Évaluation multidimensionnelle du coping: Validation du Brief COPE sur une population française

Abstract

This article aims to introduce the translation and the validation of a multidimensional measure of coping strategies: the Brief COPE (5), in a French population. The coping concept comes from psychological studies that were conducted on stress. In the conceptual analysis of stress by Lazarus and Folkman (25), coping works with two cognitive appraisals performed by the person concerning the perception of a threatening situation and his or her available resources to deal with it. Coping is defined as « cognitive and behavioural efforts to master, reduce, or tolerate the internal and/or external demands that are created by the stressful transaction ». The Brief COPE is the abridged version of the COPE inventory (9) and presents fourteen scales all assessing different coping dimensions: 1) active coping, 2) planning, 3) using instrumental support, 4) using emotional support, 5) venting, 6) behavioural disengagement, 7) self-distraction, 8) self-blame, 9) positive reframing, 10) humor, 11) denial, 12) acceptance, 13) religion, and 14) substance use. Each scale contains two items (28 altogether). This inventory has the advantage of being built from acknowledged theoretical models (Lazarus' transactional model of stress, 1984; behavioral self-regulation model, Carver and Scheier, 1981, 1998). It can be used to assess trait coping (the usual way people cope with stress in everyday life) and state coping (the particular way people cope with a specific stressful situation). As is the COPE inventory, the Brief COPE is a measure used for many health-relevant studies: drugs addiction, ageing, breast cancer, depression, AIDS... Both measures are widely used in Anglophone countries and translated in many languages. Today, the COPE inventory has been validated among Estonian, Croatian, Chinese, and Italian populations and the Brief COPE is also validated among Spanish people. Thus, the worldwide use of this coping inventory should allow a broad comparison of medical and psychological research for coping strategies regarding every kind of pathologies. Thus, we were led to present the translation and the validation of this measure in a French population. Two studies are presented in this article: the first one describes the validation of the inventory in a dispositional format (trait coping) and the second one the validation of the inventory in a situational format (state coping). The French version of Brief COPE, which was used for both studies, was back-translated and analysed by the Brief COPE author: Charles S. Carver. For study 1, 834 first year university students answered the Brief COPE in its dispositional format. To study the factor structure of the Brief COPE, we used structural equation modeling and the LISREL software. Results show that the expected theoretical structure and the observed one fit adequately (χ2 = 606, p < 0,05, RMSEA = 0,04, GFI > 0,95, AGFI > 0,92, RMR < 0,03). In order to study convergent and discriminant validity of Brief COPE, self-esteem (SEI, Rosenberg, 1979), perceived stress (PSS, Cohen et al., 1983), and psychological distress (GHQ-12, Goldberg, 1972) measures were used. Results show that functional coping strategies (eg, active coping) are linked to good self-esteem, to lower perceived stress, and to lower psychological distress, whereas less functional strategies (eg, denial or self-blame) are widely linked to poor self-esteem, to a high perceived stress, and to psychological distress. Study 1 shows also several significant gender differences. Study 2 describes the Brief COPE validation in its situational format. 178 additional students answered this version. The method that was used is the same the one developed by Lazarus and his colleagues for the WCC validation (15). Participants were asked to recall and think about the most stressful event they had experienced during the past two months. They were also to give an account of how much that situation mattered to them. They had to evaluate their capacity to control the situation and indicated whether they felt the situation was amenable to change. The study includes only people who labeled an event as « important » or « very important ». Here again LISREL was used to study the factor structure of the Brief COPE. The results emphasise - as in study 1 - that the expected theoretical structure and the observed one fit adequately (χ 2 = 391, p < 0,05, RMSEA < 0,05, GFI = 0,87, AGFI = 0,80, RMR < 0,06).