Gambling and fantasy: an examination of the influence of money on fan attitudes and behaviors.

Mahan, Joseph E., III ; Drayer, Joris ; Sparvero, Emily 等

Introduction

Gambling remains a highly regulated and, in some cases, legally prohibited industry. Holleman (2006) argued that "legislatures and courts have largely deemed gambling illegal because of the social ills it creates" (p. 74). However, while lawyers and legislators continue to debate the merits and legality of sports gambling, the prevalence of the activity is undeniable. In fact, some point to the involvement of both governments and sport organizations as lending implicit support to such activities (McKelvey, 2004a). The result is a multi-billion dollar industry and, while exact numbers are difficult to ascertain, some have estimated that as much as $380 billion is illegally wagered annually on sports (Levinson, 2006). Included among the opportunities are a variety of overseas sports betting websites such as bodog.com, as well as a growing number of domestic sports betting options (both legal and illegal).

While the social ills created by sports gambling may be of some concern, the influence of both traditional sports betting as well as wagering on fantasy sports on traditional fan attitudes and behaviors may be more salient to sport marketers, especially as some states consider deregulating sports betting and the popularity of fantasy sports continues to grow. Therefore, the purpose of the current study is to understand attitudinal and behavioral differences of individuals who participate in sports betting and/or fantasy sports as compared with individuals who do not participate in these activities. Specifically, this research looks at the relationships between participation in these activities and team identification, spending behaviors, sport interest, and enjoyment of sport.

Review of Literature

Sports Gambling

The landscape of sports gambling has been characterized as including both illegal and legal facets (Moorman, 2008). Following Moorman (2008), this paper considers 'sports gambling' as a comprehensive term to describe a variety of behaviors, including such activities as sports betting and certain aspects of fantasy sports participation. Sports betting has been defined as wagering money on the outcome of sporting events (Frey, 1992), whereas it has been suggested that fantasy sports may be considered gambling in some instances due to the potential gain or loss of money (e.g., Lee, Kwak, Lim, Pedersen, & Miloch, 2011).

While much of the scholarly literature on gambling--including betting on sports--appears to be centered on the pathological aspects of such activities (cf., Petry, 2003), there is a growing segment of research that examines the 'social gambling' phenomenon. Those studies focusing on the social aspects of sports betting can be split into three streams: economic aspects (e.g., Paul & Weinbach, 2010), policy implications (e.g., McKelvey, 2004a), and consumer behavior (e.g., Nelson et al., 2007).

Economic effects of sports betting. Perhaps the area that has received the most attention in the scholarly literature on sports betting concerns the economics of such activities. Some of these studies have identified relationships among certain betting metrics (e.g., point spreads), betting volume, and sport consumption patterns or behaviors. For example, Paul and Weinbach (2010) investigated the predictors of betting volume (i.e., number of bets placed per game) in both the National Basketball Association (NBA) and National Hockey League (NHL). Findings pointed to team quality--indicated by point spreads--and television coverage as having significant influence on betting volume (i.e., individuals bet more on contests between the best teams, where the outcome is uncertain, and when the game is widely televised). This finding led the authors to conclude that bettor behavior appears to approximate fan behavior, particularly in that both bettors and fans are drawn to the better teams as well as the outcome uncertainty associated with (professional) sport events. Paul and Weinbach's conclusions partially supported those from an earlier investigation of the role of team 'popularity' on betting odds in Spanish football (Forrest & Simmons, 2008). Based on a large sample of bets on professional football in Spain, Forrest and Simmons pointed to the existence of a 'sentiment bias.' Specifically, they concluded that team popularity appeared to exert influence on the odds of a particular match. While the scholars in this area have purported to estimate the relationship between sports betting and consumption behaviors, each utilized aggregated data. As such, future research that explores individual response (e.g., attitudes and behaviors) could have utility in further examining the nature and extent of these relationships.

Policy implications of sports betting. Some scholars have postulated about potential effects of the link between certain gambling-related phenomena (e.g., lotteries, fantasy sports) and sport organizations. Despite the outcry against gambling by sport leagues, it is offered that these same groups continue to pursue marketing and sponsorship ties with businesses in the gaming industry (e.g., casinos). McKelvey (2004a) pointed to the formalized ties between sport organizations and gambling entities as perhaps representing a blurring of the lines between organized sport and gambling enterprises. Indeed, this apparent complicity by sport leagues prompted a call for scholarly investigation of the potential effects of gambling-related sponsorship communications on sport consumers (McKelvey, 2004b). Specifically, the author proposed a line of inquiry similar to that examining the effects of alcohol and tobacco sponsorships on consumer attitudes and behaviors. Thus, investigating the nature of relationships between sport betting preferences and sport consumer behaviors could provide insight into possible benefits or consequences of a connection between gambling and sports.

