains the strong influence of moods on various cognitive, affective, and behavioral responses (Morris, 1989). Besides lacking a clear referent, moods are also assumed to be less intense and to last longer than emotions (Schwarz & Clore, 2007). Although emotions have a clear referent and are therefore unlikely to function as affective experiences in feeling-as-information processes, they might exert influence indirectly. Past research has shown that the experience of positive or negative emotions can lead to equally valenced moods (Schwarz & Clore, 2007).

People do not generally rely on their mood as a heuristic when judging a target. As stated above, they primarily refer to their mood if a complex and time-consuming judging task is at hand during a period when they lack processing capacity and/or motivation (Schwarz, 2011). In addition, they appear to be only drawing from affective experiences if they perceive their current feelings as valuable information sources. That is, the perceived value is not questioned by attributing the feelings to some other currently irrelevant cause (Schwarz & Clore, 1983). Schwarz and Clore (1983), for instance, showed that people reported better moods and evaluations of their lives on days with good weather. This positive effect on life evaluations vanished when interviewers directed participants’ attention to the weather and thereby reminded them of the actual cause of their feelings. Next to perceived information value, the influence of feelings also depends on their perceived relevance. According to Raghunathan and Pham (1999), people are more influenced by their feelings when they make decisions concerning their own preferences than the preferences of other. In the former case, feelings are considered more relevant by decision makers as their own feelings are more strongly related to their own preferences compared to the preferences of other. Finally, people are especially likely to draw on their feelings if they lack concrete, task-related information and if no other relevant information sources are available (Clore, Schwarz, & Conway, 1994).

Sports Television and Viewers’ Affective Experiences

As mentioned above, recipients of televised sports often turn on the television in the first place to gain emotional rewards (Raney, 2006). Specifically, the suspenseful action resulting from the presented fight for victory of opposing athletes or teams motivates people to tune in and carry on watching (Peterson & Raney, 2008). Thus, as wins but also defeats are inherently rooted in the competitive nature of sports events, roughly one half of the spectators should be positively affected by the events’ outcome while the other half should be negatively affected (Wann, 2006). It should be noted that feelings elicited by televised sports do not differ fundamentally from feelings elicited by non-media causes (Scherer, 1998). According to Raney (2009), such feelings can best be understood by applying the disposition theory of sports spectatorship (Zillmann, Bryant, & Sapolsky, 1989). Zillmann and colleagues (1989) suggest understanding defeats as a loss of value and wins as a gain of value. Depending on whether (1) loss of value is inflicted upon “enemies” and gain of value is obtained by “friends” or (2) gain of value is obtained by “enemies” and loss of value is inflicted upon “friends,” viewers respond in terms of appreciation or sorrow, respectively. Thus, feelings derived from watching sports depend in large part on the positive or negative dispositions a person holds toward the team or athlete, and on the respective team’s or athlete’s performance in terms of winning or losing.

Ultimately, the enjoyment of viewing a sporting event is a function of the outcome of the game in relation to the strength and valence of the disposition held toward the competitors. Specifically, enjoyment is thought to increase the more the winning team is favored by the viewer and/or the more the losing team is disliked by the viewer. Conversely, enjoyment is thought to decrease the more the winning team is disliked by the viewer and/or the more the losing team is favored by the viewer. (Raney, 2006, p. 316)

Hence, viewing televised sports will leave the spectator in the most positive mood if the strongly favored team or athlete wins. If the contrary is the case, the least enjoyment (i.e., greatest disappointment) is felt by the audience, and viewers are left in the most negative mood. If spectators are unrelated to or indifferent toward a team or athlete, flat or no affective reactions are assumed to occur (Zillmann et al., 1989). Looking at factors influencing affective dispositions toward teams or athletes, geographical closeness as well as viewing a team or athlete with the same nationality as one’s own generally result in positive affective dispositions. In addition, habitual fandom often further fosters these dispositions (Knobloch-Westerwick, David, Eastin, Tamborini, & Greenwood, 2009). Zillmann and colleagues (1989) presented much empirical support for their theory in their initial article, and the theory holds true today, having been supported by subsequent studies (Raney, 2009).

