Towards performance prediction using in-game measures

Sylvester Arnab, O. Imiruaye, F. Liarokapis, Gemma Tombs, Petros Lameras, A. Serrano-Laguna, P. Moreno-Ger

    Research output: Contribution to conferencePaperpeer-review


    The efficacy of a learning process is influenced by the quality of teaching, learning support and environment. This requires effort in tracking how students learn. This paper explores the use serious games in order to help understand the learning process, where interaction data during a play-learn session can be captured. The focus is on the use of in-game data, analyzed using Learning Analytics techniques, and discusses the potential of such an approach to predict learners’ performance. Gameplay data were collected from various play-learn sessions based on a First Aid Game. Results indicate that in-game measures can help to understand students’ progress and predict their performance, providing opportunities for individual support to be provided to learners.
    Original languageEnglish
    Publication statusPublished - 2015
    EventAmerican Educational Research Association annual meeting - Chicago, Illinois, Chicago, Illinois, United States
    Duration: 16 Apr 201520 Apr 2015


    ConferenceAmerican Educational Research Association annual meeting
    CountryUnited States
    CityChicago, Illinois


    • serious games
    • performance prediction
    • user studies
    • learning analytics
    • game-based learning

    Fingerprint Dive into the research topics of 'Towards performance prediction using in-game measures'. Together they form a unique fingerprint.

    Cite this