Impacts of computer-assisted diagnostic assessment on sustainability of L2 learners’ collaborative writing improvement and their engagement modes

Diagnostic assessment (DIA) is under-researched in second/foreign language education despite its common practice across a wide range of professions such as medicine, mechanics, and information technology. This study aimed at exploring the impact of computer-assisted DIA on the sustainable improvement of English-as-a-Foreign-Language (EFL) learners’ collaborative writing and engagement modes. To do so, 36 selected EFL university students were paired to collaborate on writing data commentary tasks whose performances were subjected to the teacher’s regular DIA in Google Meet and task-wise student reflective logs over ten weeks. Repeated-measures ANOVA indicated dyads’ significant progress in lower-level writing skills (sentence structure, word choice & grammar, mechanics), but no considerable progress in higher-level writing skills (organization and development). The results of one-way ANOVA showed the DIA impact on individuals’ sustainable writing improvement from the pretest to the immediate and delayed posttests. Students’ reflective logs were analyzed to explore their behavioral, emotional, and cognitive modes of engagement in computer-assisted DIA. Theme frequency analysis indicated the participants’ active behavioral engagement in terms of their notable amount of time per task. They conveyed their emotional engagement in the user-friendliness of Google Meet, real-time social presence, and low anxiety experienced with DIA. Their cognitive engagement was depicted by their major approval of DIA and positive self-assessment of writing improvement. Yet, some participants were critical to the prioritization of language form(s) over the content in teacher DIA practice. This study yielded pedagogical implications for the L2 teachers to blend DIA, task-based academic writing, and student collaboration in e-learning contexts.