An Evidence-Based Approach to Distractor Generation in Multiple-Choice Language Tests

The purpose of this project is to explore the feasibility a new approach for producing evidence-based distractor sets. We use Common Wrong Answers (CWAs), and associated performance data, generated by candidate responses open gap-fill tasks, produce sets multiple-choice tasks based on same texts. then investigate whether these are effective in language tests, terms empirical qualitative review, consider potential impacts production process test material. This explores innovative method content development, raises possibility item that can semi-autogenerate items shorter periods time without affecting quality or reliability. Although specific one task type, it hoped further research will expand applications deliver version may be operationalised across different types development assessments.