Investigating relationships between video gaming, spectating esports, and gambling

1 Abstract An established body of research exists in which playing video games have been associated with potentially problematic behaviours, such as gambling. However, this position has recently been questioned in respect to the contemporary environments of both video gaming and gambling. This study investigates relationships between a range of gambling activities and the consumption of video games in general, and the newly emergent phenomenon of esports in particular. In addition, these practices are considered in relation to established measures assessing game addiction and problematic gambling. The study employs Partial Least Squares modelling to investigate data gathered via an international online survey (N=613). Video game addiction was found to be negatively associated with offline gambling, online gambling, and problem gambling. Video game consumption had only small, positive association with video game-related gambling and problem gambling. Consumption of esports had small to moderate association with video game-related gambling, online gambling, and problem gambling. The primary finding of this study are that contemporary video games are not, in themselves, associated with increased potential for problematic gambling, indeed, the position that problem gaming and problem gambling are fundamentally connected is questioned. Keywords Video Games, Esports, Gambling, Problematic Gambling, Game Addiction, Convergence This is the post print version of the article Macey, J. & Hamari, J. (2018) Investigating relationships between video gaming, spectating esports, and gambling. Computers in Human Behavior 80 (March), 344-353. https://doi.org/10.1016/j.chb.2017.11.027 2 1. Introduction There is an established body of research which addresses the potential associations between playing video games and a range of problematic behaviours, from aggressive or violent behaviour (Anderson et al., 2010; Olson et al., 2009) to substance abuse (Desai, Krishnan-Sarin, Cavallo, & Potenza, 2010; Williams, Yee, & Caplan, 2008). The relationship between video gaming and gambling is an aspect which has continuously received a significant amount of attention; the case has been made that gaming may serve as a pathway that increases the likelihood of developing problematic gambling behaviours. This position is one in which the structural similarities between gaming and electronic gambling are cited as a major influence (Fisher & Griffiths, 1995; Johansson & Götestam, 2004; Wood, Gupta, Derevensky, & Griffiths, 2004), as are the social benefits accrued for successful players (Griffiths & Wood, 2000), and misperceptions related to a sense of control (Gupta & Derevensky, 1996). The concept of structural similarities between video gaming and gambling was first discussed almost three decades ago (Griffiths, 1991) and continues to be highly influential to this day (McBride & Derevensky, 2017). However, the focus of the original research was on coinoperated arcade games and gambling using slot machines (Griffiths, 1991; King, Delfabbro, & Griffiths, 2010a). It has been argued that, as both video games and gambling environments have undergone significant changes, the findings are no longer applicable to the contemporary practices of gaming and gambling (Forrest, King, & Delfabbro, 2016). Driven primarily by online technologies (King, Delfabbro, & Griffiths, 2010b; King, Delfabbro, Kaptsis, & Zwaans, 2014) the convergence of gaming and gambling has taken on new forms (Lopez-Gonzalez & Griffiths, 2016). Technological developments have not simply changed the content of games, offering sophisticated immersive environments for example, but more 3 significantly they have changed the way that games are played. The spaces of play have dispersed, no longer centralised in arcades or the home, player-versus-machine has become player-versusplayer via networked sessions. Business models such as “free-to-play” (Alha, Koskinen, Paavilainen, Hamari, & Kinnunen, 2014; Hamari, Hanner, & Koivisto, 2017) and social network games have introduced gambling-like mechanics back into video games. In addition, the expansion of virtual economies and goods (Hamari, Alha, Järvelä, Kivikangas, Koivisto, & Paavilainen, 2017; Hamari & Keronen, 2017; Lehdonvirta & Castronova, 2014) has obfuscated the use of real money for gambling-like activities in games; gambling-like mechanics are no longer easily identifiable for users (King et al., 2014; Gainsbury, Russell, King, Delfabbro, & Hing, 2016; Kim, Wohl, Salmon, Gupta, & Derevensky, 2015). All this is happening in an environment