Scale validation and latent profile identification of GenAI competence for pre-service second language teachers
Hanwei Wu, Yongliang Wang, Gurpinder Singh Lalli
Abstract
Abstract Pre-service second language (L2) teachers can benefit from Generative Artificial Intelligence (GenAI), yet there is a lack of reliable instrument to evaluate their GenAI competence. To fill this void, two studies were conducted. Study 1 aimed to create a GenAI Competence Scale for Chinese Pre-service L2 Teachers. Based on exploratory factor analysis of 350 samples, a 21-item scale was developed with a three-factor structure: Awareness and Willingness, Knowledge and Application, and Social Responsibility. Confirmatory factor analysis of 358 samples verified the scale’s ideal model fit, along with its high validity (convergent, discriminant, and criterion-related), reliability, and cross-gender invariance. Study 2 utilized Latent Profile Analysis to identify three distinct profiles among 708 Chinese pre-service L2 teachers’ GenAI competence: (i) moderate levels of Awareness and Willingness, relatively low Knowledge and Application, and moderate Social Responsibility (C1: 22.74 %), (ii) moderately high levels of Awareness and Willingness, relatively low Knowledge and Application, and moderately high Social Responsibility (C2: 65.82 %), and (iii) high levels of Awareness and Willingness, moderate Knowledge and Application, and high Social Responsibility (C3: 11.44 %). We anticipate that future scholars will adopt this scale across diverse L2 education settings, conducting some in-depth explorations to enhance the generalizability of findings, deepen the understanding of GenAI competence among pre-service educators, and contribute to the advancement of relevant theoretical frameworks and practical applications.