Validation de l’ECR-R-16F : version française courte de l’Experience in Close Relationships Revised (ECR-R)

L’attachement amoureux est connu être en lien avec le fonctionnement psychique et la psychopathologie de l’individu. L’attachement amoureux est sous-tendu par deux dimensions : l’anxiété et l’évitement. L’ Experience in Close Relationships Revised (ECR-R) est un questionnaire de 36 items largement utilisé afin d’évaluer les dimensions d’anxiété et d’évitement de l’attachement amoureux. Des versions abrégées de l’ECR-R ont été développées dans différentes langues, mais il n’existe actuellement pas de version courte validée en français. L’objectif de cet article est de proposer et de valider une version française courte de l’ECR-R, l’ECR-R-16F. Afin de constituer l’ECR-R-16F, 16 items, 8 items par dimension (anxiété et évitement) ont été sélectionnés sur la base de leur validité d’apparence. Les propriétés psychométriques ont été évaluées dans deux échantillons indépendants ( n = 600 et n = 174). Dans les deux échantillons, les analyses factorielles confirmatoires montrent que l’ECR-R-16F est valide pour évaluer les dimensions d’anxiété et d’évitement dans l’attachement amoureux et présente de bonnes propriétés psychométriques. Des analyses d’invariances ont été effectuées dans le second échantillon, les résultats montrent une invariance de la structure factorielle et de la saturation des items en fonction du genre du répondant. Les scores sur les dimensions d’anxiété et d’évitement sont négativement associés à la satisfaction relationnelle. L’ECR-R-16F peut être utilisé pour mesurer l’attachement amoureux de façon fiable et plus succincte qu’avec la version longue. L’utilisation de cette version courte permettra de faciliter l’évaluation des tendances d’attachement, que ce soit dans un contexte de recherche ou dans un contexte clinique. Romantic attachment is known to be associated with the individual's psychic functioning and psychopathology. Romantic attachment is underpinned by two dimensions: anxiety and avoidance. Anxiety is related to the fear of being abandoned or rejected by the partner, whereas avoidance is related to the avoidance of intimacy and dependency in relationships. Insecure attachment is characterized by high levels of anxiety or avoidance, whereas secure attachment is characterized by low levels on these two dimensions. Romantic attachment tendencies will guide the individuals’ behaviors in close relationships, especially the couple relationship, and insecure attachment tendencies are likely to cause difficulties in these relationships. The Experience in Close Relationships Revised (ECR-R) is a 36-item questionnaire widely used to assess the two dimensions of romantic attachment. However, the length of the ECR-R is often mentioned as a limitation for its use and shorter versions are therefore needed. Short versions of the ECR-R have been developed in various languages, but there is currently no short version in French. The aim of the present study is to propose a short version of the ECR-R in French, the ECR-R-16F. This study details the item selection process and evaluates the psychometric properties of the ECR-R-16F in two independent samples. The validation of the ECR-R-16F has been conducted in two studies. In the first study, 16 items, 8 items per dimension (anxiety and avoidance), have been selected for the short version. The item selection was conducted based on the face validity of items, in order to ensure a representation of each of the main aspects of anxiety and avoidance attachment tendencies and to avoid redundancy among items. The psychometric properties have then been tested in a sample of 600 participants (300 women and 300 men). In the second study, the psychometric properties have been tested in a second sample of 174 participants (89 women and 85 men). Measurement invariance analyses regarding the respondent gender have been conducted to ensure that the scale can be used similarly in men and women. Finally, correlational analyses between the scores on the anxiety and avoidance dimensions and relationship satisfaction have been conducted to ensure the construct validity of the scale. In both samples, confirmatory factorial analyses have shown that the ECR-R-16F is a valid tool to assess anxiety and avoidance dimensions of romantic attachment and has good psychometric properties. The analyses confirmed the two-factor structure of the questionnaire and showed a good reliability for both dimensions. Measurement invariance analyses have been conducted in the second sample and the metric invariance was achieved, showing an invariance of factor structure and item saturation according to respondent gender. However, scalar invariance was not achieved, suggesting that the mean scores on the items differs between men and women. Construct validity was supported by the correlational analyses with relationship satisfaction, showing that higher anxiety and avoidance were associated with lower relationship satisfaction. The ECR-R-16F is a valid tool to assess romantic attachment reliably and more succinctly than with the long version. The use of this short version can facilitate the assessment of romantic attachment in research and clinical settings.