where the presence of games and game-like experiences is ubiquitous (Raessens, 2006; Hamari, Huotari, & Tolvanen, 2015), one which has seen an increasing trend toward the liberalisation of gambling laws (Kingma, 2006; Fong, Fong, & Li, 2011; Markham & Young, 2015), and increased access to gambling activities via the internet and mobile devices (Choliz, 2016; Deans, Thomas, Daube, & Derevensky, 2016). The phenomenon which most succinctly encapsulates these trends is that of esports; a form of sports where play is “facilitated by electronic systems”, i.e. competitive video gaming organised into leagues and tournaments (Hamari & Sjöblom, 2017; Taylor, 2012). In esports, video games are the objects and the drivers of all activity, its “sportification” (Lopez-Gonzalez & Griffiths, 2016) has brought with it a host of activities associated with traditional sports: professionalization, regulation, fan communities, and gambling. In addition to esports (Holden, Rodenberg, & Kaburakis, 2016), the convergence of gaming and gambling is evident in social gaming (Gainsbury, King, Abarbanel, Delfabbro, & Hing 2015; King et al., 2014) and the free-play modes offered by online casinos (Bednarz, Delfabbro, & King, 2013). It is understandable, therefore, that concerns have been raised over the potential for 4 video game players to be exposed to factors which may encourage problematic gambling (Bednarz et al., 2013; Griffiths, King, & Delfabbro, 2009; Parker, Taylor, Eastabrook, Schell, & Wood, 2008). Results have been mixed, with some studies showing a significant relationship between playing video games and increased participation in gambling (Wood et al., 2004; Gainsbury et al., 2016; Kim et al., 2015; McBride & Derevensky, 2017) while others have not found a clear relationship (Delfabbro, King, Lambos, & Puglies, 2009; King, Ejova, & Delfabbro, 2012; Forrest et al., 2016). 1.1. Aims of the Research An environment has developed in which the prevalence of both video gaming and gambling as leisure activities has been accompanied by technological and cultural convergence, increased ease of access, and liberalisation of gambling regulations. As such, it is imperative that relationships between video gaming and gambling require continued investigation, with specific focus on newly emergent phenomena such as esports. Currently, there is a dearth of research which addresses esports and gambling, what published work there is largely addresses the question of legal and regulatory issues (Schneider, 2015; Owens Jr, 2016). This deficit requires urgent attention as industry analysts predict the number of global esports viewers to reach 375 million by the end of 2017, with active participants in formal, mainstream esports gambling already exceeding 2.25 million. Furthermore, it is estimated that over 3 million people actively participate in the informal markets surrounding in-game items, such as skins lotteries (Grove & Krejcik, 2015). This study, therefore, seeks to investigate relationships between the consumption of video games, esports and three different forms of gambling: offline, online, and video game-related gambling. The final category includes activities such as: betting on esports matches, playing fantasy 5 esports, paying to access randomly generated in-game items, using in-game items or currencies as wagers in third-party gambling sites, and social network gambling games. Both offline and online gambling refer to established practices such as betting, playing the lottery etc., in specific contexts. These factors give rise to the following research questions: x RQ1: Is increased consumption of video games and esports associated with increased levels of gambling? x RQ2: Are higher rates of problematic video gaming associated with higher rates of a) gambling activity, and b) problematic gambling? The convergence of gaming and sports embodied by esports suggests that if video gaming is associated with increased gambling, it would be in this environment that any relationships would be most pronounced. Therefore, in order to investigate the stated research questions, the following target population was identified: video game players who also watched esports, and/or who had gambled within the previous 12 months. Online questionnaires with selfselected respondents are considered the most appropriate method of obtaining data from such populations (Griffiths, 2010). Advantages of this method include: increased access to target population, global reach, it is more cost-efficient than traditional random sampling techniques, and responses are less likely to be affected by the desire for social acceptance. 