Sports betting behavior. The vast majority of the scholarly work on gambling behavior, including betting on sports, has focused on identifying those factors that influence an individual's preferences for engaging in such activities. Such factors range from sociodemographic indicators (e.g., gender) to individual differences, including personality traits. In a study examining gambling preferences, Petry (2003) identified 'sports gamblers' as younger males with moderate gambling problems, which differentiated them from those who preferred other gambling activities (e.g., slot machines). This finding suggests that those who bet on sports may tend to be social or 'recreational' gamblers rather than 'problem' gamblers. Moreover, this heterogeneity of gambler profiles (i.e., ranging from 'recreational' to 'problem') may signify the need for research that focuses specifically on sports betting.

Another factor, an individual's interest in sports, has received only scant attention in the literature but could play a role in explaining sports betting behavior. One attempt to explore this notion operationalized 'interest' in sports as a person who identified as a spectator or participant (Nelson, et al., 2007). In a survey of over 10,000 college students, results indicated that those who reported higher levels of sports interest were more likely to bet on college sporting events. Moreover, despite finding a significant gender effect (i.e., males more likely to bet), the authors implied that situational or contextual factors, such as sports interest, may be (more) important in explaining sports betting behavior. A more recent study employed a similar approach, examining how one's participation in sports (i.e., current or past) related to various gambling behaviors (Weiss & Loubier, 2010). Findings suggested that former athletes show preferences for skill-based gambling activities, including betting on sports.

This line of inquiry has made inroads into highlighting the relationship between one's connection (e.g., being a 'fan') to sport and certain gambling behaviors. However, it appears that application of a theoretical construct, such as sport involvement or team identification, could have utility in further explicating sports betting behavior. Scholars have begun to explore this notion in the context of sports betting acting as an economic motivation for sport consumption, but results have been, as yet, inconclusive (cf. Wann, Grieve, Zapalac, & Pease, 2008). As such, further investigation could prove useful in this regard. Based on the above review of literature, the following research questions were proposed:

RQ1: What is the nature of the relationship between sports betting behavior and level of team identification?

RQ2: What is the nature of the relationship between sports betting behavior and sport spending levels?

RQ3: What is the nature of the relationship between sports betting levels (i.e., amount wagered per month) and specific sport attitudes (i.e., interest and enjoyment) and behaviors (i.e., attendance)?

Fantasy Sports

Fantasy sports and the law. As mentioned previously, the overall landscape of sports gambling includes both the traditional forms of sports betting as well as newer activities such as fantasy sports. In fantasy sports leagues, participants may choose to include a league entry fee with the winner of the league receiving most, if not all, of the prize money. Over the years, authors have debated whether or not this activity should be considered in the same way as traditional sports betting. Ultimately, the Unlawful Internet Gambling Enforcement Act (UIGEA) was passed in 2006 as Title VII of the SAFE Port Act for the purpose of prohibiting gambling-related businesses from accepting payments associated with illegal internet gambling (Doyle & Yeh, 2009). While UIGEA effectively eliminated online sports betting as well as online poker games, the government carved out a specific exemption for fantasy sports. Despite the lawmakers' decision on the topic, the debate around whether or not fantasy sports should be considered a form of online gambling has continued. Holleman (2006) argued that while existing anti-gambling laws aim to protect the "economically disadvantaged," the demographic profile of fantasy participants is quite different than other gamblers and, consequently, fantasy participants do not require the same protections (p. 75). Fisher (2008) reported that fantasy players have an annual household income of approximately $94,000 and are much stronger consumers of many of the leading product categories than non-players. On the other hand, others have argued that fantasy sports have become a gambling culture. Indeed, Bernhard and Eade (2005) offered: "As is the case with more conventional forms of sports wagering, many claim to find the game more interesting when money is risked and awarded to the winners" (p. 36).

Within the legal system, the debate of whether or not fantasy sports should be considered an illegal form of Internet gambling has often centered on whether it is considered a game of luck or a game of skill. According to Holleman (2006), "the only games that fall under the gambling statutes are those classified as games of chance rather than games of skill" (p. 68). Within the context of fantasy sports, Holleman (2006) stated that "through research, intelligence, and skill, the participants can control the outcome of the contests" suggesting that it would qualify as a game of skill (p. 79). On the other hand, Bernhard and Eade (2005) stated that "if we broadly define gambling as an activity that risks something of value (substantial amounts of money) on an event whose outcome is undetermined (such as the whims of a professional baseball season), fantasy baseball clearly qualifies" (p. 29).

In the end, UIGEA was passed in 2006 and contained the exemption for fantasy sports, suggesting that the government was convinced that fantasy sports "are not accompanied by the same social ills as other games where elements of chance are involved" (Holleman, 2006, p. 79). Interestingly, the National Football League, which like all other major professional sports leagues has strict rules against betting on sports, has no league rules that restrict player, coach, or official participation in fantasy sports. Ultimately, understanding how paying for play in fantasy sports leagues impacts the attitudes and behaviors of participants may provide useful insights into whether the activity belongs in the same category as other traditional forms of gambling. These potential similarities and/or differences may help to guide future policies and legislation.

