Bilingual speech sound development during the preschool years: The role of language proficiency and cross-linguistic relatedness

Abstract

Purpose: The purpose of this study was to investigate typical Spanish-English speech sound development longitudinally in a group of bilingual preschoolers enrolled in a Head Start Program and to examine the extent to which such development is linked to language proficiency. The study also aimed to identify whether speech development is related cross-linguistically, and to improve our understanding of error patterns in this population. Method: Thirty-five bilingual preschool children produced single-word speech samples in Spanish and English both at the beginning of their first and their second year in a Head Start program. Conversational samples in both languages were also collected at these data points to calculate Mean Length of Utterance in words (MLUw) and thus assess the children’s linguistic proficiency. The phonetically transcribed speech samples were compared over time in terms of segmental accuracy measures and error pattern frequencies. Correlation analyses were run to examine the relation between segmental accuracy measures across languages, and between speech sound production and MLUw. Results: One-way within-subjects ANOVAs revealed significant improvements in accuracy over time in both languages, but not always for cross-linguistically unshared segments, nor for all consonant manner classes. Overall error rates decreased over time in both languages, although certain error types showed no change. Cross-linguistic interactions were low in both languages. The results also revealed significant cross-linguistic correlations in segmental accuracy between Spanish and English, as well as between MLUw and speech sound production in both languages on a range of measures, with language-specific differences in Year 2 of the Head Start Program, but not in Year 1. Conclusions: This study is the first to document developmental changes in the speech patterns of Spanish-English bilingual preschool children over one year. Accuracy rates improved significantly in both languages, suggesting that enhanced exposure to the majority language at school may not impede