Effects of Sport-Induced Mood on Viewers’ Judgments

So far, we have presented theoretical and empirical evidence of mood induction being possible through watching sports on television as well as mood influences being possible on people’s judgments. The next section will combine these two elements by reviewing existing research on the influence of sport-induced mood on spectators’ judgments. While many studies have investigated spectators’ feelings during sport competitions, few have considered affective experiences in the postgame period and even fewer have examined the consequences of these experiences (Schweitzer, Zillmann, Weaver, & Luttrell, 1992). One study dealing with subsequent consequences is a quasi-experiment by Schweitzer and colleagues (1992), who investigated differences in mood and estimations of the likelihood of war (in the period leading up to the Gulf War) between fans of two college American football teams after watching them playing against each other on television. Results showed that fans of the winning team reported better mood and estimated war to be less likely, compared with fans of the defeated team. It should be noted that the latter effect on war likelihood was primarily caused by the fans of the defeated team rating a war more likely than a control group, whereas the fans of the winning team showed no difference compared with the control group. It is possible that future prospects are rather affected in their estimation if an equally valenced mood is present, such as the fear of a military confrontation and a negative mood (Schweitzer et al., 1992). Although the authors found differences in mood and spectators’ evaluations in line with the outcome of the game, the study did not test whether the differences in evaluations were actually caused (i.e., mediated) by the spectators’ mood. As observed by Hirt and colleagues (1992), “many of the studies purported to support such an assumption do not explicitly test whether the obtained changes in mood actually mediate the observed changes in outcome variables” (p. 736). Recognizing this deficit, they tested the mediating role of mood on performance evaluations of spectators after watching a televised win or defeat of a favored basketball team. The results showed that the game’s outcome affected viewers’ mood in line with predictions of disposition theory, with highly identifying viewers being affected more strongly. Contrary to the assumptions of feeling-as-information theory, the affected mood did not influence estimations concerning future team or personal performance. Instead, different levels of postgame self-esteem mediated the influence on team and personal evaluations. The authors concluded that their performance measures were possibly primarily sensitive to self-esteem while being insensitive to mood, thereby acknowledging the possibility of mood effects on other estimations. In any case, future research is required to explicitly test the mediating role of mood (Hirt, Zillmann, Erickson, & Kennedy, 1992). Although not explicitly testing mood-as-information effects on personal evaluations, Bryant and Cummins (2009) were likewise able to show higher or lower levels of self-confidence of highly identifying American football fans after either watching a win or a loss of their favored team in line with previous results.

The moderating role of team identification was supported by a study by Wann and colleagues (1994). Highly identifying sport spectators reported an increase in positive emotions after watching a win, whereas low identifying viewers showed almost no emotional change. The same pattern was found after watching a defeat, whereas in this case the negative emotions increased for highly identifying viewers. As highly identifying sports fans view their team as being part of their social identity, “we know that for sports fans, a team loss is taken as a personal defeat, and a team win is taken as a personal victory” (Bryant & Cummins, 2009, p. 233) resulting in respective mood.

The above studies considered U.S. participants and U.S. sports, whereas Schwarz and colleagues (1987) investigated mood effects and subsequent consequences in German sport viewers during the 1982 FIFA World Cup. In this study, German residents were telephoned immediately before or after two games involving the German national team, who were victorious in one of the games and drew the other. The results showed a significant interaction between pregame and postgame measurement and game outcome; that is, viewers who watched the victory reported higher postgame ratings of global well-being than the participants interviewed prior to the game, whereas those who watched the drawn match reported lower ratings compared with those interviewed prior to the game. No such effects were found regarding satisfaction with work and income, or with national issues. The authors attributed this lack of influence to the fact that work and income, as well as national issues, represent more specific life domains compared with evaluating one’s life as a whole (global well-being). According to the judgment model of subjective well-being (Schwarz & Strack, 1985), the evaluation of specific life domains is less affected by temporary mood states because mood might not be considered informative and other information sources are more readily at hand. Again, the authors assumed influences on viewers’ mood to cause the effects of game outcome on assessing one’s global well-being, but did not actually test the mediating role of mood. Furthermore, the above studies researched mood effects measured immediately after the respective match. Hence, no previous study has assessed the longer term effects of mood, despite the fact that moods are assumed to be relatively long-lasting affective experiences (Schwarz & Clore, 2007).

Looking at different kinds of consequences, all previous studies that investigated feeling-as-information processes resulting from sports spectatorship reported effects on some personal estimation. Few studies have considered influences on impersonal matters such as national issues or the likelihood of war (e.g., Schwarz et al., 1987; Schweitzer et al., 1992), though these matters are no less important. As stated in the introduction, judging a country’s current economic situation may represent such a crucial matter. To date, studies investigating sport-induced mood effects on economic decisions have mainly been grounded in economics, meaning that such studies are typically conducted on the macro level. For instance, Edmans, García, and Norli (2007) showed that losses by national football teams were followed by a strong negative reaction on the stock market the following day. The authors assumed that changes in the investors’ mood caused this reaction. Similarly, Chang and colleagues (2012) reported that losses by local American football teams had a negative influence on the next-day stock returns of firms located close to the home city of the defeated team. Since a large holding of a firm’s stock is supposed to be held by local investors, changes in mood resulting from local American football losses were assumed to affect local trading decisions. While these studies support the hypothesis that feeling-as-information processes influence economic decisions on a macro level, the processes have yet to be verified on the individual level, since neither changes in mood nor their mediating role have been tested.

