Balancing Granularity and Context in Designing Evidence Models for Stealth Assessment: A Case of Pause Distribution in a Digital Game-Based Learning Environment

Abstract

<p> In the context of game-based learning environments, stealth assessment unobtrusively measures learning outcomes. However, designing a robust evidence model requires careful consideration of learning behaviors in context. This chapter presents a case on designing evidence models for stealth assessment in a game-based learning environment. In this study, the authors employ a finite mixture model to analyze players’ pause behaviors in the game, interpreting the meanings of these behaviors under different contexts. The findings contribute to the understanding of how to balance granularity and context in designing evidence models, proving valuable for enhancing the efficacy of stealth assessments for game-based learning. <br></p>