Eliciting and Measuring L2 Metaphoric Competence: Three Decades on from Low (1988)

Abstract

In 1988 and 2006, Applied Linguistics published Low’s and Littlemore and Low’s seminal theoretical accounts of second language (L2) metaphoric competence (MC). Meanwhile, attempts to elicit metaphor-related skills/competences have been mixed. Instrumentation has varied in reliability, been limited in scope, and used arguably flawed reliability coefficients (McNeish 2018). Factor analysis, used in first language (L1) MC and other areas of L2 research (Loewen et al. 2009; Plonsky and Gonulal 2015), has not been used to explore latent L2 MC variables. To address these issues, we developed a large battery of MC tests to elicit Low/Littlemore’s constructs, administering it to 112 L1 Mandarin speakers of L2 English and 31 L1 English speakers. Data cleaning revealed some operationalization challenges, but resulting overall reliability was high and demonstrated innovative use of ordinal omega as a powerful alternative to Cronbach’s alpha, the most common instrument reliability index in L2 research (Plonsky and Derrick 2016). Exploratory factor analysis suggested four latent L2 MC variables: Productive Illocutionary MC, Metaphor Language Play, Topic/Vehicle Acceptability, and Grammatical MC, broadly supporting Low/Littlemore’s proposals.