Hypotheses

We seek to build on previous results, and address the research deficits outlined above, by including longer term effects, by adding estimations of economic issues to the outcome variables, and by explicitly testing the mediating role of mood. In addition, we orientate our research to the study by Schwarz and colleagues (1987), as their study was conducted in the context of a FIFA World Cup and their design worked well overall. Summing up, it is proposed that according to the disposition theory of sports spectatorship (Zillmann, 1989), mood derived from watching a sporting competition depends on the victory or defeat of the favored team.

Hypothesis 1a: Television viewers of the FIFA World Cup show more positive moods after having watched their favored team win, compared with before the match.

Hypothesis 1b: In contrast, television viewers of the FIFA World Cup show more negative moods after having watched their favored team lose, compared with before the match.

Following feeling-as-information theory (Schwarz, 2011; Schwarz & Clore, 1983), changes in mood derived from watching sporting competitions affect viewers’ estimations. This is especially true if the estimations are demanding and time-consuming, and if the feelings are perceived as valuable and relevant. These conditions are met by evaluations of one’s current level of self-confidence and by the economic situation of a country. However, one’s mood should be more relevant concerning the former than the latter (Raghunathan & Pham, 1999), meaning that effects are assumed to be stronger regarding self-confidence.

Hypothesis 2a: Estimations of television viewers concerning their self-confidence are indirectly affected by the outcomes of games at the FIFA World Cup via viewers’ mood; that is, more positive moods after wins will lead to enhanced self-confidence, whereas more negative moods after defeats will lead to decreased self-confidence compared with before the match.

Hypothesis 2b: Estimations of television viewers concerning the economic situation of their home country are indirectly affected by the outcomes of games at the FIFA World Cup via viewers’ mood; that is, more positive moods after wins will lead to enhanced economic evaluations, whereas more negative moods after defeats will lead to decreased economic evaluations compared with before the match.

Hypothesis 2c: Estimations of self-confidence are more strongly influenced by current moods than are economic evaluations.

The disposition theory of sports spectatorship predicts that the postmatch mood is influenced by not only the valence but also the strength of the relationship that viewers hold toward their favored team. Various studies have supported this moderating role of team identification (Hirt et al., 1992; Wann, Dolan, McGeorge, & Allison, 1994).

Hypothesis 3: The mediating role of mood described in Hypothesis 2 is moderated by viewers’ team identification; that is, highly identifying viewers will show the assumed affective reactions and subsequent influence on their estimations, while low or lower identifying viewers will show weaker or no affective reactions and subsequent influence on their estimations.

Method

Study Design

To test our hypotheses, a quasi-experimental design was applied following the study by Schwarz and colleagues (1987). A win and a defeat by the German national team being live broadcasted and live watched by the participants during the 2006 FIFA World Cup in Germany were used as experimental manipulations. Specifically, the win was represented by the quarterfinal (June 30, 2006) which Germany won 4–2 over Argentina on penalties, while the defeat was represented by the semifinal (July 4, 2006), which Germany lost 0–2 to Italy after extra time. This particular World Cup was chosen because many German citizens identified with the competition and had responded to it in advance with a burst of enthusiasm, rendering affective experiences likely (Geese, Zeughardt, & Gerhard, 2006). Participants were asked to take part in an online questionnaire. Data were collected online because this makes it relatively easy to obtain sufficiently large subsamples for the four experimental conditions. In addition, answering questions at home in front of the computer ensured full anonymity for the participants, which was a necessary requirement considering the personal nature of the questions (Birnbaum, 2004). Since participants did not have to watch the games in a laboratory setting, but watched the games voluntarily from home, the research setting was most realistic. Participants took part in the survey either within 2 days before or 2 days after one of the games. Looking at the 2-day intervals, people took part in the survey with roughly the same frequency on each of the 2 days of the respective intervals. Thus, measuring mood and estimations within a range of 2 days before or after the games enabled longer term propositions.

Participants

A total of 328 Germans took part in the study (40 before the win, 79 after the win, 116 before the defeat, and 93 after the defeat) ensuring positive affective dispositions toward the German national team due to same nationality (Knobloch-Westerwick et al., 2009). The average age was 29.03 years (SD = 12.48). In contrast to Schwarz and colleagues (1987), both male (56%) and female (44%) television viewers participated, enabling more general assertions to be made. Looking at education, participants were rather higher educated as 56% had a high school diploma and additional 25% had a university degree. All participants completing the survey after each game indicated that they had actually watched the game. As an inducement to take part in the survey, participants had the chance of winning a jersey of the national team.

Measurement

Mood was measured using 2 items. While the first referred to mood at the present time, resembling the mood measurement by Hirt and colleagues (1992), the second item asked for mood during the past 1 or 2 days, accounting for mood as a longer term affective experience. The items were On the whole I presently feel being in a good mood and I personally was in a very good mood during the past 1–2 days. Both items were measured on the same 5-point scale ranging from This is in no way true to This is absolutely true. Since the measurement consisted solely of 2 items, split-half reliability was used to assess reliability rather than Cronbach’s α (Hulin, 2001). The Spearman–Brown coefficient yielded a satisfying value of .75 (M = 3.80, SD = .97).

