Abstract
MCQs of the multiple true/false (MTF) variety were widely used in summative assessment 25 years ago. They could test a number of skills in addition to recall of factual knowledge, and were reliable, discriminatory, reproducible and cost-effective. However, there are now considerable doubts about their construct validity, mainly because of the varying responses of examinees to negative countermarking and the ‘don’t know’ option, and the strategies they use when sitting examinations. Extended matching and one-from-five questions are now preferable, and negative countermarking is outmoded. MTF questions are still valuable in formative assessment and revision but are not recommended for summative examinations.