Proving what is thought: The efficacy of MCQs

Abstract

Multiple choice questions (MCQs) have enjoyed a renaissance. Online technology has enabled easy and personalised delivery; immediate feedback to users in response to their answers; large, crowd sources banks of questions; and adaptive selection of questions based on prior performance. This presentation considers how good MCQs can be identified. Good can be very subjective but based upon a clear understanding of the content knowledge and the pedagogic issues associated with question construction. The description of good is extended to how well the question behaves when answered by a large number of users. Item response theory (IRT) is a paradigm for the analysis and scoring of test instruments using the relationship between a person’s performance on a test item and the person’s overall performance level described in terms of probability. Finally, the full study, which identifies the relationship between computational thinking ability, responses to a self perception questionnaire and comparison with secondary data related to overall ability. The conclusions include: the computational thinking MCQs show a high degree of reliability; algebra and general school achievement and students’ perception are significant predictors of computational thinking performance; and there is no gender difference noted in computational thinking performance and perceptions of computational thinking.

More information

Identifiers

ORCID for John Woollard: orcid.org/0000-0002-4518-0784

Catalogue record

Export record