Self-confidence was measured by 2 items on the established 5-point scale. The items were I feel very confident about my personal future and I currently feel highly self-confident. Reliability was satisfactory, with a Spearman–Brown coefficient of .72 (M = 3.57, SD = .92).

Evaluation of the current economic situation of one’s home country was measured with a single item reading I assess the economic situation of our country as positive. Again, participants indicated their approval on the established 5-point scale.

Team identification was measured using 2 items from a study by Schramm and Klimmt (2003), which successfully assessed the identification of supporters of the German national team during the 2002 FIFA World Cup. The established 5-point scale was applied. The items were I watch the games because I am proud when my favorite team wins and I watch the games to keep my fingers crossed for my favorite team. Reliability was satisfactory, with a Spearman–Brown coefficient of .78 (M = 4.26, SD = 1.07).

Results

Hypothesis 1 predicted a significant interaction assuming only increased mood after having watched a football game won by the favored team, while the opposite was predicted after watching a defeat. Following the analysis by Schwarz and colleagues (1987), a two-way analysis of variance (ANOVA) was calculated. The calculation included the time of measurement (before or after the game) and the game outcome (win or loss) as independent variables, and the viewers’ mood as the dependent variable. The ANOVA resulted in the predicted significant interaction of measurement time and game outcome, F(1, 324) = 10.26, p < .01, partial η² = .03; see Figure 1. Television viewers who watched the quarterfinal victory reported enhanced mood (M = 3.91, SD = .95) compared with before the game (M = 3.33, SD = .94), while those who watched the semifinal loss reported decreased mood (M = 3.75, SD = 1.04) compared with before the game (M = 3.92, SD = .97).¹ Post hoc t-tests revealed the enhanced mood after the win to be highly significant, t(117) = –3.17, p < .01, whereas the decreased mood after the defeat was nonsignificant, t(207) = 1.23, ns. Summing up, Hypothesis 1a was supported whereas Hypothesis 1b was only supported by a statistically insignificant trend in the data.

Figure 1.

Effect of game outcome on viewers’ mood.

Hypothesis 2 assumed that estimations of television viewers concerning their self-confidence (Hypothesis 2a), as well as their country’s economic situation (Hypothesis 2b), are indirectly affected by the outcomes of games at the FIFA World Cup. Specifically, viewers’ mood, affected by game outcomes (Hypothesis 1), was predicted to mediate the relationship between game outcomes and viewers’ estimations. The mediation was tested using bootstrapping analyses with bias-corrected confidence estimates. The analyses followed the methods described by Preacher and Hayes (2004), using their Statistical Package for the Social Sciences (SPSS) macro. Bootstrapping involves computing a confidence interval (CI) around the assumed indirect effect, in our case 95 or 99% confidence. If the value of zero lies outside the interval, the indirect effect is statistically significant at p < .05 or p < .01 (two-tailed). The analyses and bootstrap estimates that follow are based on 5,000 bootstrap samples. As shown in Figure 2, watching a victory by the German national team resulted in a highly significant enhancement of the viewers’ mood, t(117) = –3.17, p < .01; see also Hypothesis 1a. In turn, viewers’ mood was a significant predictor of viewers’ estimations of self-confidence, t(117) = 11.03, p < .001, and of the economic situation, t(117) = 2.16, p < .05. In line with this, bootstrapping analyses revealed a highly significant and a significant indirect effect regarding the two types of estimation, since CIs were entirely above zero in both cases (self-confidence: 99% CI [.09, .86]; economic situation: 95% CI [.01, .34]). Calculating the total indirect effects by multiplying the point estimations of the respective a and b paths presented in Figure 2, we obtained a point estimation of .46 for the indirect effect on self-confidence and a point estimation of .14 for the indirect effect on economic situation. In other words, watching a victory by the German national team resulted in an indirect increase in viewers’ self-confidence by half a point on the 5-point scale, via enhanced viewers’ mood that in turn resulted in increased self-confidence. Following the same pattern, watching a victory also indirectly enhanced viewers’ estimation of the economic situation of their home country by about .14 points on a 5-point scale. The total and direct effects of a victory on both estimations were nonsignificant, thereby emphasizing the indirect effect.² Looking at watching a loss, no such indirect effects were found. As already revealed in the analysis of Hypothesis 1b, watching a loss did not significantly decrease viewers’ mood, t(207) = 1.23, ns; see Figure 3. Hence, there were no indirect effects of watching a defeat on viewers’ estimations, although again, viewers’ mood had a highly significant influence on viewers’ estimations, self-confidence: .62, t(207) = 13.47, p < .001; economic situation: .47, t(207) = 6.38, p < .001. Summing up, Hypotheses 2a and 2b were supported regarding watching a win. In addition, Hypothesis 2c was supported because the indirect effect on self-confidence was about 3 times greater than the indirect effect on economic situation.

