Conceptualization and validity of the compulsivity construct in potentially addictive behaviors: a replication and extension study with the brief Granada Assessment for Cross-domain Compulsivity (GRACC18) in the gambling and video gaming domains

Background. The definition of compulsivity and its role in putative addictive behaviors remains unclear, partly due to previous research conflating its conceptualization as a general transdiagnostic trait with its understanding as an acquired feature of a specific activity as it spirals out of control. Methods. This study aims to validate a short version of the GRACC90 scale (GRACC18), designed to assess the degree to which video gaming and gambling have become compulsive in two independent samples of panel members who regularly engage in one of these activities—ranging in severity from recreational to pathological (though not formally diagnosed)—. Exploratory structural equation modeling (ESEM) was applied to examine the factorial structure of the scale and to test its structural invariance across both domains. Additionally, statistical associations between compulsivity scores and gambling- and gaming-related constructs—including problem severity, positive and negative affect, motives, and quality of life—were explored. Results. Findings indicate that (a) the compulsivity scale exhibits reliability, validity, and a unifactorial structure, (b) its structure remains invariant across domains, (c) compulsivity is strongly correlated with symptom severity in both gambling and gaming, moderately associated with negative affect and quality of life, and not significantly linked to positive affect, and (d) no interaction effect between domain and compulsivity was statistically significant in the regression models tested to predict the two affect dimensions and quality of life. Furthermore, (e) severity shows a stronger correlation with affect and quality of life than compulsivity, and (f) cross-loadings between severity and compulsivity items are notably low. Conclusion. These results support the theoretical validity of compulsivity as distinct from severity, demonstrating an identical structural composition and comparable involvement in symptoms and harms across video gaming and gambling domains. Future research should further investigate the emotional, cognitive, and computational foundations of compulsivity in these and other behavioral contexts.