Collaborative Research: Using Sensing Technology and Automated Speech Recognition to Capture Teacher Language Interactions in Diverse Pre-K Classrooms

Young children’s language development is crucial for their later academic success. During the preschool years, teachers contribute to children’s language development by engaging in conversations with them. However, it is not clear how often preschool teachers have language interactions with children, or if every child in the class receives opportunities to talk with teachers.

This project uses innovative sensing technology tools to examine teachers’ language interactions with preschoolers over the course of a school year. The project employs advanced speech processing algorithms that automatically analyze language data, providing a more detailed understanding of how language is used in the classroom than has previously been possible.

The results provide insights into how preschool language experiences shape language development and whether children’s language experiences differ based on their background characteristics, such as their standardized English language skills, temperament, or gender. The eventual goal of this work is to provide teachers with concrete, actionable data on how to enrich language experiences in preschool and lay a strong foundation for future reading success.

The study seeks to answer the following research questions: How often do teacher language interactions occur in preschool classrooms? Which children in a classroom are involved in fewer or briefer teacher language interactions, as compared to their peers? Do teacher language interactions change over the course of the year?

Are teacher language interactions associated with growth in children’s English language skills? Monthly data are collected in preschool classrooms using sensing technology tools to address these questions. These sensing technology tools allow for the identification of language interactions, or moments when a teacher and child are in proximity to each other and are talking.

Information is also collected on children’s language skills, temperament, teacher-child relationships, and demographic information. The outcomes of this project are expected to inform models of language development in preschool-age children and provide fine-grained data that can be used by educators to equitably support language development in preschool settings.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.