Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
In Event: Special Poster Session 05 with Continental Breakfast Reception
In Poster Session: PS 05 - Policy Section
Data from the 2013 American Community Survey show that one in five U.S. residents spoke a foreign language at home(Camarota & Zeigler, 2014). For many of these families, English is the non-native second language of the parents, but the first language of the children. This surge in the number of bilingual children has given rise to an increase in research on characteristics of young children’s language input in either/both languages, and how this relates to children’s language development(e.g., Hoff, 2018). Most studies have focused exclusively on how relative quantities of input in each language predicts children’s lexical and sematic proficiency and development. However, few studies have focused on the qualitative aspects of children’s bilingual input, such as lexical diversity, or the variety of words spoken to young children. Considering how significant high-quality input is for children’s linguistic and cognitive development(e.g., Rowe, 2012), it is important to investigate how the quality of bilingual children’s input varies by language.
Participants were 18 families who self-reported that their children (13 male and 5 female) were learning both English and Spanish. Children were aged four to six years old (mean age= 5.22, SD= 0.808) and enrolled in either Kindergarten or pre-Kindergarten in the greater Boston area. Children were a subset of children from a larger study on academic achievement and language development, and all children with any reported Spanish exposure were included in this analysis. Parents completed the Language and Social Background Questionnaire (Anderson et al., 2017), revealing that most of the 18 parents were native Spanish speakers. Caregiver-child dyads were invited to participate in a 20-minute videotaped interaction, during which they had a snack and were provided with an engaging toy, but were otherwise given no instruction on how to interact. Interactions were transcribed in CLAN using CHAT conventions(MacWhinney, 2000) by a trained bilingual researcher. Families were divided into three subgroups based on primary language spoken during the interaction: primarily English by both speakers (n=8), primarily Spanish by both speakers (n=2), or bilingual (n=7), in which parents tended to speak primarily in Spanish while children primarily spoke in English. Type-to-token ratio (TTR) was then calculated by dividing the types (total number of different words) by the token (total number of all words) for each speaker. A higher TTR indicates a greater lexical diversity. Parents additionally reported highest completed education levels and family income, from which a socioeconomic status (SES) composite was derive.
After controlling for SES, child age and gender, caregivers who spoke primarily in Spanish (Spanish and bilingual groups combined) exhibited significantly higher TTRs than caregivers who spoke primarily in English in the recorded speech samples (t = 0.661, p < 0.01, mean Spanish TTR = 0.43, mean English TTR= 0.28), indicating that parents speaking primarily in their native language used more diverse vocabulary when speaking with their children. Given that higher lexical diversity in child-directed input is positively correlated with children’s language skills(Rowe, 2012), these results may have important implications for recommendations to caregivers on language use with their young children.
Veronica Valencia, MIT
Rachel R. Romeo, 1Department of Brain and Cognitive Sciences and McGovern Institute for Brain Research; Division of Medical Sciences, Harvard University
Julia A. Leonard, Department of Brain and Cognitive Sciences and McGovern Institute for Brain Research, Massachusetts Institute to Technology;
Meredith Rowe, Harvard University
John D.E. Gabrieli, Department of Brain and Cognitive Sciences and McGovern Institute for Brain Research