Validation of a Sports Betting Adaptation to the Problem Gambling Severity Index in Young Adults

Abstract

Background. Sports betting is a rapidly growing addictive behavior, especially among young adults. As such, there is a need for measuring problem sports betting behaviors and consequences separately from established generalized gambling measures. The present study provides support for a sports betting adaptation of the Problem Gambling Severity Index (PGSI-SB). Methods. We recruited a sample (N=221) of young adults aged 18-29 (Mage=24.4; 22% female; 13.2% Hispanic; 68.6% college degree) from 36 different US states. Eligibility criteria included ≥2 sports betting days in the past month. Results. Confirmatory factor analyses showed support for both a single and two-factor model with subscales for problematic behavior (e.g., dependence) and negative consequences. The PGSI-SB was strongly correlated with the original PGSI in terms of scale-level and item-level correlations (i.e., convergent validity). Aim 3 established predictive validity of the single-factor PGSI-SB via significant associations with three indices of past two-week sports betting: frequency, number of bets, and total amount wagered. Predictive validity for the two-factor model was impacted by multicollinearity, given high correlation between subscales. Conclusions. Findings establish the merits of a dedicated problem sports betting measure for young adults, which is a key step towards enhancing the quality and consistency of sports betting research.