1.2. Research Model This research is concerned with the relationship between consumption of digital media, in the form of video games and esports, and gambling behaviour, as such an involvement model (Binde, 2013) was developed to answer the research questions detailed above. 6 Previous research has linked increased consumption of video games to increased participation in gambling and raised likelihood of developing problematic gambling behaviours (Wood et al., 2004; McBride & Derevensky, 2017). This relationship has been explained in terms of structural similarities between gaming and gambling (Fisher & Griffiths, 1995; Johansson & Götestam, 2004), the accrual of social capital (Griffiths & Wood, 2000), and maladapted cognitions such as an overdeveloped sense of control (Gupta & Derevensky, 1996). Therefore, it is hypothesised that Video gaming habits will be positively associated with Offline Gambling Habits (H1), Online Gambling Habits (H2), and Video Game-Related Gambling Habits (H3). The association is expected to be most pronounced in relation to Video Game-Related Gambling Habits and weakest for Offline Gambling Habits. An interest in esports is born out of an initial interest in video games, esports being considered a subset of the wider gaming environment (Lee & Schoenstedt, 2011), therefore, Esports viewing habits is anticipated to be an additional predictor of involvement in gambling, one which is itself influenced positively by Video Gaming Habits (H4). Accordingly, Esports Viewing Habits is hypothesised as positively influencing Offline Gambling Habits (H5), Online Gambling Habits (H6), and Video Game-Related Gambling Habits (H7). The association is expected to be strongest for Video Game-Related Gambling Habits and weakest for Offline Gambling Habits. Problematic gaming behaviour in particular has been theorised as being associated with problematic gambling (Griffiths & Wood, 2000; Johansson & Götestam, 2004; Parker et al., 2008). It is a logical expectation that game addiction (GAS) is positively influenced by the habits surrounding the consumption of video games (H8) and, by extension, esports (H9). It has been interpreted as such, and used in the same way, in previous research (Forrest et al., 2016). Mirroring the relationships outlined in H1-H3 and H5-H7, GAS is expected to show positive associations with Offline Gambling Habits (H10), Online Gambling Habits (H11), and Video Game-Related Gambling Habits (H12). 7 Video game-related gambling is the newest form of gambling (heavily dependent upon technological developments that have facilitated contemporary video game forms, business models and online communities (Scholz, 2011; Taylor, 2012). Therefore, it is anticipated that it will be influenced by gambling habits of pre-existing formats, both Offline Gambling Habits (H13), and Online Gambling Habits (H14), with the former being weaker than the latter. Including a measure of problematic gambling when investigating possible relationships between video game consumption and gambling behaviour has been recommended by researchers in the field (Forrest et al., 2016). It is noteworthy that problematic gambling has been found to be more strongly associated with online gambling than offline (Griffiths, Wardle, Orford, Sproston, & Erens, 2009; Olason et al., 2011). As a result, problem gambling (PGSI) is expected to be positively influenced by all types of gambling behaviour, with the strongest associations predicted to be for Video Game-Related Gambling Habits (H15) and Online Gambling Habits (H16), and weakest in relation to Offline Gambling Habits (H17). The path model used to investigate relationships between the consumption of both video games and esports and gambling activities is presented in figure 1: Figure 1: Path Model. Model showing relationships between video game consumption and gambling activity. 8 2. Methods The survey included two measures of problematic behaviour, the Game Addiction Scale (GAS; Lemmens & Valkenburg, 2009) and the Problem Gambling Severity Index (PGSI), derived from the Canadian Problem Gambling Index (CPGI; Ferris & Wynne, 2001). Playing video games, watching esports, and gambling habits were assessed using items which measured frequency, average weekly hours, and average monthly spend (in US$) for each activity. 