Fantasy sports' impact on attitudes and behaviors. Although there is currently no research which has examined the effect of pay-for-play leagues on fantasy players, several studies have examined how fantasy sports participation impacts the attitudes and behaviors of the traditional sports fan. One study suggested that fantasy sports added another point of attachment for fans through an increased interest in the statistical output of individual players (Drayer, Shapiro, Dwyer, Morse, & White, 2010). The authors concluded that this additional interest in the league resulted in increased levels of consumption, particularly mediated consumption. The findings of this qualitative inquiry were largely supported by the survey-based research conducted by Karg and McDonald (2011). The authors found that, compared to non-fantasy players, fantasy sport players reported higher levels of attachment, team identification, and loyalty as well as increased sport-related consumption such as game attendance, television viewing, and secondary spending.

Although the research on the impact of fantasy sports on the consumption patterns of participants is somewhat limited, there are a few studies on the topic which have all produced similar results. For example, Nesbit and King (2010a, 2010b) confirmed fantasy sports participation's impact on mediated consumption, but they also found that it increased fan attendance. More recently, Dwyer and Drayer (2010) separated existing fantasy players into different groups based on varying levels of interest in an individual's favorite team or fantasy team. The authors found that those with a strong interest in their fantasy team (regardless of their level of interest in their favorite team) consumed sport at significantly higher levels than those fantasy participants with lower levels of interest in their fantasy team.

To date, the literature seems consistent with regard to the potential influence of fantasy participation on fan attitudes and behaviors. However, given the gambling culture that exists in fantasy sports (Bernhard & Eade, 2005) and the fact that gambling has been cited as a motive to participate in fantasy sports (Dwyer & Kim, 2011), research is needed to understand the role that money plays in the attitudes and behaviors of fantasy participants, particularly in light of the continued debate on the fantasy sports exemption within UIGEA. Moreover, while many studies in psychology have explored the connection between personality traits and gambling behavior, only recent efforts have expanded that relationship to include participation in fantasy sports (Lee et al., 2011). Whereas there has been much debate over whether or not fantasy sports should be considered gambling, some scholars have drawn a parallel based on similarities in the arousal potential of both forms of activity (Farquhar & Meeds, 2007). Following this notion, Lee et al. (2011) found certain personality traits, including sensation seeking and locus of control, to have significant influences on attitudes and intentions toward fantasy sports for male participants. This study linking fantasy sports participation with gambling behavior through the application of psychology theory represented only an initial look into the underlying motives behind fantasy sports participation. As such, additional research that incorporates other domain-specific constructs (e.g., team identification) could prove fruitful in this context (Lee et al., 2011). Subsequently, the current study is also driven by the following research questions:

RQ4: What is the nature of the relationship between participation in fantasy sports and level of team identification?

RQ5: What is the nature of the relationship between participation in fantasy sports and sport spending levels?

RQ6: What is the nature of the relationship between paying to participate in fantasy sports and specific sport attitudes (i.e., interest and enjoyment) and behaviors (i.e., attendance)?

Methods

This study surveyed individuals who had opted to receive a major U.S. metropolitan newspaper's "Sports" email newsletter. Recipients of the newsletter (N = 23,500) were emailed an invitation to participate in a survey of local sport fans' attitudes. A total of 2,429 responses were received, with a useable yield of 2,303 surveys (9.8% response rate). Survey respondents were predominantly male (82%) and white (86%). They were well educated (46% held a bachelor's degree, and 23% held a graduate degree), and affluent (88% reported incomes above the $50,000 median household income level). The respondents ranged in age from 18 to 75, with an average age of 47. When compared to the typical sports fan, our sample was younger, more affluent, and included more males. A recent report in the SportsBusiness Daily ("Fan Demographics," 2010) found that there were marginally more male than female sport fans (approximately 60% to 40% in the four major leagues), approximately 18% of fans have household incomes greater than $50,000, and approximately 40% of fans are older than 50. Our sample was consistent in terms of ethnicity (82% white).

Team identification was measured using Mael and Ashforth's (1992) six-item organizational identification scale. The items were customized based on the respondent's previous selection of one local team as their favorite. Statements were rated on a five-point Likert scale, and individual items were summed and averaged to create an index variable (TeamID).

Respondents were asked about their participation in both gambling and fantasy sports. Relative to gambling, all participants were first asked if they bet on sports (SportsBet). Those who responded 'yes' to this item were then asked which sports they bet on, as well as an open-ended question about how much they wager in an average month. For use in the analyses, frequencies of the monthly betting item were examined and a categorical variable (MonthWager) was created using a tripartite split. This resulted in three groups: low (<$24/month), medium ($25-$99/month), and high ($100+/month). In addition, individuals who indicated that they had bet on sports were asked to rate their agreement with the following statements: (1) betting increases my overall interest in sports (BetInterest), (2) betting on sports makes watching sports more enjoyable (BetEnjoy), and (3) overall, I enjoy betting on sports more than watching sports (EnjoyMore).

Similarly, participants were asked if they played fantasy sports (Fantasy). Those respondents who indicated 'yes' were presented follow-up items related to whether or not they paid to play (FantPay) as well as rating their agreement with the following statements: (1) playing fantasy increases my overall interest in sports (FantInterest), (2) playing fantasy increases my enjoyment of watching sports (FantEnjoy), and (3) I enjoy playing fantasy more than actually watching sports (FantMore).

