Phenomenography as a Tool for Investigating Understanding of Computing Concepts

Computing has become a foundational subject across the engineering disciplines and offers significant opportunities both in practice and from an educational perspective. Maximizing this potential requires deep understanding of how students learn and apply computing concepts. There has been a great deal of work exploring understanding in computing education, focused primarily on what constitutes knowledge in computing and the processes engaged to utilize this knowledge in solving computing problems. There is also a sizable body of work exploring the misconceptions held by novices in computing education. However, little work has been done exploring the types of conceptions that computing students hold for the fundamental computing concepts apart from identifying misconceptions. Uncovering the different types of conceptions held by students independent of specific computing languages or environments is essential to understanding how students learn computing concepts and ultimately to develop better pedagogical and assessment techniques. Phenomenography is a research methodology uniquely designed to uncover the different conceptions held by individuals about a given concept because the main tenet of phenomenography is that any phenomenon can be understood or experienced in a limited number of qualitatively different ways. Thus, the goal of phenomenography is to uncover those different ways of understanding. In recent years, phenomenography has begun to be used to explore the way that students experience the act of learning to program, both from a procedural and object-oriented perspective. However, it has not been used to explore the understanding held by individuals of specific concepts in computing. This paper describes how phenomenography was employed to explore the fundamental computing concepts of conditional and repetition structures. In addition, a discussion will be presented on how the results of this study, along with follow-on studies employing this methodology exploring additional fundamental programming concepts, can lay the groundwork for the development of language and computing environment independent assessment instruments. These instruments are needed for valid assessment and comparison of the pedagogical variations inherent in using the variety of programming languages, environments, and paradigms available today.