The Same or Different? Convergence of Skin Gambling and Other Gambling Among Children

New media and its associated technological infrastructure have created conditions in which forms of gambling can, and increasingly are, being incorporated into digital life and practice (Macey and Hamari 2018a; Griffiths et al. 2013). This is particularly true within video games, which incorporate relatively new and emerging practices that replicate and reproduce gambling-like activities within this media. These practices include loot boxes, where players pay to ‘open’ a virtual box in the hope of it containing in-game items of significantly higher value than their original outlay, or the gambling or betting of ‘skins’ (decorative in-game items) through various mediums (Macey and Hamari 2018a). There has been much consideration of the intersection between video game participation and engagement in other risky practices, including gambling (Macey and Hamari 2018b) and it suggested that video game engagement could serve as a gateway into other gambling activities, though as Macey and Hamari (2018b) point out, evidence on this is mixed.

These studies have tended to focus on the relationship between any form of video game play and gambling behaviour (McBride and Derevensky 2017). However, within video games, an increasing number of gambling-like activities are available, and less consideration has been given to the intersection of these ‘within game’ features with more traditional forms of gambling activity. These features, such as skin gambling/betting or the purchase of loot boxes, are facilitated by micro transactions within video games, whereby players pay to purchase in-game virtual items or access to certain game features. These transactions are an increasingly common and profitable part of the gaming ecosystem (ParentZone 2018; King and Delfabbro 2018). This then facilitates a range of other actions for players, such as the betting or trading of skins, mainly on third party websites. Skins are virtual items earned or purchased within video games, which have their own value within the gaming community. They are decorative items that have no bearing on the outcome of the game but are highly sought after nonetheless and first emerged in 2012 within the game Counter Strike: Global Offensive (ParentZone 2018). Skins are purchased from a digital marketplace and some skin items are more valuable than others, often based on rarity, popularity and potential use (ParentZone 2018; Gambling Commission 2017a). The value of these items, like ‘hard’ currency itself, gold or diamonds (or using seventeenth century examples, tulip bulbs) can fluctuate based on these features. Through third party access to digital marketplaces, where skins are bought and stored, skins can be bet or traded on other websites and thus the virtual value of the skin converted into real currency. These practices are examples of a common phenomenon within digital games, where a range of different actions and industries develop around and extend from the core game (Kerr 2006). With regards to skin betting and gambling, there is ambiguity around the nature of the practice, though the British Gambling Commission (the industry regulator) stated that they consider skins to have real world value and that betting of them represents a ‘money’s worth prize’ (Gambling Commission 2017a). Skins therefore function as a form of crypto-currency with their own value but can also be converted into ‘real world’ currency. This suggests that the betting of these items extends beyond game play and could be considered gambling conducted via processes and websites where there is no robust age verification in place and which are complex to regulate.

Concern about skin betting practices have been heightened with respect to children and the potential role they may play in shaping problematic gaming and/or gambling practices (Gainsbury et al. 2015; King and Delfabbro 2018). Described as convergence between gaming and gambling, three interlocking challenges have been considered: that these ‘convergent’ practices could prompt children to gamble more generally as a form of gateway activity, that engagement in this form of activity alone could be harmful and that these practices normalise gambling for a cohort of children (ParentZone 2018).

In Britain, as elsewhere, children are singled out for specific regulatory protections from gambling with legal age limits placed on most commercial forms. Nonetheless, it is estimated that 12% of children aged 11–15 have participated in some form of gambling activity in the past week, with over half of this activity being on commercial and (technically) legally restricted forms (Wardle 2018a). Furthermore, it is estimated that around 0.8% of children aged 11–15 in Great Britain experience problems with their gambling behaviour and early onset of gambling in childhood is a known risk factor for subsequent problems (Blinn-Pike et al. 2010; Forrest and McHale 2018). It is in this context that the challenges of these seemingly convergent digital practices are raised as they are viewed as providing the means for children to gamble and access gambling content. Politicians are giving this increasing attention, with questions asked by UK parliamentarians about the impact of skin betting on underage children (UK Parliament 2018).