2.1. Participants and Procedure A sample of 869 video gamers was collected, from a total of 2,397 responses, via an online survey publicised across social media channels and online discussion forums dedicated to video gaming and esports. The survey was available for a period of one month, between November and December 2016. As an incentive to participate, valid respondents were entered into a prize draw to win a $50 gift-card. A filter question was included, those that failed were excluded from the sample, also excluded were those who reported playing no video games within the previous 12 months. The final sample consisted of 613 respondents, 25.57% of total responses, of which: the modal range was 18 – 21 (31.5%) (table 1); 98.2% played video games once a week or more (table 2); 50.1% watched esports once a week or more (table 2); 32.8% gambled offline within the previous 12 months, 34.4% had gambled online and 47.5% had gambled in relation to video games (table 3); 91.4% were male (table 4), a figure also reflected in previous studies of both active esports players (Weiss & Schiele, 2013) and internet gamblers (Gainsbury, Wood, Russell, Hing, & Blaszczynski, 2012). 9 Age Ranges of Sample (n=613) n % Cumulative % Information Not Provided 11 1.8 1.8 14 or Under 11 1.8 3.6 15 - 17 152 24.8 28.4 18 - 21 193 31.5 59.9 22 - 25 104 17.0 76.8 26 - 29 71 11.6 88.4 30 - 33 32 5.2 93.6 34 - 37 12 2.0 95.6 38 - 41 12 2.0 97.6 42 - 45 11 1.8 99.3 46 - 49 2 .3 99.7 50 or Over 2 .3 100.0 Total 613 100.0 Table 1: Demographics – Age Video Game and esports Consumption Frequencies of Sample (n=613) Video Game Play Frequency Esports Viewing Frequency n % Cumulative % n % Cumulative % Never - - - 79 12.9 12.9 Less Than Once a Month 2 .3 .3 83 13.5 26.4 About Once a Month 4 .7 1.0 50 8.2 34.6 2 - 3 Times a Month 5 .8 1.8 94 15.3 49.9 About Once a Week 11 1.8 3.6 96 15.7 65.6 2 - 6 Times a Week 176 28.7 32.3 145 23.7 89.2 Every Day 415 67.7 100.0 66 10.8 100.0 Total 613 100.0 613 100.0 Table 2: combined video game and esports consumption frequencies Gambling Participation Rates of Sample (n=613) Offline Online Video Game-Related Gambled in last 12 months? n % n % n % Yes 201 32.8 211 31.4 291 48.5 No 412 67.2 402 68.6 322 52.5 Total 613 100 613 100 613 100 Table 3: gambling participation in last 12 months 16 of participation in offline gambling. A similar relationship is in evidence between GAS and online gambling habits. The amount of variance in online gambling habits explained by the model is higher than that of offline gambling habits, but is still very small (R2 = .058). Together these results suggest that video gaming in itself does not have any significant relationship to established gambling practices. The strong relationship between online gambling and video game-related gambling is unsurprising, however, the degree of this relationship is unanticipated. Indeed, online gambling seems to be the biggest predictor of video game-related gambling, over and above either consuming video games or watching esports. That said, the total effect value of watching esports on video game-related gambling is almost double the direct effect, (β = .296), the relationship between the two is, therefore, a strong one. The model explains 25% of the variance of PGSI, approaching the 26% required for the effect to be considered large (Cohen, Cohen, West, & Aiken, 2013). The direct effects of video game-related gambling on PGSI are substantially more than those of online gambling habits. Only 7.8% of GAS was explained, however, a clear and strong relationship with game consumption habits is evident. While the initial assumptions were not that video game consumption habits would entirely explain GAS, a more substantial overall effect was expected. Considering RQ1, the situation appears to be more nuanced than expected as, despite the fact that the consumption of video games is a predictor of esports viewing habits, their individual relationships with different gambling activities vary somewhat. Both Video Gaming Habits and Esports Viewing Habits have statistically significant relationships with video gamerelated gambling. However, only Esports Viewing Habits shows any other statistically significant 17 relationships, with Online Gambling Habits, and that is, somewhat surprisingly, stronger than with video game-related gambling. We can say, therefore, that the consumption of esports is associated with increased gambling in mediated contexts (via video games and the internet) but not with offline gambling. The situation in respect to the consumption of video games is, however, more ambiguous, with only a small association shown to exist with video game-related gambling. The situation in regard to RQ2 is more emphatic as the model shows that problematic video gaming is not associated with higher rates of either gambling activity or problematic gambling. In fact, it appears that higher rates of problematic gaming, as measured using GAS, seem to act against involvement in both online and offline gambling, and for the development of problematic gambling behaviour. 