Data for this study were collected through a cross-sectional survey of 300 Vietnamese EFL university students across four year levels, providing a robust stratified sample for structural equation modeling (SEM). Conducted over a 10-week period, the research examined the predictive associations among peer scaffolding, academic writing motivation, writing self-efficacy, and academic writing performance. Predictor variables were assessed early in the semester using theoretically validated 5-point Likert scales. The dependent variable, academic writing performance, was operationalized as a latent construct measured by final course assignments evaluated by two independent raters, achieving an inter-rater reliability exceeding .85. Self-reported English proficiency was also included as a control variable. All statistical analyses were conducted using jamovi, beginning with normality diagnostics and progressing to exploratory (EFA) and confirmatory factor analysis (CFA) to ensure a rigorous measurement model. Subsequently, the structural model and indirect effects were tested using bootstrapping with 5,000 resamples. While exploratory multi-group analyses were conducted across gender and year levels, the cross-sectional nature of the design requires that all identified pathways be strictly interpreted as theory-guided associations rather than definitive causal effects.
Huong et al. (Thu,) studied this question.