L’évaluation du langage oral chez les enfants bilingues au moyen d’une batterie standardisée pour monolingues

RésuméL’objectif de cette étude est de mesurer la fiabilité des épreuves de langage oral d’une batterie standardisée monolingue, l’Exalang 5-8, chez des enfants grandissant en contexte bilingue. Pour cela, nous avons comparé les performances d’un groupe de 21 enfants bilingues, âgés de 5 ans 11 mois à 7 ans 11 mois, à celles d’un groupe témoin de 21 enfants monolingues de même âge et de même niveau socio-culturel, à travers 17 épreuves évaluant différents domaines langagiers : la phonologie, la métaphonologie, le lexique et la morphosyntaxe. Les résultats montrent un effet négatif du bilinguisme à seulement deux épreuves : la dénomination d’images et la production de récit à partir d’une séquence animée. Pour ces deux épreuves, l’établissement d’une norme bilingue apparaît nécessaire. Ainsi, la standardisation de certaines épreuves destinées à l’évaluation du langage oral des enfants monolingues apparaît utilisable dans le cadre de l’évaluation du langage d’enfants bilingues précoces. Un impact de la longueur d’exposition et du taux d’exposition au français a également été observé sur les performances des bilingues à certaines épreuves, montrant la nécessité de prendre en compte l’hétérogénéité des profils de bilinguisme lors de l’évaluation du langage oral de ces enfants.

Assessing the language of bilingual children is highly challenging for speech and language pathologists, especially in a country like France that has a strong monolingual tradition. In France, the possibility of testing bilingual children in another language than French remains to date limited and there are virtually no available French language tests providing bilingual norms. Additionally, many factors related to the bilingualism situation of the children can influence their performance on language tests, such as the length of exposure to the language of testing, i.e. the number of years and months the child has been exposed to that language at the moment of testing, or the rate of exposure, i.e. the percentage of exposure the child gets to that language on a daily basis at home. The aim of this study was to assess the reliability for bilingual children of a French test battery, Exalang 5-8, originally standardized for French monolingual children. Using 17 different tasks, four language domains were examined: phonology, metaphonology, lexicon and morphosyntax. Performance of 21 bilingual children with a mean age of 6;10 (range: 5;11–7;11) was compared to the performance of 21 monolingual peers. The bilingual children had started being exposed to French at a mean age of 10 months (range: 0–3;0) and had an average rate of exposure to French around 50% (range: 7–81%). Monolingual and bilingual children were of similar high socio-economic status, with all mothers having achieved at least a secondary level education. The effect of bilingualism was assessed for each of the 12 tasks. We also measured the influence of length and rate of exposure to French on the performance of the bilingual children. Results showed an effect of bilingualism, with bilinguals performing significantly more poorly than monolinguals, on two tasks only: the lexical task of picture naming and the syntactic task of narrative production. These results suggest that for these two specific tasks, bilingual norms are necessary. However, the monolingual standardization of the other tasks seems to remain useful in the context of bilingualism. We also observed an influence of length and/or rate of exposure to French on the performance of the bilingual children on three tasks: the picture-naming task, the grammaticality judgment task and the task of receptive grammar. Greater length and/or rate of exposure to French led to better performance on these tasks. These results confirm the need of considering the specific situation of each bilingual child in terms of amount and pattern of language exposure when assessing his or her language.