4. Discussion The main findings of this research are that: first, there are no strong associations between the consumption of video games or esports, and gambling activity; and second, that problematic video gaming has a small, but significant, negative association with both gambling in general, and problematic gambling in particular. These, and other issues of interest arising from the study, are discussed below. This research theorised that an interest in esports is born partly out of a pre-existing interest in, and consumption of video games, however, the small amount of variance of esports consumption explained by gaming habits suggests that this view is too simplistic, although there is evidence of a fairly strong relationship between the two. An almost identical relationship seems to exist between video gaming and GAS scores. This is particularly significant as researchers in the field of addiction studies have often used either frequency of gaming or time spent gaming as a 18 primary indicator of addictive behaviour (Van Rooij, Schoenmakers, Vermulst, Van Den Eijnden, & Van De Mheen, 2011; Lemola et al., 2011; Weinstein, 2010; Festl, Scharkow, & Quandt, 2013). This research shows such an approach to be overly simplistic; even using a combination of consumption measures proves to be a poor indicator of potentially problematic gaming. This is clear evidence that problematic video game playing differs from other conditions for which consumption measures are a good indicator of addictive behaviour (Sassen et al., 2011; Rehm et al., 2013). The small amount of variance in online and offline gambling habits explained by the range of game-related measures is at odds with the stated expectations of this research. This, and the fact that the model was unable to find statistically significant relationships between video game consumption and gambling activities not related to video games, is in contrast to a large body of work (Wood et al., 2004; Gainsbury et al., 2016; Kim et al., 2015; McBride & Derevensky, 2016). Instead, it provides support for research which questions proposed links between the practices of gaming and gambling (Delfabbro et al., 2009; King et al., 2012; Forrest et al., 2016). Indeed, the negative relationship between game addiction score and both online and offline gambling suggests that problematic gaming and problematic gambling are clearly distinct from one another. It seems instead that those who score more highly on measures of game addiction are unlikely to migrate to gambling behaviours, despite the apparent structural similarities (Johansson & Götestam, 2004; Wood et al., 2004; McBride & Derevensky, 2016). Critics of this position might reasonably argue that the negative correlation between game addiction score and gambling habits is due to limited resources; people are unable to participate in both activities concurrently, and if the opportunity to play video games were removed, they would be highly likely to seek similar gratifications from gambling. However, the results of this research refute such an argument as the overall, mediated relationship between game addiction score and PGSI is both negative and statistically significant. Furthermore, video game habits show statistically significant positive relationships with both game addiction score and PGSI, while the game addiction score has a negative relationship with PGSI. 19 There appears to be, therefore, an aspect of video gaming itself which serves to prevent the uptake of gambling and the development of associated problematic behaviours (Forrest et al., 2016). Problematic gambling has a moderate to large amount of variance (25%) explained solely by behaviours relating to the consumption of gambling, (frequency, hours spent gambling per week and money spent gambling per month). This is noteworthy when compared to the low amount of variance (7.8%) for problematic gaming using the same measures, suggesting that the nature of problematic gaming is distinct from other behavioural conditions. As such, the findings support the call for specific measurement tools to be developed rather than using those derived mainly from substance use disorders (Petry, 2013; Kardfelt-Winther, 2015; King & Delfabbro, 2016; Demetrovics & Király, 2016). Although video game-related gambling habits have a more pronounced direct effect on PGSI than online gambling habits, the total, mediated, effects are comparable in strength. It is likely that the similarity of the overall influence of these two forms of gambling can be attributed to the fact that video game gambling is almost