Understanding bilingual children’s language use using cross-linguistic analysis

Background: An increased understanding of typical cross-linguistic interactions (code-mixing) is critical for describing bilingual children’s language use. In educational and clinical settings, a lack of understanding about typical bilingual language use leads to the misdiagnosis of language functioning. This study used the Index of Productive Syntax (IPSyn, n=56 structures) and token-based analyses to quantify and characterise cross-linguistic interactions in bilingual Jamaican-Creole-(JC)-English-speaking preschoolers’ productions. The findings support bilingual children’s communication by increasing educators’ and clinicians’ awareness of language use in understudied contexts.

Aim: To quantify and characterise cross-linguistic interactions of JC-English speaking preschoolers’ spontaneous language productions.

Method: JC-English-speaking preschoolers (n=61) completed 15-minute language samples in JC and English. Preschoolers’ spontaneous productions were analysed using a four-step process to complete cross-linguistic analysis including: (1) coding of cross-linguistic interactions within- and across-utterance levels; (2) within-utterance analyses using the IPSyn and token-based analyses; (3) across-utterance analysis to establish across-utterance rates; and (4) temporal analysis coding interactions at the beginning, middle, and end of samples.

Results: Item-analysis using the IPSyn revealed language structures involved in cross-linguistic interactions. A mean of 27.7/56 (SD=6.4) IPSyn structures were coded in JC and 26.4/56 (SD=5.7) in English. Token-based analysis documented domains underlying cross-linguistic interactions at the within-utterance level, with most to least frequently-coded domains as syntactic (M=13.9%), phonological (M=8.7%), morphological (M=8.1%), and lexical (M=0.4%) in JC and phonological (M=9.2%), syntactic (M=7.5%), morphological (M=4.3%), and lexical (M=0.6%) in English. Across-utterance analysis revealed statistically significant differences between the mean rate of cross-linguistic interactions across languages, t(60)=8.55, p<.001, with higher rates in JC (M=44.9%) than English (M=27.8%). Temporal analysis revealed cross-linguistic interactions occurring at the beginning (JC:29.4%, English:28.6%), middle (JC:33.7%, English:33.9%), and end (JC:36.9%, English:37.5%) of samples.

Conclusions: Using cross-linguistic analysis can inform typical patterns of bilingual language use in understudied linguistic contexts. This approach supports culturally appropriate practices in clinical and educational contexts.

Implications for children: It is amazing that many children speak two languages! We can learn from you and how you use all your languages.

Implications for families: It is important that educators have an accurate understanding of bilingual children’s languages. We can better understand how bilingual children use their languages by looking at how they use all their languages.

Implications for practitioners: This study provides an approach for analysing children’s cross-linguistic interactions (code-mixing) and specific examples of language patterns you may see when working with JC-English-speaking preschoolers.

Funding: The first author was a doctoral scholar at the time of this research and the second author is a co-Investigator on a United States Department of Education Preparation of Special Education, Early Intervention, and Related Services Leadership Personnel grant. The research was supported by an Endowment Gift Fund to the Jamaican Creole Language Project and a University of Cincinnati Vice-President for Research Start-up Funds.

Key words: children’s voices, Jamaican Creole, language development, play, vulnerable communities, preschoolers, bilingual, code-mixing

This presentation relates to the following United Nations Sustainable Development Goals:

• SDG 3: Good Health and Well-being
• SDG 10: Reduced Inequality