Games as Formative Assessment Environments: Examining the Impact of Explanations of Scoring and Incentives on Math Learning, Game Performance, and Help Seeking

To investigate whether games may serve as useful formative assessment environments, this study examined, experimentally, the effects of two aspects of formative assessment on math achievement, game play, and help-seeking behaviors: (a) making assessment criteria explicit through the explanation of scoring rules and (b) incentivizing the use of feedback. A short pre-algebra game that targeted the concepts of adding rational numbers and the meaning of a fraction was used. Data were collected from 164 students in 4th to 6th grades at five after-school programs.;Five treatment conditions were employed in the study. A control group played a different math game. Three variations of providing explanation of scoring rules were tested: (a) no scoring information; (b) feedback on when points were earned and lost; and (c) elaborated explanation of the scoring rules made available before the game was played (in the tutorials), as feedback during the game (math elaborated feedback provided after scored events), and in the general help menu. The fifth condition provided both the elaborated explanation of the scoring rules and also offered an incentive to access additional feedback by regaining a portion of points. An assessment that was administered as a pretest and posttest was used to measure math achievement. In-game performance data (e.g., number of correct moves made in the game) were used to measure game performance. The proportion of times feedback was accessed was used to measure help seeking. Background information was also collected to obtain information on relevant variables such as prior math grades on report cards, game experience, and self-efficacy. Finally, students were administered a survey to measure their motivation to play the game.;Findings indicate that although they accessed feedback less, students who were provided the combined treatment of the elaborated explanation of the scoring rules and offered the incentive to seek additional feedback had higher normalized gain scores (compared to students who received minimal scoring information). This treatment also resulted in better game performance. The combined treatment was more effective for students who reported low game experience, low self-efficacy, low math self-concept, and low preferences for cooperative learning, as well as male students, by increasing scores between pretest to posttest, yielding moderate to large effect sizes. Moreover, for students who reported receiving a "D" in math on their last report card and those with low self-efficacy, the combined treatment of explanation of scoring rules and incentive to seek additional feedback resulted in significantly higher posttest scores, with outcomes of these analyses also yielding large effect sizes. Both a significant interaction between treatment conditions and pretest scores and analyses of subsamples of students suggest that this combined treatment may promote deeper processing of the feedback for lower performing students, especially students with low academic intrinsic motivation. Overall, findings suggest that when designing games for learning, it is important to both make assessment criteria, (in this case, the scoring rules of a game) explicit to a student as well as to find ways to motivate students to use provided feedback through the use of incentives.