Speaker
Description
Problematic behavioral patterns with video games can arise or be maintained by maladaptive game-related cognitions, even in adolescents. Thus, for preventive and clinical purposes, it is important to have instruments specifically able to relieve high levels of maladaptive cognitions. The Internet Gaming Cognition Scale (IGCS) is a new and promising scale to assess problematic gaming cognitions through items with a three-ordered response scale. However, relatively few attention has been focused on its psychometric properties, and only by applying Classical Theory of Test. The aim of the present study was to investigate the psychometric properties of the IGCS using item response theory analyses in order i) to test if items really assess medium and high levels of the latent trait and ii) to maximize the scale efficiency by maintaining items that show the best functioning in measuring it. Participants were 741 adolescent video-gamers (58% males, Mage=16.38, SD = 1.53). The Graded Response Model resulted to fit data, but some items did not fit or showed problems of local dependence. By deleting those items, a shorter version of the IGCS was obtained. Total Information Function (TIF) showed that the scale was very informative from medium to high levels of the trait. Invariance across gamers with different levels of severity was tested with Differential Item Functioning (DIF), and correlations with time spent gaming and risky lootbox use were provided. The shortened version of IGCS obtained through IRT seems to be a useful tool for preventive and clinical purposes with adolescents.
