A Client-Server Architecture for Collection of Game-based Learning Data

Advances in information technology are driving massive improvement to the education industry. The ubiquity of mobile devices has triggered a shift in the delivery of educational content. More lessons in a wide range of subjects are being disseminated by allowing students to access digital materials through mobile devices. One of the key materials is digital-based educational games. These games merge education with digital games to maximize engagement while somewhat obfuscating the learning process. The effectiveness is generally measured by assessments, either after or during gameplay, in the form of quizzes, data dumps, and/or manual analyses. Valuable gameplay information lost during the student's play sessions. This gameplay data provides educators and researchers with specific gameplay actions students perform in order to arrive at a solution, not just the correctness of the solution.

This problem illustrates a need for a tool, enabling educators and players to quickly analyze gameplay data. in conjunction with correctness in an unobtrusive manner while the student is playing the game. This thesis describes a client-server software architecture that enables the collection of game-based data during gameplay. We created a collection of web services that enables games to transmit game-data for analysis. Additionally, the web application provides players with a portal to login and view various visualization of the captured data. Lastly, we created a game called "Taffy Town", a mathematics-based game that requires the player to manipulate taffy pieces in order to solve various fractions. Taffy Town transmits students' taffy transformations along with correctness to the web application. Students are able to view several dynamically created visualizations from the data sent by Taffy Town. Researchers are able to log in to the web application and see the same visualizations, however, aggregated across all Taffy Town players. This end-to-end mapping of problems, actions, and results will enable researchers, pedagogists, and teachers to improve the effectiveness of educational games.