Factors Influencing the Subjective Learning Effectiveness of Serious Games

Abstract

Aim/Purpose: This work examines which factors influence user views on the learning effectiveness of serious games. For that matter, a model was developed and tested. Background: Although the impact of serious games on learning is their most widely ex-amined aspect, research is spread thin across a large number of studies having little in common in terms of their settings, samples, and learning sub-jects. Also, there is a lack of consensus regarding which factors have an im-pact on their effectiveness. The most significant problem seems to be the fact that most assessment tools examined just a few factors. Methodology: The initial model included eleven factors responsible for shaping the learning outcomes, belonging to four groups: (a) content, (b) technical features, (c) user state of mind, and (d) learning enabling features. All possible relationships between these factors and subjective learning effectiveness were examined. Data were collected using the Serious Games Evaluation Scale. The target group was 483 university students who played two serious games. The model was tested using covariance-based structural equation modeling. Contribution: The study offers the prototype of a rather complex model, accurately explaining the intricate relationships between the substantial number of factors that were measured and their impact on user views regarding the subjective learning effectiveness of serious games. Findings: The final model fit statistics were very good, and 58.4% of the variance in subjective learning effectiveness was explained. The factor with the most significant impact was enjoyment, followed by subjective narration quality and realism. Quite interestingly, motivation did not have any effect on subjective learning effectiveness, while subjective feedback quality was not included as a construct in the final model. Moreover, the subjective ease of use and audiovisual fidelity had a minimal impact on other factors. Finally, the model proved to be invariant across genders and across the serious games that were used. Recommendations for Practitioners: Serious game developers can use the model so as to decide on which factors to focus, depending on their needs. Educators and education policymakers can also benefit from the model’s use, together with scales evaluating the quality of educational software. By assessing technical and content features and by using the model as a blueprint, they can envisage how enjoyable and motivating a serious game might be, as well as how it is going to impact user views regarding its learning effectiveness. Recommendation for Researchers: Researchers can use the model in order to understand what shapes the learning experience of users when they play serious games. They can also use it for understanding the interactions between different the factors that come into play. Future Research: Several alternative models have to be tested so as to develop a much simpler one which, at the same time, will have the capacity to adequately explain what users think of serious games. Several different target groups and serious games have to be examined in order to establish that the model is indeed invariant across a wide range of serious games genres and users. Finally, an interesting idea is to examine the relationship between subjective and objective learning effectiveness.