RESUMEN
Multiple choice question papers in which the student has a 'don't know' option are widely used in undergraduate and postgraduate examinations in Medicine. In the present study students' performance in papers with a 'don't know' option has been compared with their performance when they are instructed to answer all the questions. By completing questions left unanswered (i.e. 'don't know' options) students were able to increase their score significantly and the rank order of the students in class is changed. Answers omitted may indicate complete ignorance or various degrees of knowledge. It is concluded that the 'don't know' option in multiple choice question papers favours the bold and test-wise student and in consequence their validity as a measure of achievement may suffer. Papers in which the student is instructed to answer all the questions are to be preferred to those in common use where there is a 'don't know' option.
