Little effort has been made to systematically test the psychometric properties of the Gambling Craving Scale
(GACS; Young & Wohl, 2009). The GACS is adapted from the Questionnaire on Smoking Urges (Tiffany
& Drobes, 1991) and thus measures gambling-related urge. Crucially, the validation of scales assessing
gambling urge is complex because this construct is better conceptualized as a state (a transient and contextdetermined
phenomenon). In the present study, we tested the psychometric properties of the French version
of the GACS with 2 independent samples of community gamblers following an induction procedure delivered
through an audio-guided imagery sequence aimed at promoting gambling urge. This procedure was specifically
used to ensure the assessment of gambling urge as a state variable. Participants also completed measures
of gambling severity, gambling cognitions and motives, impulsivity, and affect. Confirmatory factor analysis
showed that the original 3-factor solution (anticipation, desire, relief) did not fit the data well. Additional
exploratory factor analysis suggested instead a 2-factor solution: an intention and desire to gamble dimension
and a relief dimension. The factorial structure resulting from the exploratory factor analysis was tested with
confirmatory factor analysis in a second independent sample, resulting in an acceptable fit. The 2 dimensions
presented good internal reliability and correlated differentially with the other study’s variables. The current
study showed that, similar to what has been reported for substance-related urges, gambling urges are
adequately probed with a bidimensional model. The findings suggest that the French GACS has good
psychometric properties, legitimizing its use in research and clinical practice.