Figure 2.

Indirect effect of watching a win on viewers’ estimations via viewers’ mood.

Figure 3.

Indirect effect of watching a defeat on viewers’ estimations via viewers’ mood.

Hypothesis 3 predicted a moderated mediation (Hayes, 2013). Specifically, the mediating role of mood described in Hypothesis 2 is assumed to be moderated by viewers’ team identification; that is, highly identifying viewers will show the assumed affective reactions and subsequent influence on estimations, whereas lower identifying viewers will show weaker effects or no effect. Since mediation could only be found in the case of watching a win, the following analysis concentrates on this situation. Moderated mediation was tested by applying bootstrapping analyses using the methods described by Hayes (2012), as well as his SPSS macro “Process.” The macro offers the advantage of an integrated testing of the moderated mediation. Specifically, indirect effects of watching a win on viewers’ estimations, as well as their significance, were calculated at different levels of the assumed moderator. Following Berndt and colleagues (2012), the mean of team identification as well as a standard deviation below and above the mean were used to represent moderate, low, and high levels of team identification. Bootstrapping analysis of 5,000 samples revealed a significant indirect effect of watching a win on viewers’ self-confidence via viewers’ mood for highly and moderately identifying viewers (see Table 1). The 95% CIs were entirely above zero in both cases ([.17, .95] and [.15, 75]), while the indirect effects came to a point estimation of .54 (highly identifying) and .43 (moderately identifying). Hence, highly and moderately identifying viewers rated their confidence about half a point higher on the 5-point scale in the case of having watched a win while the highly identifying estimated it even a bit greater compared to the moderately identifying. In contrast, there was no significant indirect effect for low identifying viewers, as the calculated CI included zero (95% CI [−.18, .77]). Similar results were obtained regarding the indirect effect on viewers’ estimation of the economic situation (see Table 1). Looking at highly and moderately identifying viewers, the indirect effects were significant and came to a point estimation of .18 (95% CI [.02, .51]) and .14 (95% CI [.02, .38]). Highly and moderately identifying viewers rated the current economic situation of their home country about .2 points higher on the 5-point scale in the case of having watched a win. In contrast, low identifying viewers were again not indirectly influenced (95% CI [−.03, .36]). Summing up, television viewers of football games appear only to be indirectly influenced by game outcome via mood if they highly or moderately identify with one of the teams. Hypothesis 3 was supported.

Table 1.

Conditional Indirect Effect of Watching a Win on Viewers’ Estimation of Self-Confidence and the Economic Situation at Levels of Team Identification (N = 119).

	Self-Confidence		Economic Situation
Levels of Team Identification	Point Estimate	95% Bias-Corrected Bootstrap Confidence Interval	Point Estimate	95% Bias-Corrected Bootstrap Confidence Interval
Low (3.20)	.26	[−.18, .77]	.09	[−.03, .36]
Moderate (4.26)	.43	[.15, .75]	.14	[.02, .38]
High (5.00^a)	.54	[.17, .95]	.18	[.02, .51]

^a5.00 equals not the value of one standard deviation above the mean (5.33 equals), but 5.00 represents the highest value possible as the moderator was measured on a scale from 1 to 5.

Discussion

The goal of the present study was to test whether the outcomes of games at the FIFA World Cups can affect viewers’ estimations regarding their own self-confidence and the current economic situation of their home country. The relationship was assumed to be such that game outcomes influence viewers’ estimation indirectly via viewers’ mood. The results showed that the mood of television viewers was influenced by the outcome of the game. In line with the disposition theory of sports spectatorship (Zillmann et al., 1989), viewers of televised football games were in a significantly better mood after watching their favored team win compared with before the game. In addition, their mood was decreased after watching their favorite team lose. It is important to note that the drop in mood after watching a defeat was nonsignificant and was approximately one third of the magnitude of the enhanced mood that accompanied a win (see the results for Hypothesis 1). The relatively small magnitude of this change in mood can be attributed to the fact that viewers whose mood was negatively affected after a loss would be less likely to take part in an online survey compared with viewers being positively affected after a win. It could be assumed that people most strongly affected by a loss would be especially unmotivated to take part in the survey, whereas this is unlikely to hold true for viewers with a strongly enhanced mood after a win. Future studies researching negative mood effects after negative sports results might apply survey methods that are less dependent on participants’ initial motivation, such as the telephone survey of Schwarz and colleagues (1987).

