77

RésuméLe Memory Compensation Questionnaire est un outil qui mesure la manière dont les individus compensent leurs pertes de mémoire, qu’elles soient objectives ou perçues. L’objectif de cette étude est de valider en français la version abrégée du Questionnaire de compensation mnésique (MCQ) [1]. Après avoir sélectionné méthodiquement 25 items sur les 43 de la version longue, nous avons testé cet outil sur un échantillon de 260 personnes âgées sans troubles cognitifs entre 50 ans et 85 ans. Une analyse factorielle confirmatoire a été effectuée pour mesurer la validité interne et la fidélité. Une analyse de corrélation de Spearman nous a permis de mesurer la validité externe et la normalisation. Les résultats montrent que la structure factorielle de la version brève BF-MCQ correspond bien avec le modèle hypothétique, ainsi qu’avec les différentes versions longues existantes. La consistance interne est excellente et la validité de critère montre une corrélation avec le sentiment d’efficacité mnésique, la personnalité et les affects, ainsi qu’un effet de l’âge et du genre. Finalement, cette étude nous permet de valider le BF-MCQ comme un instrument statistiquement valide et fidèle.

French Validation of the Brief French Memory Compensation Questionnaire with aging peopleThe Memory Compensation Questionnaire [2] is a seven-factor scale that measures individual differences in the tendency to select particular strategies and to overcome perceived or real memory losses. French versions called the French Memory Compensation Questionnaire (F-MCQ) have been validated [1]. The purpose in this study is to develop and examine the factor validity and reliability of a short version of the F-MCQ (BF-MCQ) with half on the initials items of the long version. After selecting methodically the items in the shorter version, we tested this tool on 260 cognitively healthy people (67% of female) aged between 50 and 85 years old (M = 65 years). A confirmatory analysis was done for the internal validity. Reliability was measured by Cronbach's alpha and the criterion validity was tested with Spearman's correlation analysis. The results of the confirmatory analysis show that the hypothetical model of the short version of the F-MCQ adequately fit with that obtained with the data and corresponded well with the long French version. The average standardized factor loading of the items were high ( i.e., M = [0,56 ; 0,87]), comparatively with the long French version ( i.e., M = [0,52 ; 0,71]). The reliabilities of the scales of this short version were all bred (Cronbach's alpha values = [0,65 ; 0,86]), and regarding the long version (Cronbach's alpha values = [0,62 ; 0,75]). In line with the literature, the Spearman's correlation put in evidence correlations between memory compensation strategies and memory self-efficacy, personality traits, affects, as age and genre. In summary, these results allow us to consider that the brief version of the French Memory Compensation Questionnaire is a valid and accurate tool that can be used in research context as in clinical practice situation.