exclusively facilitated via online media. Aspects of online gambling such as increased ease of access, anonymity and the use of digitised/virtual currencies have been identified as characteristics of online gambling which facilitate problematic behaviours (Derevensky & Gupta, 2007; Gainsbury, Hing, Delfabbro & King, 2014; LopezGonzalez & Griffiths, 2016). Further evidence of the strong relationship between online gambling and video gamerelated gambling can be found in the high β value between the two. Although the development of the model theorised that causality to run in a certain direction, it cannot be proved as this study is correlational in nature. It may be that those who are interested in games and who also gamble online are likely to then begin gambling in relation to video games, alternatively it may be that those who develop an interest in video game gambling then go on to explore other forms of gambling in the online environment. 20 A somewhat surprising finding was that video gaming habits had a reasonably small correlation with video game-related gambling, indeed it was the smallest of all observed relationships, both in direct and mediated effects. Of those variables related to video gaming in general, it was the consumption of esports that displayed the strongest relationship to video gamerelated gambling. Furthermore, it was the only measure that had any statistically significant, positive, relationship with either online or offline gambling. It seems clear, therefore, that rather than playing video games, it is the consumption of esports that is a more significant predictor of increased participation in gambling. Whether this is due to any specific characteristics of esports itself, or if gambling is associated with esports in the same way that it is with traditional sports (Hill & Clark, 2001; Udovicic, 1998) is something that requires further investigation. 4.1. Implications The first notable implication of this research is that the use of gaming frequency or time spent gaming as a shorthand for addictive behaviour is over-simplistic and inaccurate. Therefore, researchers and professionals in the field of addictive behaviours must utilise more robust measures in order to minimise the risk of misdiagnosis. A further lesson is that different approaches are required to understand and address problematic gaming and problematic gambling, ones which are based more on the individual circumstances and characteristics of each activity. Therefore, the approach whereby problem gaming is understood through the lens of gambling is questionable and likely to be ineffective. As such, criteria for assessing problematic gaming which have been developed from those based on problematic gambling or Substance Use Disorder, require a thorough overhaul. Finally, the role and effect of esports, rather than video gaming per se, should be taken into consideration when evaluating the potential to develop problematic gambling behaviours. And, 21 consequently, particular attention should be paid to this context when developing therapeutic approaches or treatment programmes. 4.2. Limitations This research incorporated the lessons of previous studies by utilising more robust measures for consumption than simply using frequency of gaming, and by including a measure of problematic gambling (Forrest et al., 2016). Nevertheless, it remains the case that the most notable limitation of this research was the collection of data via an online survey, as such it is open to the standard criticisms directed at self-selected samples. The benefits of this approach, however, include greater access to the target population than techniques employed in probability sampling, and reduced scope for responses to be guided by social acceptance or feelings of embarrassment. This is especially pertinent in relation to potentially sensitive topics such as gambling or problematic gaming (Griffiths, 2010). A potential issue specific to this particular survey was the seeming lack of diversity in respondents, with only 6.2% of participants being female. Whilst this is similar to other research (Weiss & Schiele, 2013; Gainsbury et al., 2012; Sjöblom, Törhönen, Hamari, & Macey, 2017) it is significantly lower than estimated levels of female participants in either video gaming, 41% (ESA, 2016), or watching esports casually, 36% (EEDAR, 2015). The characteristics of this dataset may be the result of the channels by which the data was collected; the most significant source of respondents was Reddit, a social news and discussion website which has previously been criticised for its lack of diversity (Speed, 2015; Zuckerman, 2012). An additional reason for the disparity between male and female respondents may be that this research is concerned with the relationship between video gaming and gambling, the latter is a pastime in which male participation largely outweighs female, and in which males favour sports betting, casino games and internet gambling in 22 general (Gainsbury et al., 2012; McDaniel & Zuckerman, 2003; Welte, Barnes, Wieczorek, Tidwell, & Parker, 2002; Gupta & Derevensky, 1998; Hing & Breen, 2001) all of which are the predominant forms of gambling associated with video games. 