The present results also reveal that the outcomes of televised football games influence viewers’ estimation of self-confidence and economic evaluations, as people make judgments based on mood states influenced by football outcomes. Supporting feeling-as-information theory (Schwarz, 2011; Schwarz & Clore, 1983), people report themselves to be more self-confident and estimate the current economic situation to be more positive if being in a better mood caused by having seen their favored team win a football match. Hence, judgments being completely unrelated to sports events in the first place can still be influenced by the broadcasting and watching of such events, as they alter viewers’ mood and consequently their judgments. Despite the missing empirical evidence in the study by Hirt and colleagues (1992), the present work was able to support the mediating role of mood in feeling-as-information processes evoked by sports spectatorship. Interestingly, there were no additional direct effects of game outcomes on viewers’ estimations, indicating that viewers’ estimations depend primarily on their current mood instead of being directly inferred from sports events. As indirect effects can occur regardless of significant direct or total effects, according to current mediation analysis practice (Rucker, Preacher, Tormala, & Petty, 2011), the missing direct effects do not contradict the results.

In fact, the indirect effects indicate that viewers’ estimations are only influenced if their mood has been changed. As shown by the results and predicted by the disposition theory of sports spectatorship (Zillmann et al., 1989), such changes in mood not only depend on the performance of a favored team but also on the viewer’s particular identification with the team. Specifically, of the people who watched their favored team win, only those who were highly or moderately identifying with their team were in a better mood and consequently reported enhanced estimations. In contrast, low identifying viewers showed flat or no affective effects and thereby no estimations reflecting the game outcome. The present findings support the moderating role of team identification (Wann, 2006), which is recommend to be included in future studies.

Beyond replicating previous results—especially the results by Schwarz et al. (1987) toward the present study was oriented—previous results were extended in several ways. While Schwarz et al. (1987) solely tested mood-as-information effects regarding male participants watching football, the present results can be transferred to the whole German population as female viewers were included matching the increasing number of women watching FIFA World Cups (Geese et al., 2006; Gerhard et al., 2010). In addition, the present study actually statistically tested and confirmed the mediating role of mood in the assumed mood-as-information process. Schwarz and colleagues (1987) as well as others (Schweitzer et al, 1992) simply assumed differences in viewers’ estimations after different game outcomes as effects of different mood levels without testing their mediating influence. Furthermore, mood-as-information effects on economic estimations evoked by sports events were explicitly tested and confirmed, while previous studies either did not focus on economic measures (e.g., Hirt et al., 1992; Schwarz et al., 1987; Schweitzer et al., 1992) or did not test the actual influence of mood on economic estimations or behavior (e.g., Chang, Chen, Chou, & Lin, 2012; Edmans, García, & Norli, 2007). Finally, this study showed that mood changes evoked by football games as well as mood-as-information effects stay in effect for longer periods, in line with the general understanding of moods (Schwarz & Clore, 2007). As mood and estimations were assessed within 2 days after each match and participants attended the survey roughly equally on each of the 2 days, effects on mood and estimations can be viewed to last for at least 24 hr on average. In contrast, previous studies measured effects on mood and estimations generally right after the sports event accounting only for short-term effects (Hirt et al., 1992; Schwarz et al., 1987; Schweitzer et al., 1992).

The assessment of mood and estimations after a longer period has passed since the mood-precipitating event probably also accounts for the rather small effect sizes, as moods level off over time or are changed by new events. Still, highly identifying viewers rated their self-confidence more than half a point higher, and the economic situation about .2 points better, on a 5-point scale after watching their team win compared with before the game. The potentially enormous audience simultaneously affected by a televised sport competition (e.g., 29.7 million German viewers in the case of the semifinal loss in the 2006 FIFA World Cup) indicates the importance of even the rather low effect sizes. In fact, effects on stock market returns (Chang et al., 2012; Edmans et al., 2007) might easily occur if more than a third of a country’s population would simultaneously feel more self-confident and estimate the current economic situation in a more positive light. From this perspective, the role of television in this process should not be underestimated. In addition, television probably enforces such phenomena, as emotionalized broadcasting strategies are routinely employed (Bryant, Comisky, & Zillmann, 1977).