Finally, all participants were asked about their spending behaviors related to professional sport. Specifically, participants were asked to estimate their average monthly expenditures on tickets to games (SpendTix), team merchandise (SpendMerch), other gameday expenses (e.g., parking, concessions) (SpendExtra) and other sport-related expenses (e.g., cable subscriptions, sport magazines) (SpendOther). A continuous graphic scale was used that was anchored numerically (from $0 to $2000). Participants used a sliding bar to indicate their level of expenditure in each category.

Given the nature of the proposed research questions, t-tests as well as analysis of variance procedures were employed. Dichotomous independent variables (i.e., SportsBet and Fantasy) allowed for the use of independent t-tests in exploring both RQ1 and RQ4. For the remaining RQs, ANCOVA and MANCOVA were selected because household income was applied as a covariate in order to isolate potential effects of the categorical independent variables on the chosen dependent variables. The sample showed the greatest difference from the general sports fan population on household income, and the decision to control for income was in an attempt to improve the generalizability of the results.

Results

Descriptive Statistics

The distributions of variables related to an individual's sport affiliation in general (SportsFan) as well as with a particular organization (TeamID) were examined for normality. The TeamID measure (M = 2.99, SD = 0.80) was normally distributed (skewness = -.115; kurtosis = -.197). Cronbach's alpha for this scale was in line with existing research (a = .83), indicating acceptable internal consistency of the measure. Correlations among constructs of interest (i.e., TeamID, SportsBet, Fantasy, and FantPay) are shown in Table 1.

Sports Betting Analyses

In exploring the relationship between sports betting (SportsBet) and TeamID (RQ1), results of a t-test indicated significant differences in level of TeamID between those who responded 'yes' (M = 3.06) and 'no' (M = 2.94) to engaging in sport betting (see Table 2). ANCOVA results demonstrated that, after controlling for income, both TeamID (F = 11.24, p < .001) and SportsBet (F = 44.26, p < .001) explained 4% of the variance in reported number of sport events attended (NumEvents). The interaction (TeamID x SportsBet) was not significant. These findings provided a response for RQ1 by suggesting those who bet on sports are more highly engaged as sports fans than those who do not bet on sports.

MANCOVA analysis (see Table 3) revealed significant differences in spending behaviors (RQ2) as explained by both TeamID and SportsBet after controlling for respondent income level, though there was no significant interaction (TeamID x SportsBet). Univariate results showed both TeamID and SportsBet to account for a small amount of variance (4%) in spending on sport event tickets (SpendTix) and team-related merchandise (SpendMerch), but only SportsBet was significant (p < .001) for other event-related spending (SpendExtra). Conversely, the main effect of non-event spending (SpendOther) as well as interactions (TeamID x SportsBet) for all four univariate tests were not significant. These results demonstrated that those who bet on sports may engage in more sport-related spending, but perhaps only in the specific context of events (e.g., tickets).

Examination of attitudinal and behavioral differences among those who reported betting on sport (i.e., only 'yes' responses to SportsBet item) utilized a tripartite split of the MonthWager variable (RQ3). This resulted in monthly spending on sports betting classified as 'high spending,' 'moderate spending,' and 'low spending.' The one-way ANOVA of MonthWager on TeamID was not significant (p > .05), indicating no differences in one's affiliation with a team based on how much is spent per month on sport betting. As such, TeamID was not included in the remaining analyses of attitudes and behaviors within the sports betting group.

MANCOVA results (see Table 4) indicated significant effects of MonthWager on the betting-related attitudes explored in this study, after controlling for income level. Follow-up univariate tests demonstrated that sport betting increased overall interest in sports (BetInterest) for those in the 'high spending' category in relation to those in either the 'moderate' or 'low' spending groups. Further, enjoyment of sports as a result of sports betting (BetEnjoy) was significantly higher (p < .001) for those in the 'high spending' group than the two comparison groups. There were no significant differences (p > .05) among those who bet on sports on the third attitudinal item (i.e., "I enjoy betting on sports more than actually watching sports"). Thus, it would seem that the more individuals wager on sports per month, the more likely they are to have positive attitudes toward sport.

Examination of the behaviors among those who bet on sports revealed significant differences between MonthWager groups on the majority of attendance and spending indicators employed in this study. ANCOVA results showed, after controlling for income, significant variance (3%) in NumEvents (F = 9.24, p < .001) as explained by an increase in the reported amount spent per month on sports betting (i.e., MonthWager). Results of a MANCOVA on spending behaviors revealed MonthWager to explain significant levels of spending (5% total variance) related to sports events (see Table 5). Univariate tests demonstrated that spending on tickets, team-related merchandise, and other event-related spending was greater for those who spent more per month on sports betting (i.e., 'high spending') than for those who did not spend as much (i.e., 'moderate' and 'low spending'). Differences in non-event spending were not significant (p > .05). Taken together, these results indicated that increases in spending on sports betting appear to parallel increases in 'traditional' spending on sports.

