Using cross-game behavioral markers for early identification of high-risk internet gamblers.

Abstract

Using actual gambling behavior provides the opportunity to develop behavioral markers that operators can use to predict the development of gambling-related problems among their subscribers. Participants were 4,056 Internet gamblers who subscribed to the Internet betting service provider bwin.party. Half of this sample included multiple platform gamblers who were identified by bwin.party's Responsible Gambling (RG) program; the other half were controls randomly selected from those who had the same first deposit date. Using the daily aggregated Internet betting transactions for gamblers' first 31 calendar days of online betting activities at bwin.party, we employed a 2-step analytic strategy: (a) applying an exploratory chi-squared automatic interaction detection (CHAID) decision tree method to identify characteristics that distinguished a subgroup of high-risk Internet gamblers from the rest of the sample, and (b) conducting a confirmatory analysis of those characteristics among an independent validation sample. This analysis identified two high-risk groups (i.e., groups in which 90% of the members were identified by bwin.party's RG program): Group 1 engaged in three or more gambling activities and evidenced high wager variability on casino-type games; Group 2 engaged in two different gambling activities and evidenced high variability for live action wagers. This analysis advances an ongoing research program to identify potentially problematic Internet gamblers during the earliest stages of their Internet gambling. Gambling providers and public policymakers can use these results to inform early intervention programs that target high-risk Internet gamblers.