Answers are grounded in this paper and cite their sources.
Purpose This study tests a dual-process Self-Determination Theory (SDT) model in L2 classrooms by examining engagement and disengagement concurrently, retaining their multidimensional structures, and investigating their higher-order and facet-level associations with end-of-course achievement.Design/methodology/approach In required College English courses in China, Phase 1 (N = 1,088) used structural equation modeling to examine cross-sectional associations among perceived autonomy support and teacher control, need satisfaction and frustration, and higher-order engagement and disengagement. Phase 2 used an availability-based subsample (N = 237) to examine achievement associations in separate engagement and disengagement models at higher-order and facet levels, using final examination scores as the outcome.Findings Autonomy support was associated with need satisfaction, which was associated with engagement; teacher control was associated with need frustration, which was associated with disengagement. Higher-order engagement positively predicted achievement, whereas higher-order disengagement negatively predicted it. In separate facet models with four facets entered simultaneously, no engagement facet uniquely predicted achievement, whereas behavioral disengagement was the only disengagement facet showing a unique negative association. These findings suggest a provisional cross-level asymmetry: unique facet-level information emerged for disengagement but not for engagement.Originality/value The study contributes to understanding pedagogical innovation by identifying complementary learner experiences and forms of engagement and disengagement that can inform the evaluation and adaptation of classroom practices. Its findings highlight the value of considering both broad classroom involvement and specific withdrawal when identifying support needs. Autonomy-supportive teaching and supportive responses to recurring behavioral withdrawal offer directions for instructional adaptation that require testing in intervention research.
No takes yet. Share an insight, caveat, or question.
Zhou et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: