Effects of feedback on deep-level features of argumentative writing over multiple drafts: Insights from an intervention study with secondary EFL students

Abstract Providing feedback on writing is a vital task in language education, yet its time-consuming nature may lead to limited feedback in large language classes at secondary schools. Rubric and exemplar feedback is more time-efficient than traditional in-text comments, yet its effect on written argumentation requires further investigation. Our experimental study ( N = 202 EFL students) addressed this research lacuna and examined the effects of a learning unit with three types of written formative feedback (EG1: rubric + exemplar, EG2: in-text comments, EG3: rubric + exemplar and in-text comments) on written argumentation. EGs were compared to a control group, which received the learning unit without additional feedback. We used an analytic rating to assess the development of deep-level features (structure and coherence) in 808 argumentative essays (across four measurement points comprising writing and revision phases). Results revealed progress for all groups, with EG1 (rubric + exemplar) outperforming the control group at measurement point three and outperforming EG2 and the control group at measurement point four. Progress regarding rebuttals was largest in EG1. Our findings indicate that rubric + exemplar feedback benefits written argumentation, in particular the inclusion of rebuttals, and that more feedback does not necessarily result in more writing improvements.