A methodology for assessing the effectiveness of serious games and for inferring player learning outcomes.

Although serious games are proven to serve as educational tools in many educational domains, there is a lack of reliable, automated and repeatable methodologies that measure their effectiveness: what do players know after playing serious games? Do they learn from them? Previous research shows that the vast majority of serious games are assessed by using questionnaires, which is in stark contrast to current trends in the video game industry. Commercial videogame developers have been learning from their players through Game Analytics for years via non-disruptive game tracking. In this paper, we propose a methodology for assessing serious game effectiveness based on non-disruptive in-game tracking. The methodology involves a design pattern that structures the delivery of educational goals through a game. This structure also allows one to infer learning outcomes for each individual player, which, when aggregated, determine the effectiveness of a serious game. We tested the methodology by having 320 students play a serious game. The proposed methodology allowed us to infer players' learning outcomes, to assess the game effectiveness levels and to identify issues in the game design.