The Interplay among EFL Students' Epistemic Beliefs, Language Learning Strategies, and L2 Motivational Self-System: A Structural Equation Modeling Approach

Examining learners' beliefs about the essence of knowledge, how they are conceptualized, and the ways they influence the learning process have gained attention in the second language (L2) learning. This study employed a multivariate statistical framework to model complex relationships among three constructs, i.e., epistemic beliefs (EB), language learning strategies (LLSs), L2 motivational self-system (L2MSS), and their sub-factors (N = 12). The data were collected in two phases. At the preliminary stage, the structural equation modeling (SEM) was conducted to visualize a hypothesized model and to map the conceptual framework of the study. At the secondary step, three questionnaires, EB, LLSs, and L2MSS, were distributed among junior high school students (N = 300). The questionnaires were collected over four months. The collected data were screened for incomplete responses and sample attrition. Notably, 774 questionnaires met a valid response rate of 95 percent. Correlational and SEM analyses were utilized to probe the causal relationships among the constructs. The findings revealed that there was a significant positive relationship between the subscales of L2MSS and LLSs. However, there was a significant negative relationship between EB and LLSs. Besides, confirmatory factor analysis underpinned the fitness of the hypothesized model after two stepwise corrections. The findings revealed that the path coefficient for EB had a significant impact on the LLSs with the mediating role of L2MSS. In particular, the path analysis revealed that 48% of LLSs might be explained through EB and L2MSS. This paper suggests that the more EB students experience, the less likely they adopt learning strategies, and the more L2MSS they hold, the more LLS students employ.

Keywords:

Epistemic Beliefs , Language learning strategies , Motivational Self-System , Second Language Learning , Structural Modeling Approach