The Relationship of Lexical Richness to the Quality of ESL Learners’ Oral Narratives

This paper examines the relationship of lexical richness to the quality of ESL learners’ oral narratives. A computational system is designed to automate the measurement of three dimensions of lexical richness, i.e., lexical density, sophistication, and variation, using twenty-five different metrics proposed in the language acquisition literature. This system is used to analyze large-scale data from the Spoken English Corpus of Chinese Learners (Wen, Wang, & Liang, 2005) together with the vocd utility of the Computerized Language Analysis programs (MacWhinney, 2000), which offers an additional measure of lexical variation, the D measure (Malvern, Richards, Chipere, & Durán, 2004; McKee, Malvern, & Richards, 2000). This comprehensive analysis allows us to identify measures that correlate strongly with the raters’ judgments of ESL learners’ oral narratives as well as to understand the relationships between these measures. This research provides ESL teachers and researchers with a robust tool for assessing the lexical richness of ESL language samples and insights into how lexical richness measures may be effectively used as indices of the quality of ESL learners’ speaking task performance.