Examining the comparability of parallel English and Chinese elicited imitation tasks

Abstract

The suite of elicited imitation tasks (EITs) developed originally by Ortega et al. (2002) has been validated as an effective measure of second language (L2) proficiency and adopted in different L2 studies. However, the comparability of the crosslinguistic parallel EITs, all developed based on the English version, has yet to be explored. This exploratory study investigated the comparability of Ortega et al.’s parallel English and Chinese EITs by analyzing L2 English data collected from 82 L2 English participants in Wu et al. (2022) and L2 Chinese data collected from 80 L2 Chinese participants in Wu and Ortega (2013). Despite the crosslinguistic variation observed between the English and Chinese EITs, the converging evidence gathered from internal consistency, item discriminatory power, correlations with external proficiency measures, and abilities to discriminate four speaking levels indicates that the two parallel EITs tap roughly the same language proficiency components in each language. Crucially, the classification drawn from the same CAF measures across both languages suggests that the five proficiency levels of the EIT scores (i.e., Novice = 0–24; Low = 25–48; Intermediate = 49–72; High = 73–96; Advanced = 97–120) can consistently discriminate language abilities into distinctive levels across both languages. We hope the proposed five proficiency levels will enhance the application and interpretation ability of Ortega et al.’s parallel EITs in crosslinguistic, bilingual, and multilingual research.