Approaches to learning in the non-academic context: construct validity of learning approaches test in video game (lat-video game)

The subject-object interaction of knowledge studied by the field of learning approaches has been evaluated exclusively in the school/academic context. However, the field does not assume that these interactions are manifested only in this context. This article studies the validity of the Video Game Approach Test (LAT-Video Game), with the novel proposal to evaluate approaches in a non-academic context. The structural validity and its generality were investigated, as well as the predictive and divergent validity of the LAT-Video Game in two independent samples. Three models were tested in the first sample and the constrained bifactorial model gave the best fit and parsimony. By comparing the two samples, this model proved to be invariant even to the scalar level. The LAT-Video Game predicts the self-declaration of people as gamer or non-gamer at 84% [65% -100%]. The deep approach in video games does not correlate with approaches in the academic context, measured by the Learning Approaches Scale (LAS). The motivation related to the practice of video games correlates positively with the superficial approach and negatively with the deep approach. The LAT-Video Game shows structural validity, invariance, predictive and divergent validity.