Zusammenfassung

This paper presents a fine-grained process analysis of 22 students in a classroom-based learning setting. The students engaged (and failed) in problem-solving attempts prior to instruction (i.e., the Productive-Failure approach). We used the HeuristicsMiner algorithm to analyze the data of a quasi-experimental study. The applied algorithm allowed us to investigate temporally structured think-aloud data, to outline productive and unproductive problem-solving strategies. Our analyses and findings demonstrated that HeuristicsMiner enables researchers to effectively mine problem-solving processes and sequences, even for smaller sample sizes, which cannot be done with traditional code-and-count strategies. The limitations of the algorithm, as well as further implications for educational research and practice, are also discussed.

Hartmann, C., Rummel, N., & Bannert, M. (2022). Using HeuristicsMiner to Analyze Problem-Solving Processes – Exemplary Use Case of a Productive-Failure Study. Journal of Learning Analytics, 9(2), 66–86. DOI:10.18608/jla.2022.7363