Answers are grounded in this paper and cite their sources.
Traditional second language (L2) Korean writing assessment relies on periodic scoring and static feedback. This study implements a closed-loop web-based assessment system that collects writing frequency, number of edits, and feedback responses to model competency in real time. Intact classes of 200 intermediate L2 Korean learners were randomly assigned to experimental (n = 100) or control (n = 100) conditions for six weeks. High-engagement students in the experimental group (n = 46) gained 15.2 points on average, versus 4.7 points for low-engagement students (n = 54) and 2.8 points for the control group (F(2,197) = 124.4, p < 0.001). The interaction between writing frequency and feedback responses significantly improved model fit (ΔR2 = 0.03, p < 0.001). The effect of the system depends on whether students participate in writing exercises and engage with feedback, and the results can only be interpreted as stable correlations rather than strict causal effects.
No takes yet. Share an insight, caveat, or question.
Bo Sun (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: