Understanding Expressive Language for Culturally and Linguistically Diverse Preschoolers in a Post Pandemic Setting

Abstract

Record ID: 59

Student Major: Speech-Language Hearing Sciences

Project Advisor: Leslie Kokotek

Abstract: This project aims to describe the potential relationships in expressive language performance in preschool-age children who speak different dialects of English (i.e., Standard American English [SAE] and African American English [AAE]) following over a year of remote learning and social distancing restrictions during the COVID-19 pandemic. We used both standardized and authentic measures of language expression to conduct our analysis. With some of the methods for authentic Language Sample Analyses (LSAs) requiring complex and time-consuming calculations, this project also describes the feasibility and benefits of using automated computerized examinations of LSAs, which provided a more robust set of data for comparison to the traditional expressive language assessment (i.e., the Structured Photographic Expressive Language Test - 3rd edition [SPELT-3]). The automation process for LSAs included both the Systematic Analysis of Language Transcripts (SALT) and Computerized Language Analysis (CLAN). To capture these results, five preschool-aged children, (n=3, SAE; n=2, AAE) completed the SPELT-3 and completed a 15-minute play-based language sample. LSAs included: Mean Length Utterancew/m, Number of Different Words, Type-Token-Ration, Developmental Sentence Scores, and the Index of Productive Syntax. A spearman analysis was completed to understand how children's LSAs compared to the SPELT-3. Results indicated no statistically significant relationships between LSAs and children's performance on the SPELT-3. This suggests that even the SPELT-3 which allows for the inclusion of dialectically appropriate responses, does not comprehensively detect features of children's grammatical complexity as noted by the lack of relationships between the SPELT-3 and all measures of LSAs.