Fantasy Participation Analyses

Similar to the sports betting analyses, the research questions related to fantasy sport participation (Fantasy) were explored through the use of several analyses. In response to RQ4, t-test results indicated significant differences in TeamID between those who responded 'yes' (M = 3.10) and 'no' (M = 2.91) to fantasy sport participation (see Table 2). After controlling for income, results of an ANCOVA on NumEvents demonstrated significant main effects for both TeamID (F = 9.40, p < .001) and Fantasy (F = 80.44, p < .001) while the interaction (TeamID x Fantasy) was not significant. These findings responded to RQ4 by suggesting that fantasy sports participants are more highly engaged in sports than their non-fantasy playing counterparts.

The MANCOVA (see Table 6) run on spending behaviors revealed significant main effects for both TeamID and Fantasy but not a significant interaction (TeamID x Fantasy); together, these explained 4% of total variance in the selected spending behaviors. Univariate tests indicated significant main effects only for Fantasy on SpendMerch (F = 20.50, p < .001) and SpendExtra (F = 16.83, p < .001). All other main effects and interactions were not significant, thus providing an answer for RQ5 and indicating that fantasy sports participants spend more for sport-related purchases than those who do not play fantasy sports.

Investigation of attitudes and behaviors among fantasy sports participants (i.e., only 'yes' to Fantasy item) (RQ6) was performed by employing FantPay, a dichotomous (i.e., yes/no) variable indicating whether or not respondents pay money to participate in fantasy sports. The t-test of FantPay on TeamID (see Table 2) was not significant (p > .05), indicating no differences in one's affiliation with a team based on whether a respondent pays for fantasy sports participation. As such, TeamID was not included in the remaining analyses of attitudes and behaviors within the fantasy sports participant group.

MANCOVA results pointed to significant effects for FantPay on attitudes (see Table 7). Follow-up univariate tests revealed that participation in fantasy sports significantly influenced (p < .001) overall interest in sports (FantInterest) for those who pay to participate as opposed to those who play for free. Likewise, enjoyment of sports as a result of playing fantasy (FantEnjoy) was significantly higher (p < .001) for those who pay to play than for those who do not. Moreover, those individuals who pay for fantasy sports participation were significantly more likely than their counterparts to agree (p < .001) with the statement, "I enjoy playing fantasy more than actually watching sports" (FantMore). As such, it would appear that individuals who pay to play fantasy sports generally have more positive attitudes toward sport than those who do not pay to play.

Behavioral differences among fantasy sport participants were found for attendance and certain spending behaviors. An ANCOVA on NumEvents showed significant effects for FantPay (F = 15.13, p < .001), after controlling for income. The MANCOVA analysis on spending behaviors showed a significant multivariate test for FantPay (see Table 8), while the only significant univariate test was for SpendTix (F = 12.60, p < .001), indicating that those who pay for fantasy sport participation spend more on sports event tickets than those who play fantasy for free.

Discussion

The current study revealed some interesting trends with regard to how betting on sports in the traditional sports or through paying to play in a fantasy sports league may influence consumer attitudes and behaviors. The following sections will present implications for these findings as well as provide suggestions for future research. The first research question (RQ1) considered the relationship between sports betting and team identification. Individuals who bet on sports had higher identification with their favorite team than those who did not bet. Yet, team identification was not related to the amount that was bet (i.e., high spending, moderate spending, and low spending). This could imply that the decision to bet on sports may be yet another way in which an individual can demonstrate his or her identification with the team, regardless of the amount or even the outcome of the bet. Similarly, individuals who play fantasy sports (RQ4) have higher identification with their favorite team than those who do not play fantasy sports. However, there is no difference in team identification based on whether individuals pay to play or not. In both cases, the key relationship appears to be between the decision to participate in the ancillary activity (i.e., gambling or fantasy sports) and team identification, regardless of the monetary commitment to the said activity.

The remainder of the research questions related to sports betting (RQ2 & RQ3) and fantasy sports participation (RQ5 & RQ6) explored the extent to which each complemented other forms of sport consumption. We found that individuals are not simply spending more on gambling or fantasy sport because they have discretionary income to do so. Rather, these individuals have prioritized spending on sport--including gambling and fantasy sports--as an adjunct to more traditional spending on tickets, merchandise, and concessions. That is, both sports betting and participation in fantasy sports appear to serve as 'complementary investments' by sport consumers.

Sports Betting as a Complementary Investment in Sports

The extant literature on sports betting has largely focused on three areas: economics, policy implications, and behavior. The current study largely drew upon--and extended--the work in two of these areas (i.e., economics and behavior) through investigation into the relationship between sports betting and sport consumption behaviors. The results support existing studies that have suggested sports betting behavior exhibits a positive linear relationship with identification or interest in sports (Nelson et al., 2007). Indeed, whereas earlier studies have not specifically defined 'interest,' the present application of a theoretical construct (i.e., team identification) extended this line of inquiry by characterizing a relationship between one's identification with a sports team and participation in sports betting.

