A feedback model for data-rich learning experiences

Feedback has been identified as one of the factors with the largest potential for a positive impact in a learning experience. There is a significant body of knowledge studying feedback and providing guidelines for its implementation in learning environments. In parallel, the areas of learning analytics or educational data mining have emerged to explore how to analyse exhaustive digital trails produced by technology mediation to improve learning experiences. Current conceptualisations of feedback do not take into account the presence of these trails nor the presence of knowledge extracted through analytics techniques. This paper presents a model to reconceptualise feedback in data-rich learning experiences. It acknowledges the presence of algorithms to analyse and predict learner behaviour and proposes an integrated view between these elements as part of the feedback process and aspects conventionally present in previous models. This new conceptualisation offers instructors, designers and researchers a framework to formulate feedback processes in scenarios with comprehensive data capturing, while retaining a solid connection with well-established educational theories.