Formative Assessment in College English Writing Instruction: Effects on Revision Quality and Learner Autonomy

Writing remains one of the weakest productive skills among Chinese EFL learners, and teacher grading under a summative assessment regime often fails to translate into substantive revision on the part of students. This study employed a quasi-experimental, mixed-methods design to examine the effects of formative assessment on revision quality and learner autonomy in college English writing instruction. Over a 16-week semester, two parallel classes received either a formative assessment intervention (combining transparent assessment criteria, teacher formative feedback, peer review, and reflective journals) or conventional teacher grading. Results indicated that (1) the experimental class significantly outperformed the control class on the "global revision" dimensions of content improvement and discourse organization, whereas no significant between-group difference was found in "local" language corrections; (2) learners in the experimental class showed significant gains in autonomy, particularly in goal setting and self-monitoring; and (3) interview and journal data suggested that the internalization of assessment criteria and the "reader awareness" activated through peer review served as key mediators of autonomy growth. The study argues for an operational formative assessment cycle feasible in large-class EFL writing instruction, one that shifts students from "correcting errors" to "revising meaning."