The second language trait emotional intelligence (L2-TEI) scale and its relation to language engagement

This study tested the psychometric properties and factorial validity of the language domain-specific trait emotional intelligence (L2-TEI) scale among 415 language learners and assessed its criterion validity in predicting language engagement. The scale consistes of four factors-emotionality (EM), self-control (SC), wellbeing (WB), and sociability (SO). Four models we considered to validate the scale: confirmatory factor analysis (CFA), exploratory structural equation modeling (ESEM), bifactor CFA, and bifactor ESEM were employed to evaluate the factorial validity of the L2-TEI scale. The findings showed that ESEM provided more valid results than CFA with the best fitted model being the bifactor ESEM. The bifactor ESEM has substantiated four L2-TEI specific factors while accounting for the global factor of TEI. The test of criterion-related validity indicated that the structural bifactor ESEM explained 73 % of the variance in learner engagement as an outcome variable, with the global TEI and the specific factors of SO and WB being the predictors of language engagement. These findings support the validity of the L2-TEI scale for assessing students' TEI in the L2 domain. Finally, the findings highlight the imprtance of adopting specific trait emotional intelligence to understand students' language learning outcomes such as engagement.