Algorithms used to classify new observations into a set of categories based on a training set of data in which categories are known.
Published in Chapter:
Advancing Research in Game-Based Learning Assessment: Tools and Methods for Measuring Implicit Learning
Elizabeth Rowe (EdGE at TERC, USA), Jodi Asbell-Clarke (EdGE at TERC, USA), Erin Bardar (EdGE at TERC, USA), Ma. Victoria Almeda (EdGE at TERC, USA), Ryan S. Baker (University of Pennsylvania, USA), Richard Scruggs (University of Pennsylvania, USA), and Santiago Gasca (EdGE at TERC, USA)
Copyright: © 2020
|Pages: 25
DOI: 10.4018/978-1-7998-1173-2.ch006
Abstract
Digital games provide engaging opportunities to support and assess implicit learning—the development of tacit knowledge and practices that may not be explicitly articulated by the learner. The assessment of implicit learning reveals learning not captured by traditional tests and may be critical to meet the needs of a broad range of neurodiverse learners. This chapter describes tools and methods designed to build implicit game-based learning assessment (GBLA), where research-grounded automated detectors identify implicit learning in gameplay. The detectors are based upon theoretical and empirical underpinnings, including extensive hand-labeling. The authors present a detailed overview of a six-step process for emergent GBLA, which has been applied and refined across multiple game-based learning studies. This chapter also includes a description of the data architecture and tools the authors designed and developed specifically for this approach.