Measuring Gambling Outcome Expectancies in Adolescents: Testing the Psychometric Properties of a Modified Version of the Gambling Expectancy Questionnaire

The Gambling Expectancy Questionnaire (GEQ; Gillespie et al. 2007a) is a 23-item scale assessing three positive outcome expectancies (Enjoyment/Arousal, Money, SelfEnhancement) and two negative outcome expectancies (Over-Involvement, Emotional Impact) related to gambling. It is the most used instrument to assess gambling outcome expectancies in adolescents and it has good psychometric properties. To allow a greater and more useful application of the scale, the present study aimed to modify the GEQ to make it usable with all adolescents, regardless of their gambling behaviour and to verify its psychometric properties. To that aim, the items were modifed and the response scale was reduced from a seven-point to a fve-point Likert scale. To verify the adequacy of the modifed scale, two studies were conducted among Italian adolescents. In the fIrst study (n=501, 75% males, Mage=16.74, SD=.88), after having removed four items and relocating another through explorative factor analysis, the original fve-factor structure of the scale was confrmed by applying a confrmatory factor analysis. Reliability and validity evidence were also provided. The second study (n=1894, 61% males, Mage=15.68, SD=.71) attested its invariance across gambling behaviour status and gender. The modifed version of the GEQ (GEQ – MOD) can be proftably used for research and preventive purposes with youth.