How (not) to write an introductory programming exam

Abstract

The computing education literature shows some recent interest in summative assessment in introductory programming, with papers analysing final examinations and other papers proposing small sets of examination questions that might be used in multiple institutions as part of a benchmarking exercise. This paper reports on a project to expand the set of questions suitable for use in benchmarking exercises, and at the same time to identify guidelines for writing good examination questions for introductory programming courses – and, by implication, practices to avoid when writing questions. The paper presents a set of ten questions deemed suitable for use in the exams of multiple courses, and invites readers to use the questions in their own exams. It also presents the guidelines that emerged from the study, in the hope that they will be helpful to computing educators writing exams for their own courses.