Naturally, the present research is not free of limitations. Despite the fact that various reasons spoke in favor of collecting data using an online survey (Birnbaum, 2004), the present usage of an online survey limited the results in two different ways. First, since participants took part in the survey voluntarily, choosing the day of their participation as they preferred, they were not randomly allocated to the different experimental conditions and did not respond to the survey at exactly the same time within each group. Consequently, measured differences between the groups could possibly have arisen from sources other than the intended win versus defeat manipulation, although there were no differences between the groups concerning possible influential variables such as age, sex, or team identification. Thus, results can only be conditionally interpreted in terms of clear causal relations. The second limitation caused by the online survey pertains to the fact that participants had to motivate themselves to take part in the study. While this might not be a problem when possible participants are in an elated mood after experiencing a win of a favored team, it clearly can be become a problem when trying to survey moody people having been most negatively affected by the defeat of a favored team. Consequently, we assume that such people were not motivated to take part in the survey and missing effects after the German defeat were probably due to such sample deficits. As mentioned above, we recommend using a different research method to reduce such sample deficits in future studies. Furthermore, the present results are limited to large sports events that probably exert a stronger influence on viewers’ mood compared with, for instance, a routine football game in a national league. In addition, the selected competition took place in the viewers’ home country and naturally meant something special to them. Thus, it seems worth testing whether the effects hold true for less momentous sporting events. A further limitation pertains to the fact that the game used as the experimental win manipulation ended very close after penalty. Past research “has revealed that suspense corresponds to the closeness of a game such that smaller score differentials produce more suspense and subsequently more enjoyment” (Knobloch-Westerwick et al., 2009; see also Peterson & Raney, 2008). Hence, it is clearly debatable if games featuring less closeness and suspense result in similar effects. Finally, our outcome measures limit the study. Mood and self-confidence were solely measured by 2 items and the estimation of the economic situation only by 1 item. Applying more comprehensive measures in future studies will most likely give a more thorough understanding of the investigated effects and more strongly consolidate the empirical evidence.

Nevertheless, the present study showed that television viewers watching sports events are affected in their mood depending on the competition’s outcome and provided that there is some level of identification with one of the competing teams. More importantly, altered mood states affect viewers’ estimations; that is, viewers feel more self-confident and would rate the economic situation of their home country more highly after watching a win by their favored team. As a result, large sports events televised to an entire nation or even worldwide are clearly able to affect spheres of public life totally unrelated to the sports event itself at first glance. From such a perspective, it seems unsurprising that politicians and decision makers would seek to link themselves to such events, hoping to transfer some of the possibly positive mood of viewers to their perceptions of the politicians and consequently to important issues. Returning to the 2006 FIFA World Cup in Germany, the Economist (2006) put it in a nutshell when previewing the tournament in the context of Germany’s poor economic situation at the time: “The greatest show on earth kicks off in Germany ( … ) For the German government there is more riding on the tournament than [mere] sporting success” (Let the games, 2006).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Berndt

N. C.

Hayes

A. F.

Verboon

Lechner

Bolman

Vries

H. De

. (2012). Self-efficacy mediates the impact of craving on smoking abstinence in low to moderately anxious patients: Results of a moderated mediation approach. Psychology of Addictive Behaviors, 27, 113–124.

Birnbaum

M. H.

(2004). Human research and data collection via the internet. Annual Review of Psychology, 55, 803–832.

Bloom

(2009). The impact of uncertainty shocks. Econometrica, 77, 623–685.

Brown

Bryant

(2006). Sports content on U.S. television. In Raney

A. A.

Bryant

(Eds.), Handbook of sports and media (pp. 77–104). Mahwah, NJ: Lawrence Erlbaum.

Bryant

Comisky

Zillmann

(1977). Drama in sports commentary. Journal of Communication, 27, 140–149.

Bryant

Cummins

R. G.

(2009). The effects of outcome of mediated and live sporting events on sports fans’ self- and social identities. In Hundley

H. L.

Billings

A. C.

(Eds.), Examining identity in sports media (pp. 217–238). Thousand Oaks, CA: Sage.

Chang

S.-C.

Chen

S.-S.

Chou

R. K.

Lin

Y.-H.

(2012). Local sports sentiment and returns of locally headquartered stocks: A firm-level analysis. Journal of Empirical Finance, 19, 309–318.

Clore

G. L.

Schwarz

Conway

(1994). Affective causes and consequences of social information processing. In Wyer

R. S.

Srull

T. K.

(Eds.), Handbook of social cognition (pp. 323–418). Hillsdale, NJ: Lawrence Erlbaum.

Let the games begin (2006). The Economist. Retrieved from http://www.economist.com/node/7050113

10.

Edmans

García

Norli

Ø.

(2007). Sports sentiment and stock returns. The Journal of Finance, 62, 1967–1998.

11.

Geese

Zeughardt

Gerhard

(2006). Die Fußball-Weltmeisterschaft 2006 im Fernsehen. Daten zur Rezeption und Bewertung [The 2006 FIFA World Cup on Television. Data concerning reception and assessment]. Media Perspektiven, 9, 454–464.

12.

Gerhard

Kessler

Gscheidle

(2010). Die Fußball-Weltmeisterschaft 2010 im Fernsehen. Daten zur Rezeption und Bewertung [The 2010 FIFA World Cup on Television. Data concerning reception and assessment]. Media Perspektiven, 9, 382–389.

13.

Hayes

A. F.

(2012). PROCESS: A versatile computational tool for observed variable moderation, mediation, and conditional process modeling. Retrieved from http://www.afhayes.com/public/process2012.pdf

14.

Hayes

A. F.

(2013). An introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford.

15.

