Comparing the factor structure of the Problem Gambling Severity Index: Implications for conceptualisation, measurement and practice

Introduction: Problem gambling behaviour has historically been viewed through a clinical lens, however is also considered from a broader public health perspective. Similarly, the measurement of problem gambling has taken its cues from clinically-significant markers of behavioural addiction and their consequences, which may preclude sub-clinical but concerning manifestations of gambling behaviour and related consequences.

Design and Methods: The current study reviews the Problem Gambling Severity Index and subjects the measure to confirmatory factor analysis, comparing a one- and two-factor solution in terms of model fit. Data from the 2015 Northern Territory Gambling Prevalence and Wellbeing Survey was used in the analysis.

Key Findings: Findings from confirmatory factor analyses confirmed both a one- and two-factor solution, however a two-factor solution offered significant improvement in model fit.

Discussions and Conclusions: Though the two factors were highly correlated, it may be beneficial to consider problematic gambling as comprising behavioural and consequential elements that have application in different populations or contexts.

Implications for Practice or Policy: Isolating behavioural and consequential elements of problem gambling may have utility in public health interventions where behaviour is concerning, yet falls below the threshold for a clinically-significant disorder. Similarly, the predictive validity of the Problem Gambling Severity Index may not capture more subclinical elements of problem gambling behaviour.