Answers are grounded in this paper and cite their sources.
Interactive virtual makerspaces generate process data, but platform affordances can complicate their interpretation as evidence of computational thinking (CT). This exploratory study presents the Design Affordance Diagnostic Framework (DADF), which evaluates indicators through five gates: variance, reliability, association with outcomes, robustness to non-independence and multiple-testing control. Artifacts from two cohorts of pre-service teachers completing a Roblox game design activity were examined. Using LLM-coded CT outcomes, network density showed a positive unadjusted association in 2024 but was not statistically significant when pooled (r = .16, n = 55). Reflection depth showed a positive pooled association (r = .36, n = 55) and remained provisional because of coding reliability limitations. In 2025, logic density showed a positive unadjusted association (r = .40, n = 25) and was retained as provisional; code volume was not statistically significant (r = −.18, n = 25). Available revised-analysis records do not establish cluster-robust or multiplicity-adjusted inference, so full passage through all gates is not established. DADF provides a procedure for identifying indicators that require further validation across cohorts, instruments and platforms.
No takes yet. Share an insight, caveat, or question.
Moon et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: