Factor Structure and Psychometric Properties of the Italian Version of the Gambling Related Cognitions Scale (GRCS-I)

• Published in: Journal of gambling studies 2015-03, Vol.31 (1), p.225-242
• Main Authors: Iliceto, Paolo, Fino, Emanuele, Cammarota, Camillo, Giovani, Eleni, Petrucci, Francesca, Desimoni, Marta, Sabatello, Ugo, Candilera, Gabriella, Oei, Tian Po
• Format: Article
• Language: English
• Subjects: Addictive behaviors; Adults; Analysis of Variance; Behavior, Addictive - diagnosis; Behavior, Addictive - psychology; Cognition; Cognition & reasoning; Community and Environmental Psychology; Economics; Factor Analysis, Statistical; Female; Gambling; Gambling - diagnosis; Gambling - psychology; Humans; Hypotheses; Italy; Language; Male; Medicine; Medicine & Public Health; Original Paper; Psychiatry; Psychometrics; Quantitative psychology; Questionnaires; Regression analysis; Reproducibility of Results; Sociology; Statistical analysis; Surveys and Questionnaires - standards; Validity
• ISSN: 1573-3602; 1573-3602
• DOI: 10.1007/s10899-013-9405-6

The past decade has witnessed an expanded accessibility and popularity of gambling worldwide, and in Italy the phenomenon significantly increased. Nevertheless, little is known about the role of gambling cognitions among Italian individuals, and few scales assessing problem gambling have been validated. The purpose of the present study was to examine and validate the Gambling Related Cognitions Scale-Italian version (GRCS-I), based on the 23-item Gambling Related Cognitions Scale (GRCS). Two-tailed t tests, ANOVA, MANOVA, Pearson’s correlation, and multiple regression analyses were used for continuous variables, while χ 2 tests with Yates’s correction for categorical variables. Cronbach’s α was utilized to determine the internal consistency, and logistic regression analysis and the receiver operating characteristic curve analysis to determine discriminant validity. Principal axis factoring with Oblimin rotation was applied, and then confirmatory factor analysis was used to cross-validate the factor structures. We extracted a five-factor solution that accounted for 60 % of variance. All 23 items had communalities and factor loadings were satisfactory, and the factor structures were similar to the original version of the measure. The Cronbach’s α coefficients were adequate, and concurrent and discriminant validities of the GRCS were also confirmed. GRCS-I presented good psychometric properties and it demonstrated good validity and reliability, providing a valid and suitable tool for the assessment of gambling related cognitions among Italian individuals.