Simple feedback can do the job: Analyzing the effects of simple computer based feedback for fundamental programming tasks

摘要

Although, the importance of Computer Science has been increasing, it must be noted that the drop-out rate among students of STEM subjects unfortunately is among the highest we know in education, worsen the fact that the enrollment rates are relatively small. Thus, we might have to look closer at pedagogical approaches to overcome this problem. The New Zealand educator John Hattie introduced a meta-study called “Visible Learning” in which he mentions feedback to be one of the major factors affecting the success of learning. However, its influence on fundamental programming tasks cannot be investigated properly unless objective, reliable and valid tasks exist that allow a clear inference of the observed performance for the given feedback. For example it would be possible to argue that observed educational “improvements” are in fact the result of a reliability or validity problem. Since tasks for programming fundamentals were often created intuitively, it had to be explored how to get objective, reliable and valid tasks. After their development, an empirical study has shown that simple feedback can indeed make a contribution to the successful learning of programming fundamentals if the corresponding tasks are less complex according to Anderson's taxonomy.