Validation of the French version of the « Meta-Cognition Questionnaire » for adolescents (MCQ-Af): Evolution of metacognitive beliefs with age and their links with anxiety during adolescence

INTRODUCTION: The Meta-Cognitions Questionnaire for Adolescents (MCQ-A) measures individual differences of metacognitive beliefs and monitoring thought to be involved in the onset and maintenance of psychological disorders, especially in those involving anxiety. This assessment tool has been employed in research and clinical settings involving French-speaking adolescents, but appropriate validation has yet to be conducted. This article aims to first validate the francophone version of the MCQ-Af using measures sensitive to the expression of anxiety, and secondly, to examine the influence of age and gender on metacognitive beliefs, anxiety and their links. METHOD: 214 adolescents (114 females) between 13 and 17 completed the MCQ-Af (French version) as well as the Revised-Children's Manifest Anxiety Scale (R-CMAS), French version, to assess anxiety manifestations. Structural validity was examined with confirmatory factor analyses. Three models were compared to the higher order five factor model proposed in the original validation study. Internal consistency and test-retest reliability were also performed. Student’s t tests as well as simple and stepwise regressions were conducted to assess for age and gender. RESULTS: The five correlated factors retained in the original version of the MCQ were replicated, and confirmatory factor analyses yielded comparable fit indices for a covariate factor model, as well as for a bifactor model. The bifactor model was privileged for theoretical reasons. Analyses were performed on a shortened questionnaire of 27 items as 3 items (2, 12 and 14) had non-significant loadings in prior path analyses. Age and gender differences were found in specific sub-factors of the MCQ-Af; positive and negative metacognitive beliefs seem to increase with age; girls seem to score higher on the negative metacognitive beliefs and thought control dimensions. The study further reports specific links between anxiety manifestation and negative and positive metacognitive beliefs, as well as confidence in one’s memory. A modest influence of age and gender on this link was also highlighted. CONCLUSION: The present research provides the first evidence that the MCQ-Af is a valid and reliable instrument to assess individual differences of metacognitive beliefs in French-speaking adolescents. Nevertheless, it highlights that caution should be taken in regards of 3 items in particular (items 2, 12 and 14). Furthermore, age and gender in assessed samples of adolescents might influence the scores of the different dimensions of the questionnaire.