Having helped compile and use multiple-choice questions for computer-marked assessments with undergraduates for more than 20 years ("Computer marking splits unis", THES, June 4). I feel that well-written MCQs are an excellent medium for student learning but have limited assessment value.
They measure outcome, not the process by which it is achieved. There is no facility to discriminate between correct answers selected through sound reasoning and those picked by guesswork. If process is to be as important as outcome in assessment, MCQs should be used with caution.