The betting and gambling of skins is popular among children and young people. A recent survey of 13–18 years olds in Great Britain estimated that 10% had ever gambled or bet skins (ParentZone 2018). In 2017, report by the British Gambling Commission found that 11% of 11–16 years olds had ever bet skins, and that 4% had done so in the past week. This made skin gambling/betting as popular as playing on fruit/slot machines and more popular than most other ‘traditional’ forms of gambling activity (Gambling Commission 2017b). Yet to date, there has been (to the author’s knowledge) little empirical examination of the extent to which skin gambling among children is combined with other forms of gambling; empirical insight which is needed to explore the whether gambling and gaming are mutually reinforcing consumptive practices, and if so, to what extent.

Understanding the potential impact of engagement in skin gambling among children requires a greater consideration of children’s behaviours in order to map the basic epidemiology of practices. To date, notions of convergence between gambling-like activities and more traditional forms of gambling have tended to be examined by focusing on the products, with researchers noting the similarities of these practices, their common structural features and reward system mechanisms (King and Delfabbro 2018; McBride and Derevensky 2017). It is, however, vitally important to understand convergence of behaviours, especially if theories about one practice leading to another are to be better explored. This research uses nationally representative data of children aged 11–16 to explore this and to estimate: It is hypothesised that skin gambling and other forms ‘traditional’ forms of gambling will cluster together, given the similarities between the practices meaning that those who are interested in one form of practice are also likely to be interested in others (H1). It is also hypothesised that this clustering will be socially patterned, being more common among certain types of children, especially boys (H2) and those from more disadvantage backgrounds (H3). Finally, it is hypothesised that children who participate in skin gambling and other forms of gambling will display greater levels of at-risk or problem gambling (H4), as a function of their greater involvement with gambling and gambling-like activities more generally.

Both skin gambling/betting and gambling on other activities were relatively common among British children aged 11–16, despite some legal restrictions on participation. In Britain, participation on most forms of commercial gambling, including the National Lottery, is age prohibited yet many children still find ways to access these activities, with over half of children’s gambling activity estimated to be on age-restricted forms (Wardle 2018a, b). Playing video games is even more common among this age group and among boys, the gambling or betting of skins was the most prevalent form of ‘gambling’ activity. Evidence from this analysis shows that there is some overlap in who gambles or bets with skins and who takes part in other forms of gambling (confirming hypothesis 1), with 3% of children aged 11–16 saying that they did both. Whilst this may seem like a small number, this equates to around 100,000 children aged 11–16 in Britain. Furthermore, observed/expected ratios show that these two behaviours co-occur more than would be expected given their independent population prevalence, indicating greater overlap between these behaviours than is expected. Notably, the greatest level of overlap was between skin betting and gambling and other forms of online betting or gambling. This is perhaps unsurprising, given the common media underpinning these consumptions. This therefore supports the notion of a ‘convergence’ in behaviours among some children who are engaging in both activities. These patterns of behaviour ‘convergence’ were greatest for boys, older children and those who felt they were doing less well at school, confirming hypothesis 2. However, there was little evidence that this clustering of behaviour occurred disproportionately among those from more disadvantage backgrounds. This may be related to the measure (receipt of free school meals) used to proxy low income households.

However, the evidence is not unequivocal. The most common pattern among those who bet or gambled with skins was that they did not also engage in other forms of gambling. At younger age groups, children tended either to bet on skins or to gamble on other things, if they did this at all. Among older children, skin gambling/betting was more likely to be combined with gambling on other activities, though half of skin gamblers did this activity alone. This suggests a need for greater clarity when talking about processes of convergence between gambling and gaming. As Macey and Hamari (2018a) have highlighted, there is often a tendency with newly emerging consumptive practices to view them in silos rather than to situate them within the broader context of existing behaviours. This paper attempts to address this issue and suggests that there are four distinct groups of children: the majority who engage in neither skin gambling or other forms of gambling; a significant minority who gamble but do not bet with skins (which includes a disproportionate number of girls given their lesser propensity to play video games); a minority who only bet or gambled skins and a further minority who bet and gambled skins and gambled on other things. For the vast majority of children, these behaviours are not converging simply because do not engage in these practices; yet for a minority they are and these behaviours cluster.

