Abstract:

Although many countries have not yet included Computational Thinking (CT) into their national school curricula, CT is an integral skill to obtain for young people in the 21st century (Kong & Abelson, 2019). Key aspects of CT encompass – but are not limited to – abstraction, algorithmic thinking (AT), debugging and generalisation (Bocconi et al., 2016). In countries where teaching CT is not mandatory in K-12 education, students might acquire these skills only implicitly, for example in STEM classes.

Individually interviewing three students from a local German enrichment program for gifted students in the field of computer science, we aimed to analyse which, if any, CT related skills can be identified in their approach to a graph problem. Their heightened interest in computer science lead us to the assumption that they would demonstrate at least basic skills in each targeted area of CT. During the interview, students worked with a street map on which cities and connecting streets as well as travelling times for each street were indicated. Participants were asked a series of questions ranging from simple descriptive tasks (“What can you see on the picture?”) to complex problem-centred questions (“Can you name a strategy to find the shortest path between any two cities?”).

All three students struggled to remove unnecessary details from the map to find a more general representation, e.g. in the form of a graph, even though two of them already had basic knowledge of the concept of graphs. When asked to find a strategy for the shortest path, their first approach was to compare all possibilities, however, all of them realised that that was not an efficient solution. All three participants were able to refine their first idea during the course of the interview and, to some extent, correct wrong assumptions they had made at first. At the end, one student even intuitively described a procedure resembling the steps of Dijkstra’s Algorithm. In sum, the participants demonstrated advanced skills in AT, debugging and generalisation but only basic abstraction skills. As abstraction is essential to CT, if not even its most integral part, and also relevant to many other areas such as mathematical modelling, this skill should be examined more closely in future work.

Keywords:

Computational Thinking, Algorithmic Thinking, Abstraction, Graph Problems, Shortest Path Algorithms, STEM education.