Elimination marking: A new standard for multiple choice exams?

Erasmus University Rotterdam and other Dutch universities employ standard setting (SS) for their multiple choice (MC) examinations. SS invites students to answer all exam questions but requires them to meet a raised standard because they might get some answers right by chance. SS thus encourages students to guess, which is questionable from a pedagogical and measurement perspective. It gives the signal that it is better to guess than to admit that one is uncertain or unknowledgeable. Guesses also introduce noise, deflating the accuracy, reliability, and informativeness of the exam scores. 

Steven Verheyen will present the results of a study investigating whether elimination marking (EM) is a viable alternative for SS. In EM, examinees eliminate the answer alternatives they consider mistaken. It rewards students for displaying partial knowledge, does not penalize them for expressing doubt, and allows for more differentiated feedback. Steven will discuss the relative benefits and downsides of both scoring methods with a focus on their susceptibility to individual differences in risk adversity that are related to gender and nationality.