Using assessment questions based on Bloom’s taxonomy to generate personalised feedback for students

The feedback provided to students in education is essential for acquiring knowledge and skills. However, giving personalised feedback can be a very time-consuming process. Therefore, the automation of providing personalised feedback based on student responses is critical. This study aims to analyse students’ exam responses with the goal of identifying potential building blocks for giving personalised, response-dependent feedback. The study was conducted within the context of a computer science curriculum, where Bloom’s taxonomy was used in formulating assessment questions. This allows for a structured approach to evaluating students’ cognitive skills at various levels, from basic recall to higher-order thinking. The alignment of questions with Bloom’s taxonomy was evaluated through statistical analysis. As expected, correctly answering lower-level questions according to Bloom’s taxonomy increased the likelihood of correctly answering more challenging, complex questions. Thus, students who performed well on lower-level questions were more likely to succeed in answering more complex ones. This finding highlights the importance of designing assessments that reflect the progressive nature of cognitive development. The results of this study can be used in the development of a domain model for learning and teaching, which will serve as the foundation for personalised feedback systems.