Predicting Spontaneous Language Growth with Standardized Language Assessments in Bilingual Children

• In Event: Poster Session 12
  In Poster Session: PS 12 Topic - Language, Communication

Abstract

Bilingualism continues to become more common in the United States, however, there is still a smaller body of research examining the development of two languages simultaneously in young children when compared to the amount on monolingual language development. Important research into how two languages develop simultaneously is an essential topic to further our understanding of the developing bilingual brain. One way to do this would be to use more ecologically valid languages assessments to measure children’s use of both their languages (Wood et al., 2018). In this study, we focus on whether children’s scores on a standardized language assessment predict growth on a spontaneous language sample in English and Spanish. Use of more ecologically valid assessments of language alongside standardized measures can provide further insight into how bilingual children use their languages and how they may develop over time.
Participants were 59 bilingual children (Mage at T1 = 15.3 months) attending a center-based childcare program serving low-income families completed a battery of language measures once every six months over two years. Each child completed the Spanish-English bilingual version of the Preschool Language Scales (PLS). They also completed a free play session with their parent or caregiver, five minutes of which was transcribed. From this transcript, data on the number of different English words and Spanish words used by the child were extracted for the analysis.
Growth models were used to analyze the rate of growth of the number of different words (NDW) in English and Spanish across each timepoint. Each language was analyzed separately. Unconditional models were run to evaluate the functional form of growth across time. Unconditional models for English NDWs indicated that a model with a random intercept and slope provided the best fit to the data. Next, PLS scores were added to this model as a predictor. This model indicated that PLS scores significantly predicted initial level of English NDWs produced by children. Finally, an interaction between time and PLS scores was added, and this model indicated that PLS scores significantly predicted rate of growth such that children with higher PLS scores experienced more growth of NDWs over time. The same method was used for Spanish NDWs. Unconditional models indicated the best fit for the Spanish data was a model with a random intercept and fixed slope. Next, PLS scores were added as a predictor. PLS scores significantly explained individual differences in initial level of Spanish NDWs. However, when the interaction between time and PLS scores was added to the model, there was no indication that PLS scores predicted rate of growth of Spanish NDWs. These results indicate that children with higher PLS scores also had higher initial levels of English and Spanish NDWs. Additionally, higher PLS scores predicted a faster rate of growth in English NDWs, but this pattern did not hold for Spanish NDWs. These findings highlight the need for early language intervention, as early conceptual language development sets the stage for future acquisition of the second language for bilingual children.

Authors

• Lauren Thayer, Texas A&M University - College Station
  Presenting Author

• John Marc Goodrich, University of Nebraska - Lincoln
  Non-Presenting Author

• Lisa L Knoche, University of Nebraska - Lincoln
  Non-Presenting Author