A Method for Creating Assessment Items for Computer Science

This paper presents a method for creating assessment items in Computer Science courses. The method is based on Bloom's Taxonomy of Educational Objectives in the Cognitive Domain. The paper shows how the courses in a Computer Science curriculum are divided into the following three categories: problem-solving, theory-based, and both problem-solving and theory-based. A table has been devised recommending a percentage of the questions in an assessment instrument that should target one of the three main cognitive levels of Bloom's Taxonomy. Another table gives examples of questions that can be asked along the different cognitive dimensions. The method has been used for the past three years in an undergraduate Computer Science programme offered by a well-established university in a developing country. Overall, the performance of students in theory-based and problem-solving/theory-based courses has improved. However, the performance of students in problem-solving courses has not improved significantly. This suggests that there are other factors that influence performance in these types of courses which need to be addressed.