Affinity propagation clustering of gameplay attempts yielded an interpretable six-cluster solution (part.6)¶

Claim · All claims
Evidence · 1 study · associational r? · q2
Currency · newest study 2017 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person · overdue for a recheck since 2020: tool and model accuracy changes quickly

Subclaims¶

q2 i? AP clustering converged on 145 proto-clusters, which agglomerative collapsing reduced to six clusters at a 0.905 altitude cut, with a heat map showing three well-delimited cohesive clusters plus internal structure. → Martinez-Garza 2017

Evidence¶

Martinez-Garza 2017¶

Martinez-Garza, M. M., & Clark, D. B. (2017). Investigating Epistemic Stances in Game Play with Data Mining. International Journal of Gaming and Computer-Mediated Simulations, 9(3). https://doi.org/10.4018/IJGCMS.2017070101

q2 · i? · associational · r?

Unsupervised clustering of Study 2 attempts on seven theoretically-selected variables using Gower's similarity coefficient and affinity propagation. The algorithm "converged on a set of 145 proto-clusters after 260 iterations"; visual inspection of the dendrogram suggested a cut at 0.905 altitude yielding the part.6 six-cluster solution.

"The AP clustering algorithm converged on a set of 145 "proto-clusters" after 260 iterations."

Discussion¶

Related Claims¶

• Students favoring fast iterative solution strategies achieved lower learning gains than students with slow, elaborate, or balanced play
• K-Means++ clustering of self-reported behavioral proxies yields six distinct, stable learner profiles (k = 6, Silhouette = 0.62)
• Exploratory K-means clustering identifies four heterogeneous student AI-user profiles differing in experience, competence, usefulness, trust, and control, but the solution's optimality and stability are not established
• K-means clustering of K–8 students' platform trace data yields seven distinct engagement profiles
• Cluster analysis identifies three stable learner profiles: Engaged and Literate, Uncritical Reliant, and Skeptical Minimal users
• Cluster analysis across both studies identifies three meta-attention knowledge profiles, including an overconfident group with a self-knowledge/strategy mismatch
• Affinity for technology interaction differed across clusters, with text reproducers highest and integrators scoring higher than reconstructors (d = 0.59)
• AI-supported argumentative writing yields three distinct uptake clusters (text reproducers, text integrators, text reconstructors) based on recurring n-gram overlap