Effective feedback with adaptive questions


  • Dick Bacon University of Surrey


Adaptive e-questions designed to aid learning have been implemented for a first year electronics course. The questions are designed to help students who answer incorrectly by recognising the mistake made (by means of mal-rules where appropriate) and then providing constructive targeted feedback and further tries at the question before moving on to the next. A feature of the system is that the next try can be attempted whilst the feedback from the first try is still being displayed. This is aimed at maximising the chances of the feedback being both read and applied. The questions, some choice-based but mostly numeric, all contain randomised parameters so that students can be encouraged to work together without major plagiarisation issues. The invitation to have further tries at the question appeared to act as a strong incentive to apply the feedback; very few students failed to take advantage of the opportunity to re-try a question when it was offered. The questions have been prepared for use with an assessment system that is a direct implementation of a new question-test interoperability specification. The features used, such as adaptive questions, randomisation of parameters, multiple-response conditions and alternative feedback messages are therefore likely to become more frequently available in future versions of commercial assessment systems. This in turn will support a wider use of such questions aimed at effective learning within formative assessments. In the case reported here, the combination of prompt, targeted feedback and an immediate opportunity to apply the new information led to a rise in (eventual) correct answer rate from 58% to 85%.