Abstract : While instructional scaffolding is widely recognized in Second Language Acquisition (SLA), its synchronous integration and the subsequent effects of scaffold fading in Teaching Chinese as a Second Language (TCSL) remain underexplored. This study proposes and empirically validates a multi-tiered scaffolding model—comprising linguistic, practice, and experiential scaffolds—tailored for TCSL oral instruction. A quasi-experimental, mixed-methods design was employed over a 15-week semester with 60 Vietnamese L1 learners. Quantitative data from pre- and post-tests, alongside pragmatic role-plays across three time points, demonstrated that the intervention significantly enhanced oral proficiency, yielding a massive effect size (Cohen’s d = 2.15) and effectively mitigating L1 tonal interference. Repeated Measures ANOVA confirmed the accelerated proceduralization of communicative competence (partial η² = .52). Furthermore, linear regression analysis of post-intervention survey data revealed that systematic scaffold fading is a significant predictor of sustained pragmatic adaptability and learner autonomy (R² = .41, p < .001). These findings advance Vygotskian Constructivism by providing a cohesive empirical evaluation framework, underscoring the necessity of intentional pedagogical withdrawal to cultivate an autonomy-driven learning ecosystem in global language education.
Thien et al. (Mon,) studied this question.