5. Conclusions Increased consumption of video games has a positive association with both game addiction score and video game-related gambling. However, as game addiction score has a negative correlation with both video game-related gambling and PGSI category there exists an unidentified aspect of video game play which serves to reduce the appeal of gambling for heavy gamers. On the other hand, increased consumption of esports is strongly associated with increased participation in online and video game -related gambling and moderately associated with increased potential for problematic gambling behaviour. The findings of this study are that modern video games do not, in themselves, act as developmental pathways to gambling. Furthermore, they question the claims that problem gaming and problem gambling are fundamentally connected. Instead, it seems that video games are simply a vehicle, like many other activities, employed to fulfil particular needs derived from the activity of gambling. The “sportification” (Lopez-Gonzalez & Griffiths, 2016) of video games, in the form of esports, is just one example of the way in which it is the convergence of digital culture, rather than video games themselves, that facilitates gambling. 6. Declaration The research has been carried out as part of research project (40009/16) funded by the Finnish Funding Agency for Innovation (TEKES). In addition, this research was supported by a 23 grant from The Finnish Foundation for Alcohol Studies. The authors wish to thank both organisations for their support. At no point were any of the funding bodies, or project partners involved in: the design of the study; data collection, analysis, or interpretation; the writing of the report; or decisions relating to submissions. 7. References Alha, K., Koskinen, E., Paavilainen, J., Hamari, J., & Kinnunen, J. (2014). Free-to-play games: Professionals’ perspectives. Proceedings of Nordic Digra, 2014 Anderson, C. A., Shibuya, A., Ihori, N., Swing, E. L., Bushman, B. J., Sakamoto, A., . . . Saleem, M. (2010). Violent video game effects on aggression, empathy, and prosocial behavior in eastern and western countries: A meta-analytic review.no title. Psychological Bulletin, 136(2), 151-173. Bednarz, J., Delfabbro, P., & King, D. (2013). Practice makes poorer: Practice gambling modes and their effects on real-play in simulated roulette. International Journal of Mental Health and Addiction, 11(3), 381-395. Binde, P. (2013). Why people gamble: A model with five motivational dimensions. International Gambling Studies, 13(1), 81-97. Chin, W. W., Marcolin, B. L., & Newsted, P. R. (2003). A partial least squares latent variable modeling approach for measuring interaction effects: Results from a monte carlo simulation study and an electronic-mail emotion/adoption study. Information Systems Research, 14(2), 189-217. Choliz, M. (2016). The challenge of online gambling: The effect of legalization on the increase in online gambling addiction. Journal of Gambling Studies, 32(2), 749-756. 24 Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2013). Applied multiple regression/correlation analysis for the behavioral sciences Routledge. Deans, E. G., Thomas, S. L., Daube, M., & Derevensky, J. (2016). “I can sit on the beach and punt through my mobile phone”: The influence of physical and online environments on the gambling risk behaviours of young men. Social Science & Medicine, 166, 110-119. Delfabbro, P., King, D., Lambos, C., & Puglies, S. (2009). Is video-game playing a risk factor for pathological gambling in Australian adolescents? Journal of Gambling Studies, 25(3), 391-405. Demetrovics, Z., & Kiraly, O. (2016). Commentary on Baggio et al. (2016): Internet/gaming addiction is more than heavy use over time. Addiction, 111(3), 523-524. Derevensky, J. L., & Gupta, R. (2007). Internet gambling amongst adolescents: A growing concern. International Journal of Mental Health and Addiction, 5(2), 93-101. Desai, R. A., Krishnan-Sarin, S., Cavallo, D., & Potenza, M. N. (2010). Video-gaming among high school students: Health correlates, gender differences, and problematic gaming. Pediatrics, 2706. Diamantopoulos, A., & Winklhofer, H. M. (2001). Index construction with formative indicators: An alternative to scale development. Journal of marketing research, 38(2), 269-277. Diamantopoulos, A., & Siguaw, J. A. (2006). Formative versus reflective indicators in organizational measure development: A comparison and empirical illustration. British Journal of Management, 17(4), 263-282. Esports consumer analysis whitepaper: EEDAR 2015 syndicated report. (2015). Electronic Entertainment Design And Research. Essential facts about the computer and video game industry: 2016 sales, demographic and usage data. (2016). Entertainment Software Association. 