TRADUCTION ET VALIDATION CANADIENNE-FRANÇAISE DE L'ÉCHELLE DE L'ESTIME DE SOI DE ROSENBERG*

Since the early days of psychology, the self-concept and more specifically global self-esteem has been addressed by numerous researchers in the field. Among the widely used instruments for measuring global self-esteem, Rosenberg's Self-Esteem scale (RSE) has displayed high levels of validity and reliability (Harter 1983; Wylie 1974). Through a brief (10 items) and direct approach, the RSE can be answered quickly by different age and social groups. Considering the good psychometric properties of the RSE mentioned above, and the importance of self-esteem for research in psychology, the purpose of the present four studies was to translate and validate in French the RSE Internal consistency (study 1) and confirmatory factor analyses through LISREL VI (study 2) were conducted with a French-Canadian junior college student population in order to verify the internal consistency and the unidimensional factorial structure of the scale. In addition, the construct validity (study 3) and temporal stability (study 4) of the scale were assessed. Although the present findings are preliminary in nature, they nevertheless support the reliability and validity of the Rosenberg's Self-Esteem scale in French.     

COMPARAISON INTERLINGUISTIQUE DE MATÉRIEL VERBAL PROPRE A L'APPRENTISSAGE DE CONCEPT1

Inter-language comparison of verbal material for the study of concept formation. — A list of 215 French words, which were translation equivalents of the words in Underwood and Richardson's (1956) published list, were presented to 119 French students from which single, restricted to sense impression associates were obtained. A systematic inter-language comparison is established and differences between French and English speaking subjects discussed. The piocedure adopted by Mednick and Halpern (1962) to study the ease of concept attainment as a function of Associative Rank has been applied to French subjects. A replication using a direct translation of the English nouns shows no difference between concepts at Associative Rank 1 and 2. It appears indeed from norms of the French group that the concepts selected are balanced not only with respect to Dominance Level but also with respect to Mean Associative Rank. But further experiment, with a material established according to the norms of the French list of associates, shows a very significant difference in the mean number of trials taken to solve each concept. It thus strengthened the assumption of homogeneity of associative response hierarchies in a given language community.