Confirmation des dimensions de l’échelle des peurs pour enfants révisée (FSSC-R) d’Ollendick (1983)

Abstract

L’objectif de la présente étude est de vérifier la stabilité factorielle de la FSSC-R à partir d’une analyse de confirmation de la structure factorielle. Le questionnaire a été donné à 612 adolescents (271 garçons et 341 filles), d’âge moyen de 14,17 ans (2,47). Les résultats de l’analyse de confirmation sont en faveur des cinq dimensions décrites par Ollendick et collaborateurs. L’ensemble des participants obtient un score total moyen de 113,92 (21,56), le premier facteur peur de l’échec et de la critique obtient une moyenne égale à 26,95 (6,01), le second facteur peur de l’inconnu a une moyenne à 24,96 (5,53), le troisième facteur peur des blessures et des animaux à 30,76 (6,37), le quatrième facteur peur du danger et de la mort à 22,26 (5,60), enfin le cinquième facteur peurs médicales à 8,98 (2,06). Le score total donne une évaluation globale des peurs de l’enfant alors que les cinq dimensions aident le clinicien à préciser le diagnostic psychopathologique. The revised version of the Fear Survey Schedule for Children (FSSC-R; Ollendick, 1983 [1] ) is one of the most widely used questionnaires to measure the prevalence of children's fears and to provide an overall assessment. The factorial structure groups the items into five scales, providing a better understanding of children's fears. The scales are: Fear of Failure and Criticism, Fear of the Unknown, Fear of Minor Injuries and Small Animals, Fear of Danger and Death, and finally Medical Fears. These scales evoke fairly common specific phobias in children (Ollendick et al., 1989 [4] ). The psychometric properties of the FSSC-R are well established and the French version was tested in two general population studies (Rusinek et al., 1998; Turgeon et al., 2005 [2] , [11] ). The French version of the FSSCR has been studied and adapted for a population from first grade (6.5 years) to third grade (15.5 years) by Rusinek et al., 1998 [2] . The Quebec version is newer (Turgeon et al., 2005 [11] ) and therefore more accurately takes into account of current reality in item formulation. The objective of this study is to examine the factorial stability of the FSSC-R using a confirmatory factor analysis. The Quebec version of the FSSC-R was given to 612 adolescents (271 boys and 341 girls), with a mean age of 14.17 years (SD: 2.47). The five factorial structure model of the FSSC-R (Ollendick et al., 1989 [4] ) was studied using AMOS software. The results show that the relative Chi 2 is equal to 2.56 and the other two indications have acceptable values (RMSEA is 0.05 and SRMR is 0.06) confirming the five-scale model described by Ollendick, 1983 [1] . Subjects had a total fear score of 113.92 (21.56); for the individual scales subjects scored 26.95 (6.01) on Fear of Failure and Criticism, 24.96 (5.53) on Fear of the Unknown, 30.76 (6.37) on Fear of Minor Injuries and Small Animals, 22.26 (5.60) on Fear of Danger and Death, and finally 8.98 (2.06) on Medical Fears equals to. The means observed are generally lower than those reported by Turgeon et al., 2005 [11] and Ollendick, et al., 1985 [15] . It is important to note that the current sample consists of adolescents and young adults (range 11–21 years) while the study of Turgeon et al., 2005 [11] includes only children between 9 and 12. The study of Ollendick et al., 1985 [15] reports data for four age groups: 7–9 years, 10–12 years, 13–15 years and 16–18 years without showing any difference between these age groups, but this study was conducted in 1985. Conversely, Rusinek et al., 1998 [2] reported a change of fears from childhood to adolescence. It is possible that the emergence of technology has led today's teenagers and young adults to have fears different from those common to the 1980's. The internal consistency coefficient for the 79 items is 0.95. As for the original scale, four of the five factors show satisfactory internal consistency: the Fear of Failure and Criticism factor (0.87), the Fear of the Unknown factor (0.86), the Fear of Minor Injuries and Small Animals factor (0.85) and the Fear of Danger and Death factor (0.82). The Medical Fears factor shows a slightly lower coefficient alpha (0.64). Thus, the results of Turgeon et al., 2005 [11] and those of Ollendick, 1983 [1] are confirmed in terms of adequate internal consistency coefficients for the first four factors and a slightly weaker coefficient (less than 0.70) for the Medical Fears factor. In conclusion, the instrument appears to retain its original validity through good internal consistency and the five factor scales described by Ollendick et al., 1989 [4] in this French validation.