University of Leicester
Browse
2016HeintzSAPhD.pdf (7.26 MB)

Digital Educational Games: Methodologies for Evaluating the Impact of Game Type

Download (7.26 MB)
thesis
posted on 2016-05-19, 15:22 authored by Stephanie Alexandra Heintz
The main research question addressed in this thesis is how the choice of game type influences the success of digital educational games (DEG), where success is defined as significant knowledge gain in combination with positive player experience. Games differ in type if they differ at least by one game feature. As a first step we identified a comprehensive set of unique game features, summarised in the Game Elements-Attributes Model (GEAM), where elements are the defining components that all games share (e.g. Challenges) and attributes are their possible implementation (e.g. time pressure). To deepen the understanding of relationships amongst game features, we conducted a survey based on the GEAM and received 321 responses. Using hierarchical clustering, we grouped 67 games, selected by the survey respondents, in terms of similarity and mapped the identified clusters on a 2D space to visualise their difference in distance from each other. On the resulting Game Genre Map, five main areas were detected, which proved to conform mostly to a selection of existing game genres. By specifying their GEAM attributes, we redefined these genres: Mini-game, Action, Adventure, Resource, and Role-play. Based on the aforementioned groundwork, two empirical studies were conducted. Study 1 compared three DEGs of the Mini-game genre, differing in a single GEAM attribute - time pressure vs. puzzle solving and abstract vs. realistic graphics. Study 2 compared DEGs of different genres which vary in the implementation of several GEAM attributes. For both studies, statistically significant differences were found in learning outcome, for Study 2 also in the player experience dimensions: Flow, Tension, Challenge, and Negative Affect. However, the influences of the covariates - learning and play preconditions, learning style, and personality traits - were not confirmed. Further research based on the methodological frameworks developed is needed.

History

Supervisor(s)

Law, Effie; Hoffmann, Michael

Date of award

2016-01-11

Author affiliation

Department of Computer Science

Awarding institution

University of Leicester

Qualification level

  • Doctoral

Qualification name

  • PhD

Language

en

Usage metrics

    University of Leicester Theses

    Categories

    No categories selected

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC