Interaction Effects of Feedback Explicitness, Target Structure Complexity, and Learner Proficiency on Written Corrective Feedback Effectiveness

Despite extensive research on written corrective feedback (WCF), prior studies have largely examined feedback, learner, and linguistic variables in isolation, with limited attention to how these factors are modeled over time. Addressing this gap, this study examined feedback explicitness, learner proficiency, target-structure type/complexity, and measurement time in relation to controlled L2 grammatical accuracy. Using a fully crossed factorial design, 180 Mandarin L1 EFL learners of differing proficiency levels were assigned to four feedback conditions or a control group and completed text-reconstruction tasks targeting three grammatical structures across five measurement points: pretest, immediate posttest, 8-week delayed posttest, 18-week delayed posttest, and an additional posttest three weeks later. Data were analyzed in R using mixed-effects modelling. Model comparison indicated that an AR(1) repeated covariance model provided the best fit. Results showed a significant main effect of time, but feedback condition, learner proficiency, target structure, and feedback-related interactions were not statistically significant. Descriptive means showed that the control group had the highest baseline accuracy, and gain-score summaries indicated broadly similar improvement across conditions. The findings therefore support a cautious interpretation: controlled grammatical accuracy changed over time, but the data did not show statistically distinguishable developmental trajectories across feedback conditions. Pedagogically, the results suggest caution in assuming that greater feedback explicitness necessarily produces superior accuracy gains, particularly when baseline differences and longitudinal dependency are taken into account.