This notion of sports betting as a complementary investment appeared to be reinforced by the spending behavior findings (RQ2). In addition to indicating an association between sports betting and game-related spending (i.e., tickets, merchandise, and extras), results showed the amount spent per month (MonthWager) was positively related to certain sport-related expenditures. This correspondence between fans' spending on betting and more traditional purchases perhaps indicates that some fans seek multiple outlets of consumption as opposed to selecting one in favor of the other. The apparent complementarity of sports betting and game attendance as demonstrated by findings related to RQ3 would seem to lend further credence to this idea. These findings notwithstanding, additional scholarly inquiry--including development of a model of such relationships--would be helpful in investigating this proposition.

While the policy implications of sports betting were not explicitly explored in the current study, the findings herein could be interpreted as reinforcing McKelvey's (2004b) call for empirical investigation of the effects of gambling-related sponsorships on sport consumption. If fans who spend (more) money on sports betting are also attending more events, the resultant exposure to gambling images (e.g., sponsor signage) could have effects--either positive or negative--on consumer attitudes. Certainly, future research could prove fruitful in helping to determine the impact of such sponsor-property relationships.

Fantasy Sports as a Complementary Investment in Sports

Sports fans are famous for the significant emotional and financial investments they make in their favorite sport, team, and/or player. They spend heavily to attend games, purchase authentic jerseys and other merchandise, subscribe to expensive cable television packages, watch hours of programming, and have strong emotional reactions to success or failure. Subsequently, sport marketers are constantly trying to find new ways to engage fans. Recent research has suggested that participation in fantasy sports is one way that existing sports fans have found an additional way to invest, both emotionally and financially, in sports (Drayer et al., 2010; Dwyer & Drayer, 2010; Karg & McDonald, 2011). Although some have suggested that a new focus on individual players instead of entire teams might compromise fans' identification to their favorite team, each of the aforementioned studies concluded that fantasy sports participants maintained high levels of identification to their favorite team and simply utilized fantasy sports as an additional way to follow a particular sports league. Based on this additional channel through which to follow that league, fantasy players generally consume at higher levels than those who do not participate in fantasy sports.

The results of the current study generally support the findings of these previous studies. That is, there appear to be some individuals who both participate in fantasy sports and spend more money on game- and team-related (e.g., merchandise) expenses. However, the primary purpose of the current study was to expand upon these findings to see how the presence of an increased financial investment might influence consumer attitudes and behaviors. The evidence suggests that playing fantasy for money represents yet another way that fans may become even more emotionally and financially invested in sports. Specifically, although whether respondents played fantasy for money or not was not related to differences in team identification, those who did play for money had higher levels of interest in sports. This group also reported significantly greater enjoyment of sport as a result of participation in fantasy sports than those who do not play for money. So it appears that playing for money still does nothing to compromise traditional fan attitudes and actually increases participants' enjoyment of sports overall. Perhaps as a result of this increased enjoyment in sports, those for play fantasy sports for money spend significantly more on event tickets.

Ultimately, playing fantasy sports appears to be a complementary activity for highly-involved sports fans. Within this group of fantasy participants, those who play for money appear to be even more invested in sports, leading to moderate differences in attitudes and behaviors. However, the nature and magnitude of these differences does nothing to suggest that playing fantasy sports for money should be considered among the traditional forms of gambling. Rather, they appear to represent additional opportunities for some highly engaged individuals to augment their level of consumption.

Practical Implications

The findings of this study present both opportunities and challenges for marketers of professional sports teams and/or leagues. On one hand, it appears as though there is a group of individuals that is perhaps the most invested of all sports fans. Further, given the exorbitant amount of money spent on sports betting and fantasy sports each year, this group is quite large and ready and willing to spend. On the other hand, it is difficult to appeal to sports bettors as the activity is largely illegal outside of Las Vegas and a few other locations. While playing fantasy sports for money is legal, leagues may be reluctant to make mention of playing fantasy for money in any of their promotional materials or take sponsorship and/or advertising money from groups that provide any sort of gambling services, given the lingering association between fantasy sports and other illegal forms of gambling. So who can use this information?

Perhaps the group that can capitalize on these findings the most is the wide variety of media outlets (television, radio, or the Internet). These outlets can provide information related to sports betting (point spreads, over/under, etc.) without necessarily giving the appearance that they are encouraging or profiting from sports betting. At the same time, they may be able to attract what appears to be a very highly involved group that is also very willing to spend aggressively, thus increasing the value of advertising time and space.

Additionally, some fantasy sports platforms such as CBSSports.com provide a link in each league to "finances" where each participant can track what they owe and what they have won. Again, providing this simple feature may entice groups of highly engaged sports fans to use that particular platform. Not only are these websites anxious to grow the size of their consumer base, but the value of these consumers is even higher considering that their profile is highly desirable for potential advertisers.

References

Bernhard, B. J., & Eade, V. H. (2005). Gambling in a fantasy world: An exploratory study of rotisserie baseball games. UNLV Gaming Research and Review Journal, 9, 29-42. Retrieved from http://www.gamblib.org/catalogue/publication/682/

Doyle, C., & Yeh, B. T. (2009). Unlawful Internet Gambling Enforcement Act and its implementing regulations. Retrieved from http://ipmall.info/hosted_resources/crs/RS22749_091019.pdf

Drayer, J., Shapiro, S., Dwyer, B., Morse, A., & White, J. (2010). The effects of fantasy football participation on NFL consumption: A qualitative analysis. Sport Management Review, 13, 129-141.

