Empirical Internet gambling research (1996–2008): Some further comments

I was interested to read the systematic search of peerreviewed Internet gambling research by Shaffer, Peller, LaPlante, Nelson and LaBrie (2010). Given the number of peer-reviewed papers that were omitted from their analysis, particularly ones from my own research unit, this article as it stands is a misrepresentation of the empirical work that has been carried out into Internet gambling. Given the relatively small number of empirical papers that were published between January 1996 and March 2008 (the cut-off point to be included in Shaffer et al.’s (2010) analysis), the number of omissions is relatively large. The review also gives the impression that the vast majority of published outputs from my research unit are commentaries; this is clearly not the case and it misrepresents my research in this area. It is evident that the two databases used to identify peer-reviewed papers on Internet gambling did not fulfil the authors’ objective. For instance, no empirically peer-reviewed papers that were published in either the International Journal of Mental Health and Addiction (e.g. Derevensky & Gupta, 2007; Griffiths & Barnes, 2008) or the Journal of Gambling Issues (e.g. Griffiths, 2001; Ialomiteanu & Adlaf, 2001, which also happen to be the first empirical peer-reviewed studies in the area) were included (Table I). The authors also cited the fact that their database search did not cite two gambling prevalence surveys that had data on Internet gambling because they did not have the key search terms required for inclusion in their review (i.e. LaBrie, Shaffer, LaPlante, & Wechsler, 2003; Welte, Barnes, Wieczorek, Tidwell, & Parker, 2002). However, the main findings of these two studies were still outlined. Given the authors had described the results relating to Internet gambling in these two prevalence studies, it was surprising to see that relevant data from both the peer-reviewed British adult (n1⁄4 9003) and adolescent (n1⁄4 8017) national gambling prevalence surveys were missing from the review (Griffiths & Wood, 2007; Wardle et al., 2007; Wood, Griffiths, Stevens, Bartlett, & Pye, 2006). Table I outlines 10 empirical studies on Internet gambling that were omitted from Shaffer et al.’s (2010) review, the majority being studies by myself and/or my colleagues (n1⁄4 7). The review also gives the impression that there is little place in the field for either commentaries or selfreport studies. This article presents a somewhat dismissive tone to any study that has not looked at actual behavioural data. Many commentaries bring particular issues to the fore, help set research agendas and generally ‘kick-start’ particular lines of empirical enquiry. Self-report methods, while clearly having methodological flaws, have some distinct advantages over tracking data. Behavioural data collected from only one gambling site tells us nothing about the person’s Internet gambling in general (as Internet gamblers typically gamble on more than one site) and it does not account for the fact that more than one person can use a particular account. Pure behavioural data tell us nothing about why people gamble and relationships between gambling and other behaviours (e.g. alcohol and tobacco use) and (at present) cannot examine problem gambling using current diagnostic criteria (Griffiths, Wardle, Orford, Sproston, & Erens, 2010a). It is only through self-report methods that this type of information can be found. The authors do mention some of the limitations of behavioural tracking data but they do not balance this out by pointing out some of the advantages of self-report over behavioural data (Griffiths, 2010; Wood & Griffiths, 2007a, 2007b).