An exploration of the role of interactivity in video game-based learning

Bruce Davis, Purdue University


Video games have become a popular pastime in which many people of many ages, ethnicities, and gender participate. Serious games have demonstrated effectiveness in terms of assessment in improving the learning of students. Although serious games can be effective as a teaching tool, they have made little impact in breaking through as a mainstream technique for use with students. An important consideration in game development is the level of interactivity present in a game. It is presumed that software that is more interactive is more effective. The purpose of this research was to find out what effect the degree of interactions had in a commercial game that has been modified to teach Boolean logic problem solving. The commercial game, Half-Life 2©, was modified such that game entities were created that simulated Boolean logic problem solving concepts. Two games were developed: Game A, one with fully interactive puzzles, and Game B, a game identical to Game B but with no interactive puzzles—the player was presented the puzzle in its solved state. A pre-test and post test relating to Boolean logic was also given. Our primary hypothesis stated that students who had interaction with the puzzles would affect posttest scores more than no interactivity with the puzzles. The subjects consisted of 36 college students of Purdue Calumet enrolled in the English 105 course. Mean test scores between Game A and Game B did not demonstrate a statistically significant difference. It must noted however that because of the study's small sample size, true statistical significance could not be ascertained. Though this study did not show a statistically significant difference, an observed difference was detected between the posttest means of groups A and B in which the group A, the high interactive group, did perform better on the posttest than did group B, the low interactive group. Though small sample sizes were certainly an issue, qualitative feedback provided by the subjects suggests, combined with the observation of a (non-statistical) difference between groups, that differences likely exist but our measurement devices were too course to detect them.^




David M. Whittinghill, Purdue University.

Subject Area

Design and Decorative Arts|Multimedia Communications|Education, Technology of|Computer Science

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server