Theoretical considerations for game-based e-learning analytics

Abstract

In an interactive digital game or gamified e-learning experience, mapping a learner’s progress, problem-solving attempts, self-expressions and social communications can entail highly detailed and time-sensitive computer-based traces that capture the context, actions, processes and products. New educational measurement and analysis considerations are needed to address the challenges of finding patterns and making inferences concerning what someone knows and can do. Methods based in data-mining, machine learning, model-building and complexity theory are discussed as theoretical foundations for dealing with time sensitivity, spatial relationships, multiple layers of aggregations at different scales, and the dynamics of complex performance spaces. Examples of these considerations in game-based learning analytics are presented and discussed in the hope of making a contribution to the development and refinement of theoretical approaches to measurement.