25 Ferris, J. A., & Wynne, H. J. (2001). The Canadian problem gambling index Canadian Centre on Substance Abuse Ottawa, ON. Festl, R., Scharkow, M., & Quandt, T. (2013). Problematic computer game use among adolescents, younger and older adults. Addiction, 108(3), 592-599. Fisher, S., & Griffiths, M. (1995). Current trends in slot machine gambling: Research and policy issues. Journal of Gambling Studies, 11(3), 239-247. Fong, D. K., Fong, H. N., & Li, S. Z. (2011). The social cost of gambling in Macao: Before and after the liberalisation of the gaming industry. International Gambling Studies, 11(01), 43-56. Forrest, C. J., King, D. L., & Delfabbro, P. H. (2016). The gambling preferences and behaviors of a community sample of Australian regular video game players. Journal of Gambling Studies, 32(2), 409-420. Gainsbury, S., Hing, N., Delfabbro, P. H., & King, D. L. (2014). A taxonomy of gambling and casino games via social media and online technologies. International Gambling Studies, 14(2), 196-213. Gainsbury, S. M., King, D. L., Abarbanel, B., Delfabbro, P., & Hing, N. (2015). Convergence of gambling and gaming in digital media. Gainsbury, S. M., Russell, A. M., King, D. L., Delfabbro, P., & Hing, N. (2016). Migration from social casino games to gambling: Motivations and characteristics of gamers who gamble. Computers in Human Behavior, 63, 59-67. Gainsbury, S., Wood, R., Russell, A., Hing, N., & Blaszczynski, A. (2012). A digital revolution: Comparison of demographic profiles, attitudes and gambling behavior of internet and noninternet gamblers. Computers in Human Behavior, 28(4), 1388-1398. 32 Appendix A: Outer Loadings Outer Loadings 95% CI β t p Lower Upper GAS Sum <- GAS 1.000 1.000 1.000 Off. Gam. Freq. -> Off. Gamb. 0.974*** 19.038 <0.001 0.809 0.999 Off. Gam. Spend -> Off. Gamb. 0.833*** 9.838 <0.001 0.620 0.944 Off. Gam. Hours -> Off. Gamb. 0.877*** 8.542 <0.001 0.595 0.986 On. Gam. Freq. -> On. Gamb. 0.976*** 49.151 <0.001 0.920 0.996 On. Gam. Spend -> On. Gamb. 0.878*** 21.235 <0.001 0.785 0.945 On. Gam. Hours -> On. Gamb. 0.893*** 21.477 <0.001 0.799 0.960 PGSI Group <- PGSI 1.000 1.000 1.000 VG Gam. Freq. -> VG Gamb. 0.923*** 33.091 <0.001 0.856 0.966 VG Gam. Spend -> VG Gamb. 0.895*** 25.825 <0.001 0.812 0.948 VG Gam. Hours -> VG Gamb. 0.923*** 30.332 <0.001 0.856 0.974 VG Play Freq. -> VG Habits 0.688*** 7.568 <0.001 0.462 0.821 VG Play Spend -> VG Habits 0.472*** 3.505 <0.001 0.218 0.742 VG Play Hours -> VG Habits 0.948*** 21.437 <0.001 0.822 0.990 Esp. Watch Freq. -> Esp. Habits 0.836*** 11.127 <0.001 0.650 0.942 Esp. Watch Spend -> Esp. Habits 0.653*** 4.881 <0.001 0.366 0.880 Esp. Watch Hours -> Esp. Habits 0.916*** 15.979 <0.001 0.761 0.980 GAS Sum/GAS = score for Game Addiction Scale. PGSI Group/PGSI = Problem Gambling Severity Index Categorisation. Off. Gam. = Offline Gambling Habits. On. Gam. = Online Gambling Habits. VG Gamb . = Video Game-Related Gambling Habits. VG Habits = Video Game Playing Habits. Esp. Ha bits = Esports Watching Habits. Off. Gam. = Off. Gambling. On. Gam. = Online Gambling. VG Gam. = Video Game - Related Gambling. VG Play = Video Game Playing Habits. E sp. Watch = Esports Watching Habits. Freq. = Frequency (of activity). Spend = Average Monthly Spend (on activity, in US$). Hours = Average Weekly Hours (on activity). Appendix B: Outer VIF Values Outer VIF Values VIF Off. Gam. Freq. 2.69 Off. Gam. Spend 3.658 Off. Gam. Hours 3.426 On. Gam. Freq. 3.125 On. Gam. Spend 3.552 On. Gam. Hours 3.81 VG Gam. Freq. 2.934 VG Gam. Spend 2.757 VG Gam. Hours 2.858 VG Play Freq. 1.334 33 VG Play Spend 1.091 VG Play Hours 1.415 Esp. Watch Freq. 2.17 Esp. Watch Spend 1.177 Esp. Watch Hours 2.152 Off. Gam. = Off. Gambling. On. Gam. = Online Gambling. VG Gam. = Video Game - Related Gambling. VG Play = Video Game Playing Habits. Esp. Watch = Esports Watching Habits. Freq. = Frequency (of activity). Spend = Average Monthly Spend (on activity, in US$). Hours = Average Weekly Hours (on activity).