Dog or chien? Translation equivalents in the receptive and expressive vocabularies of young French-English bilinguals.

The longitudinal sample was small: 57 children started Wave 1, but 34 remained for all three waves, and missing data or changed language exposure led to exclusions. Experiment 2 analysed 20 toddlers, so the parent-report discrepancy should not be treated as a stable estimate for all bilingual children. Language exposure and vocabulary size were measured with parent or expert reporter forms, and the authors note possible reporting error, especially at later waves. The sample was French-English bilinguals in Montréal, with 82% of mothers holding a university degree, so results may not generalise to other bilingual settings. Translation-equivalent proportions were higher than the approximately 30% often reported, possibly because the checklist method can count more translation equivalents than direct observation. The study observed typical bilingual vocabulary development and did not test therapy, home language plans, or changes in exposure.

Do not conclude that parent report is inaccurate for all bilingual language assessment. The over-reporting was for comprehension of 40 translation-equivalent words on one direct task, and the authors suggest parents may be counting partial, context-supported understanding.