Data science knowledge integration: Affordances of a computational cognitive apprenticeship on student conceptual understanding

Abstract This study implements a computational cognitive apprenticeship framework for knowledge integration of Data Science (DS) concepts delivered via notebooks. also explores students' conceptual understanding the unsupervised Machine Learning algorithm K‐means after being exposed to this method. The learning DS methods and techniques has become paramount new generations undergraduate engineering students. However, little is known about effective strategies support student machine (ML) algorithms. research questions are: How do students conceptualize their an ML method engaging with interactive visualizations designed using approach? affordances or hinder method? Design‐based allowed iterative design, implementation, validation pedagogy in context working classroom. For this, data collection often take form artifacts. We performed qualitative content analysis written responses reflections elicited during process. Results suggest that promoted integration. After interacting notebooks, most had accurate conceptions goal nature identified factors affecting output algorithm. Students found it useful have concrete representation method, which supported its showcased acquisition strategic appropriate execution. we important misconceptions held