Notably, rates of at-risk and problem gambling were highest among gamblers who also engaged in skin gambling/betting (confirming hypothesis 4). This is to be expected. By definition, those engaging in both skin gambling/betting and other forms of gambling have higher levels of gambling involvement because they both gambled on traditional forms of gambling and engaged in a similar practice within video games. However, it is also evident that this group were also more involved in ‘traditional’ forms of gambling alone, with the average number of traditional forms of gambling undertaken being higher among this group also. Involvement theory postulates that the more someone engages in gambling the more likely it is that that they will experience harm from that engagement. This is often explored using the number of gambling activities someone undertakes as a measure of their breadth of gambling engagement (LaPlante et al. 2014; Dixon et al. 2016). The results of the regression analysis showed that the relationship between skin gambling and at-risk gambling attenuated once involvement in a number of other forms of gambling was taken into account. This suggests that it is the combination of skin gambling with other forms that needs further consideration. Therefore, whilst children who gamble with skins as well as other forms of gambling should be considered a high-risk group for the attendant experience of harms, this is likely related to their broader gambling repertoires than their engagement in skin/gambling or betting alone.

Notions of convergence underpin much academic thought about the seemingly mutually reinforcing practices of gaming and gambling. This has tended to approach this issue through analysis of the product, with examples of gambling-like practices embedded within the video game eco-system heralded as examples of convergent of practice and activities. However, there is notable conceptual ambiguity around the demarcation of gambling and play (Caillios 1958; Juul 2003), with some theorists querying where gaming stops and gambling begins. It is therefore important to assess the extent to which these are shared consumptive practices among individuals. This is especially so with children who have been subject to much concern around these developments. Whilst this paper provides some evidence that these behaviour co-occur for some, it does not explore how and why this occurs or, indeed, what type of practice children believe skin gambling to be. Previous research on young people’s perceptions of gambling and gaming noted considerable ambiguity around how young people understand and define gambling activities (Korn 2005; Skinner et al. 2004). This ambiguity may arguably be heightened among children specifically because of the different values they attach to objects in lieu of access to monetary resources (Wardle 2018b). It is imperative, therefore, to understand how children themselves differentiate these consumptive practices and the meanings they attach to them.

Analyses presented are based on self-reported behaviours from a survey of school-aged children and inherits the attendant issues of this methodology. It is secondary analysis, meaning that the analysis presented is limited to the questions designed and funded by the original survey commissioners (for example, only four questions being asked about video games and skin gambling/betting). Only a very limited number of socio-economic characteristics were included in the original survey, limiting the extent to which it is possible to explore how behaviours vary among different types of children. This also limits the range of covariates available to include in the multivariate models and caution should be taken not to view these as models exploring the full range of factors associated with either skin gambling or at-risk gambling. They are presented to give greater descriptive insight into the relationships highlighted through the bi-variate analyses. The definition of skin gambling/betting used is broad and is likely to include private betting/gambling among peers as well as the betting and gambling of skins on third party websites. However, as the definition of gambling used in the survey also includes betting and gambling for money among peers, these are comparable. There is no data about the sequence of activities, only that they were undertaken at broadly the same time. The measure used to represent past month gambling is likely to slightly under-estimate gambling behaviour as frequency of gambling was only collected for six main forms of gambling and excluded less prevalent forms (for example, betting in a bookmakers). Finally, the digital world is fast moving and new products and practices emerge within a short pace of time. Whilst this data was collected in 2017, meaning it is relatively recent, it is possible that the digital landscape has changed in the intervening period.

Convergence of digital practices is often examined via the lens of the product, where consideration is given to how seemingly similar practices are transferred from one medium to another. Whilst theories about the demarcation between gambling and gaming may be contested, the assimilation of gambling cues within gaming practices and ambiguity about where gaming ends and gambling begins cannot be denied. When considering these issues, it is vital to understand how such conceptual ambiguity manifests in everyday consumption and practice. This paper has shown that, among children, whilst gambling and gaming behaviours do cluster, and do so more for some groups than others, there is also a sizeable majority of children who engage in neither activity or who do one but not the other. This paper also provides some evidence of co-occurring practices among children, especially those conducted through the same medium, where there is a high level of concordance between skin gambling/betting and online gambling. Children who engage in both skin gambling/betting and other forms of gambling should be considered an at-risk group for the experience of harms because of their heightened engagement in gambling and gambling-like activities.