Dwyer, B., & Drayer, J. (2010). Fantasy sport consumer segmentation: An investigation into the differing consumption modes of fantasy football participants. Sport Marketing Quarterly, 19, 207-216.

Dwyer, B., & Kim, Y. (2011). For love or money: Developing and validating a motivational scale for fantasy football participation. Journal of Sport Management, 25, 70-83.

Fan demographics among major North American sports leagues. (2010, June 9). SportsBusiness Daily, 185. Retrieved from http://www.sports businessdaily.com/Daily/Issues/2010/06/Issue-185/The-Back-Of-The-Book.aspx

Farquhar, L. K., & Meeds, R. (2007). Types of fantasy sports users and their motivations. Journal of Computer-Mediated Communication, 12, 1208-1228.

Fisher, E. (2008, November 17). Study: Fantasy players spend big. Street &Smith's SportsBusiness Journal, 11, 1-2. Retrieved from http://www.sportsbusinessjournal.com/article/60598

Forrest, D., & Simmons, R. (2008). Sentiment in the betting market on Spanish football. Applied Economics, 40, 119-126.

Frey, J. (1992). Gambling on sport: Policy issues. Journal of Gambling Studies, 8, 351-360.

Holleman, M. C. (2006). Fantasy football: Illegal gambling or legal game of skill? North Carolina Journal of Law & Technology, 8, 59-80.

Karg, A. J., & McDonald, H. (2011). Fantasy sport participation as a complement to traditional sport consumption. Sport Management Review, 14(4), 327-346. doi: 10.1016/j.smr.2010.11.004

Lee, W. Y., Kwak, D., Lim, C., Pedersen, P., & Miloch, K. (2010). Effects of personality and gender on fantasy sports game participation: The moderating role of perceived knowledge. Journal of Gambling Studies, 27(3), 427-441.

Levinson, M. (2006). Sure bet: Why New Jersey would benefit from legalized sports wagering. Sports Lawyers Journal, 13, 143-178.

Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior, 13, 103-123.

McKelvey, S. (2004a). U.S. professional sport organization policies shift to embrace legalized gambling entities: A roll of the dice? Journal of Legal Aspects of Sport, 14, 23-45.

McKelvey, S. (2004b). The growth in marketing alliances between US professional sport and legalised gambling entities: Are we putting sport consumers at risk? Sport Management Review, 7, 193-210.

Moorman, A. M. (2008). Fantasy sports leagues challenged as illegal gambling. Sport Marketing Quarterly, 17(4), 232-234.

Nelson, T. F., LaBrie, R. A., LaPlante, D. A., Stanton, M., Shaffer, H. J., & Wechsler, H. (2007). Sports betting and other gambling in athletes, fans, and other college students. Research Quarterly for Exercise and Sport, 78, 271-283.

Nesbit, T. M., & King, K. A. (2010a). The impact of fantasy football participation on NFL attendance. Atlantic Economic Journal, 38(1), 95-108.

Nesbit, T. M., & King, K. A. (2010b). The impact of fantasy sports on television viewership. Journal of Media Economics, 23(1), 24-41

Paul, R. J., & Weinbach, A. P. (2010). The determinants of betting volume for sports in North America: Evidence of sports betting as consumption in the NBA and NHL. International Journal of Sport Finance, 5(2), 128-140.

Petry, N. M. (2003). A comparison of treatment-seeking pathological gamblers based on preferred gambling activity. Addiction, 98, 645-655.

Wann, D. L., Grieve, F. G., Zapalac, R. K., & Pease, D. G. (2008). Motivational profiles of sport fans of different sports. Sport Marketing Quarterly, 17, 6-19.

Weiss, S. M., & Loubier, S. L. (2010). Gambling habits of athletes and nonathletes classified as disordered gamblers. Journal of Psychology, 144(6), 507-521.

Joseph E. Mahan III, PhD, is an assistant professor in Sport & Recreation Management at Temple University. His research focuses on sport consumer psychology.

Joris Drayer, PhD, is an assistant professor in Sport & Recreation Management at Temple University. His research interests include sport pricing and sport consumer behavior.

Emily Sparvero, PhD, is an assistant professor in Sport & Recreation Management at Temple University. Her research focuses on sport policy.

Table 1
Zero-Order Correlations Among Constructs of Interest.

Construct 1 2 3 4

TeamID --
SportsBet .07 *** --
Fantasy -.12 *** -.22 *** --
FantPay .04 (ns) -.21 *** -- (a) --

* p < .05; ***p < .001.

(a) Cannot be computed because at
least one of the variables is constant.

Table 2.
Results of t-tests of SportsBet, Fantasy, and FantPAY, on Team ID.

