One of the questions that we often ask ourselves when doing bilingual research and when conducting bilingual assessment is how to describe and characterize children’s bilingualism. This question is important for making educational decisions that involve language of instruction. For assessment and diagnosis of speech and language impairment it is critical that we document children’s bilingual profiles. But, it’s not as easy as we would like. We explore some of these issues in an article that appears in Bilingualism: Language and Cognition.In this study, we looked at about 1000 preschool children’s current input and output in Spanish and English and first year of English exposure relative to their performance on a screener of morphosyntax and semantics. This is the same participant pool as the risk paper we published in AJSLP, but here we focus on bilingual profiles. We were interested in the relationship between performance on morphosyntax and semantics in both English and Spanish.
One of our findings is that current use/exposure to English and Spanish was a better predictor of performance on language tasks compared to year of first exposure to English. This is an important finding because typically, we ask parents about first exposure to English. When children first started speaking English is thought to influence children’s proficiency in a language– and it does. We did find a significant relationship between year of first exposure and performance. But, we found a STRONGER association between current exposure and performance– this association contributed unique variance to children’s performance. I think a take-away message here is that we need to not only ask when children first started learning English as a second language, but how much they use each language currently. Other researchers suggest that the kind of exposure (school vs. home; adult vs. child) also has effects on performance. We haven’t broken it out that far, but it seems logical to understand under what circumstances children use each language to get a better idea of what they might know in each language.
The other interesting thing was that in terms of percent correct performance, children needed a little less exposure to one or the other language to perform relatively well in semantics (about 50%), but they needed a lot more exposure to perform well on morphosyntax (about 75%). The patterns are a little different for Spanish vs. English. But, I think the important thing here is that when we use direct measures of language performance, the measures we use may give us different results. Different levels of proficiency in L1 and L2 may be seen across different domains of language. This is important to know and interpret apparent “gaps” or inconsistencies in dominance. For this reason we think it’s best to look at exposure and use in each of the two languages.