Dataset on the Effects of Agent Type and Feedback Style on Self-Directed Learning

This dataset contains experimental data from a study investigating the effects of AI agent type (customized vs. general-purpose) and feedback style (directive vs. Socratic) on self-directed learning. A total of 51 postgraduate students participated in a 2 × 2 mixed factorial design experiment, completing two instructional design tasks over two consecutive weeks.
The dataset includes:
Feedback quality (accuracy, specificity, relevance, etc.)
Self-regulatory behaviors (task orientation, comprehension monitoring, feedback regulation)
Learning experience (cperceived usefulness, cognitive load, etc.)
Learning outcomes (gain scores between initial and revised drafts)
Data are structured to support analyses of main effects, interaction effects, and qualitative feedback coding. The dataset can be used for research on AI-assisted learning, instructional feedback evaluation, self-regulated learning, and human-AI interaction in educational contexts.