 Team ID

 M (a) SD df t

SportsBet
Yes 3.06 0.81 2246 3.16 **
No 2.94 0.79

Fantasy
Yes 3.10 0.81 2240 5.56 ***
No 2.91 0.78

FantPAY
Yes 3.08 0.79 2240 0.99
No 3.14 0.84

(a) Measure is on a five-point scale; higher
number signifies higher team identification.
** p < .01. *** p < .001.

Table 3.
Multivariate and Univariate Analyses of Covariance
for Team ID x SportsBet on Spending Variables.

 ANCOVA
 MANCOVA SpendTix SpendMerch
Source F F F

Income (a) 13.95 (b) *** 53.37 (d) *** 2.73 (d)
Team ID (ID) 7.53 (c) *** 0.36 (e) 23.21 (e) ***
SportsBet (SB) 7.29 (b) *** 12.28 (c) *** 10.60 (d) **
ID x SB 1.02 (c) 0.74 (e) 0.95 (e)

 SpendExtra SpendOther
Source F F

Income (a) 14.24 (d) *** 0.92 (d)
Team ID (ID) 1.73 (e) 0.18 (e)
SportsBet (SB) 27.79 (d) *** 0.64 (d)
ID x SB 0.26 (e) 0.38 (e)

Note. Wilks's F used for multivariate analysis.

(a) Covariate.
(b) df = 4, 1649. (c) df = 8, 3298.
(d) df = 1,1659 . (e) df = 2, 1659.
** p < .01. *** p < .001.

Table 4.
Multivariate and Univariate Analyses of
Covariance for MonthWager on Attitudes.

 ANCOVA

 MANCOVA BetInterest BetEnjoy EnjoyMore
Source F F F F

Income (a) 0.19 (b) 0.578 (d) 0.10 (d) 0.00 (d)
MonthWager 17.55 (c) *** 13.30 (e) *** 43.36 (e) *** 26.96 (e) ***

Note. Wilks's F used for multivariate analysis.

(a) Covariate.
(b) df = 3, 686. (c) df = 6, 1372.
(d) df = 1,692. (e) df = 2, 692.
*** p < .001.

Table 5.
Multivariate and Univariate Analyses of Covariance
for MonthWager on Spending Variables.

 ANCOVA

 MANCOVA SpendTix SpendMerch
Source F F F

Income (a) 5.52 (b) *** 19.69 (d) *** 0.04 (d)
MonthWager 5.20 (c) *** 8.69 (e) *** 14.00 (e) ***

 SpendExtra SpendOther
Source F F

Income (a) 6.50 (d) * 0.04 (d)
MonthWager 14.06 (e) *** 1.86 (e)

Note. Wilks's F used for multivariate analysis.

(a) Covariate.
(b) df = 4, 562. (c) df = 8, 1124.
(d) df = 1, 569. (e) df = 2, 569.
* p < .05. ***p < .001.

Table 6.
Multivariate and Univariate Analyses of Variance
for Team ID x Fantasy on Spending Variables.

 ANCOVA
 MANCOVA SpendTix SpendMerch
Source F F F

Incomea 5.81 (b) *** 56.20 (d) *** 3.39 (d)
Team ID (ID) 7.76 (c) *** 0.82 (d) 22.94 (d) ***
Fantasy (F) 5.81 (b) *** 20.50 (d) *** 2.19 (d)
ID x F 0.32 (c) 0.00 (e) 0.06 (e)

 SpendExtra SpendOther
Source F F

Incomea 15.55 (d) *** 0.96 (d)
Team ID (ID) 1.31 (d) 0.29 (d)
Fantasy (F) 16.83 (d) *** 0.35 (d)
ID x F 0.26 (e) 0.76 (e)

Note. Wilks's F used for multivariate analysis.

(a) Covariate.
(b) df = 4, 1644. (c) df = 8, 3288.
(d) df = 1,1654. (e) df = 2, 1654.
*** p < .001.

Table 7.
Multivariate and Univariate Analyses of Variance FantPay on Attitudes.

 ANCOVA
 MANCOVA FantInterest FantEnjoy FantMore
Source F F F F

Income (a) 1.89 (b) 0.02 (c) *** 0.07 (c) 5.48 (c) *
FantPay 11.42 (b) *** 23.38 (c) *** 21.84 (c) *** 16.38 (c) ***

Note. Wilks's F used for multivariate analysis.

(a) Covariate.
(b) df = 3, 852. (c) df = 1, 857.
* p < .05. *** p < .001.

Table 8.
Multivariate and Univariate Analyses of
Variance for FantPay on Spending Variables.

 ANCOVA
 MANCOVA SpendTix SpendMerch
Source F F F

Income (a) 9.43 (b) *** 36.37 (d) *** 8.56 (d) **
FantPay 5.70 (b) *** 12.60 (e) *** 1.94 (e)

 SpendExtra SpendOther
Source F F

Income (a) 10.72 (d) ** 1.77 (d)
FantPay 3.15 (e) 0.74 (e)

Note. Wilks's F used for multivariate analysis.

(a) Covariate.
(b) df = 4, 681. (c) df = 1, 687.
** p < .01. *** p < .001.