Hirt

E. R.

Zillmann

Erickson

G. A.

Kennedy

(1992). Costs and benefits of allegiance. Changes in fans’ self-ascribed competencies after team victory versus defeat. Journal of Personality and Social Psychology, 63, 724–738.

16.

Hulin

(2001). Cronbach’s alpha on two-item scales. Journal of Consumer Psychology, 10, 55–69.

17.

Knobloch-Westerwick

David

Eastin

M. S.

Tamborini

Greenwood

(2009). Sports spectators’ suspense: Affect and uncertainty in sports entertainment. Journal of Communication, 59, 750–767.

18.

Morris

W. N.

(1989). Mood: The frame of mind. New York, NY: Springer.

19.

Peterson

E. M.

Raney

A. A.

(2008). Reconceptualizing and reexamining suspense as a predictor of mediated sports enjoyment. Journal of Broadcasting & Electronic Media, 52, 544–562.

20.

Preacher

K. J.

Hayes

A. F.

(2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, and Computers, 36, 717–731.

21.

Raghunathan

Pham

M. T.

(1999). All negative moods are not equal: Motivational influences of anxiety and sadness on decision making. Organizational Behavior and Human Decision Processes, 79, 56–77.

22.

Raney

A. A.

(2006). Why we watch and enjoy mediated sports. In Raney

A. A.

Bryant

(Eds.), Handbook of sports and media (pp. 313–330). Mahwah, NJ: Lawrence Erlbaum.

23.

Raney

A. A.

(2009). The effects of viewing televised sports. In Nabi

R. L.

Oliver

M. B.

(Eds.), The Sage handbook of media processes and effects (pp. 439–453). Thousand Oaks, CA: Sage.

24.

Rucker

D. D.

Preacher

K. J.

Tormala

Z. L.

Petty

R. E.

(2011). Mediation analysis in social psychology: Current practices and new recommendations. Social and Personality Psychology Compass, 5, 359–371.

25.

Scherer

K. R.

(1998). Emotionsprozesse im Medienkontext. Forschungsillustrationen und Zukunftsperspektiven [Emotion Processes in Media Contexts. Research Illustrations and Future Prospects]. Medienpsychologie, 10, 276–293.

26.

Schramm

Klimmt

(2003). “Nach dem Spiel ist vor dem Spiel.” Die Rezeption der Fußball-Weltmeisterschaft 2002 im Fernsehen: Eine Panel-Studie zur Entwicklung von Rezeptionsmotiven im Turnierverlauf [“After the Game is Before the Game.” The reception of the 2002 FIFA World Cup on Television: A panel study concerning the development of reception motives in the course of the tournament]. Medien & Kommunikationswissenschaft, 51, 55–81.

27.

Schwarz

(2011). Feelings-as-Information Theory. In van Lange

P. A. M.

Kruglanski

A. W.

Higgins

E. T.

(Eds.), Handbook of theories of social psychology (pp. 289–308). Los Angeles, CA: Sage.

28.

Schwarz

Clore

G. L.

(1983). Mood, misattribution, and judgements of well-being. informative and directive functions of affective states. Journal of Personality and Social Psychology, 45, 513–523.

29.

Schwarz

Clore

G. L.

(1988). How do I feel about it? The informative function of affective states. In Fiedler

Forgas

J. P.

(Eds.), Affect, cognition, and social behavior (pp. 44–62). Toronto, Canada: Hogrefe & Huber.

30.

Schwarz

Clore

G. L.

(2007). Feelings and phenomenal experiences. In Kruglanski

A. W.

Higgins

E. T.

(Eds.), Social psychology. Handbook of basic principles (pp. 385–407). New York, NY: Guilford.

31.

Schwarz

Strack

(1985). Cognitive and affective processes in judgments of well-being: A preliminary model. In Brandstaedter

Kirchler

(Eds.), Economic Psychology (pp. 439–447). Linz, Austria: Trauner.

32.

Schwarz

Strack

Kommer

Wagner

(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.

33.

Schweitzer

Zillmann

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.

34.

Seidman

(2012). Eli, Eli, Eli! Super Bowl XLVI Is most-watched television show in U.S. History. Retrieved from http://tvbythenumbers.zap2it.com/2012/02/06/eli-eli-eli-super-bowl-xlvi-is-most-watched-television-show-in-u-s-history/118826/

35.

Wann

D. L.

(2006). The causes and consequences of sport team identification. In Raney

A. A.

Bryant

(Eds.), Handbook of sports and media (pp. 331–352). Mahwah, NJ: Lawrence Erlbaum.

36.

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 & Exercise Psychology, 16, 347–364.

37.

Zillmann

Bryant

Sapolsky

B. S.

(1989). Enjoyment from sports spectatorship. In Goldstein

J. H.

(Ed.), Sports, games, and play: Social and psychological viewpoints (pp. 241–278). Hillsdale, NJ: Lawrence Erlbaum.