John Blandon, Arnau Carmona, Bianda J González-Santos, Jan Ivern, Marta Beranuy, Mark D Griffiths, Xavier Carbonell, Diana X Puerta-Cortés
The IGDS9-SF appears to be a robust instrument for assessing IGD severity among high-risk Latin American samples, supporting cross-national comparisons at the metric level, and more directly across sex and server. However, the findings identified the 'escapism' criterion (Item 8) as psychometrically weak, challenging its inclusion as a core DSM-5 diagnostic symptom. These results highlight the need to reconsider the conceptualization of IGD.
INTRODUCTION: The present study examined the psychometric properties of the nine-item Internet Gaming Disorder Scale-Short Form (IGDS9-SF) in five Latin American countries. The study aimed to: (i) estimate internet gaming disorder (IGD) prevalence; (ii) test cross-national measurement invariance in a high-risk sample; and (iii) evaluate construct validity by testing the impact of removing Item 8 (escapism) in this population.
METHODS: A total of 10,151 League of Legends players aged 18 to 35 years (19.9% female; M age = 20.84 years, SD = 2.9) from Mexico, Colombia, Peru, Chile, and Ecuador completed the IGDS9-SF. General population prevalence was estimated via a weighted estimator. A probable IGD subsample (n = 1,248; 20.1% female; M age = 20.3 years, SD = 2.9) was used for multigroup confirmatory factor analysis.
RESULTS: The observed rate of probable IGD among recruited players was 12.3%; an exploratory extrapolation to the general population (51.38% regional gamer proportion) yielded 6.32% (95% CI [5.99, 6.65]), with national rates ranging from 4.88% (Chile) to 8.27% (Ecuador). In the probable IGD subsample, metric invariance was supported across countries, whereas scalar and strict invariance were not established. A one-factor model excluding Item 8 and freeing the residual correlation between Items 2 and 3 provided a significantly better structural fit than the original nine-item model. Strict invariance was supported across sex and server. For preferred game mode, metric invariance was supported, but scalar and strict invariance were not established.
CONCLUSIONS: The IGDS9-SF appears to be a robust instrument for assessing IGD severity among high-risk Latin American samples, supporting cross-national comparisons at the metric level, and more directly across sex and server. However, the findings identified the 'escapism' criterion (Item 8) as psychometrically weak, challenging its inclusion as a core DSM-5 diagnostic symptom. These results highlight the need to reconsider the conceptualization of IGD.