An Individualized Vocabulary Intervention for Dual Language Learners

University Of Colorado, Boulder CO

Investigators

Abstract

Project Summary/Abstract The goal of this proposed project is to examine the feasibility of an individualized vocabulary intervention program for preschool dual language learners (DLL) from low socioeconomic (SES) backgrounds. Children who grow up in low SES and language minority homes (L1) and learn English (L2) as a second language in school settings are likely to be at risk for reading difficulties and poor academic performance (e.g., August et al., 2006). In order to serve the particular needs of DLLs from diverse backgrounds, scientific evidence is critically needed about the intervention strategies for these preschoolers. In this proposed study, we examine the feasibility of using machine learning methods to generate individually tailored interventions for low SES dual language learners who learn two typologically different languages, Cantonese (L1) and English (L2). Two important strategies will be used in this study. First, a computation model will be built to predict and select appropriate bilingual target words for individual DLLs. Second, this intervention will be integrated into the extant preschool curriculum, thus resulting in a potentially sustainable, scalable approach to decreasing language proficiency gaps. There are two specific aims in this proposed study: 1. Model normative lexical development in Cantonese-English DLLs. We will leverage data previously collected by Dr. Kan on Cantonese-English vocabulary development at Head Start Centers, and computational models of typical lexical development in monolingual English speakers, to build a computational model of typical bilingual lexical development in Cantonese-English dual language learners. 2. Evaluate the feasibility and effectiveness of a model-based individualized vocabulary intervention program. We will use the computational model to make individual level target word recommendations for 200 Cantonese-English DLLs, and work with teachers at 8 Head Start centers to integrate the recommendations into their existing curriculum.