Addiction aux jeux vidéo en ligne : étude comparative des outils de mesure en langue française

Abstract

During these last years, online video game addiction has been a growing research topic, highly debated among researchers around the world. This addiction is evaluated by an important number of questionnaires which were validated with different statistical methods. In France, clinicians and researchers are also confronted to a large choice of questionnaires. However, these French questionnaires do not seem to evaluate the same object and they must be used for different types of populations. In this study, we compare four questionnaires often used in therapy and research about online video game addiction. These tools are the English and French version of the Problem Video game Playing (PVP), the Internet Addiction Test (IAT), the Questionnaire de Mesure de l’Intensité aux Conduites Addictives (QMICA, French version only) and the Game Addiction Scale (GAS). They are compared to a questionnaire in English: the Internet Gaming Disorder-20 questionnaire. Several statistical characteristics comparisons are highlighted in this study. Factorial analysis statistical determinants, concurrent validity, number of items, the specificity of the sample (age, psychological disorders), construction of the scale and the theory behind each questionnaire are detailed and compared in this study. For all the questionnaires, the addiction criterion of the DSM-IV or the DSM-5 are considered. However, no distinction between online and offline video game addiction is made for all the questionnaires. No distinction of the game genre is made either. There is a high variation of the items number, the age and the number of participants evaluated for the statistical validation. It varies from 7 items for the GAS to 20 items for the IGD-20. The number of participants varies from 50 for the French version of the PVP to 1003 for the IGD-20. Some questionnaires are validated on a population of teenagers (PVP, GAS), and other are for adults (IAT, QMICA, IGD-20). Cronbach’ alphas are good for all questionnaires, except for the English version of the IAT (“moderately weak”). A factorial analysis is commonly used for the validation and/or the translation of questionnaires. The French version of the IAT and the IGD-20 inform us about the Root Mean Square Error of Approximation (RMSEA). Other authors used the total variance explanation. Finally, different concurrent validities are highlighted, such as depression and anxiety for the English version of the PVP, playing time, life satisfaction and socials skills for the English version of the GAS, or other addictions like tobacco for the QMICA. Considering all these statistical characteristics, The Internet Gaming Disorder-20 questionnaire seems to be the best tool to evaluate the video game addiction, but it is not validated in French yet. For teenagers, the GAS seems to be the more useful. In order to evaluate other addition, the QMICA is the best questionnaire. The statistic elements of the French versions of the PVP and the IAT are questionable for the use in therapy and research.