Removing Lectures from a Computer Programming Course - A Quantitative Study

Computer programming is a discipline that is becoming increasingly important to today’s engineering practice and society overall, and it is used extensively and intensively in several fields. Examples can be found in mechanical, electrical, or chemical engineering, and many other areas. As such, it is common that higher education institutions teach at least one basic programming course on the subject in every engineering degree program. This means that programming is taught to a variety of engineering students with significantly different backgrounds and expectations. Consequently, it is important that all students — in particular those that had less prior exposure to computational methods and programming in their past — get sufficient time to gain experience with their programming tools. Lectures are often chosen as the standard teaching method when designing a course structure. This also applies to programming courses. Two main reasons for this are the effectiveness of informing a large group of students in a very short period of time, and the efficiency of doing so while involving only a single teacher. However, we believe that programming is much better taught and learned with a ''hands-on'' approach. Therefore, we argue that lectures can be safely removed from programming courses and replaced with extended lab exercise sessions where teaching assistants circulate and interact with the students individually when needed. This gives students additional time to build up experience with the programming environment consisting of the programming language, interface, and associated tools. We support our argument with a discussion of both quantitative metrics and a summary of qualitative statements drawn from programming course evaluations and student feedback. These evaluations were gathered in courses over a series of semesters — before and while the course followed our suggested model. We conclude from the observed data that removing lectures from the course not only increased overall student satisfaction, but also bolstered the learning outcome. Furthermore, we show that an appropriate adaptation of better teaching material, after having removed the lectures, yielded further improvement in these categories.