Document Type


Publication Date



Prior work has shown that computer supported homework can lead to more learning over traditional paper-and-pencil homework. This paper examines various factors such as, immediate feedback and tutoring, for causing learning. The primary study about learning from homework involved the comparison between instant-feedback with tutoring versus a control condition where students got feedback the next school day in math class. Eight grade students were assigned homework from these two conditions. The computer supported web-based homework provided instant feedback in terms of step-by-step scaffolding to solve a problem. After analyzing the students who participated in both conditions, it was found that students learned significantly more (effect size of 0.40) with computer supported homework. The difference was reliable. This result has enormous practical significance as it suggests an effective improvement over the widely used paper-and-pencil homework assignment method. We followed this main result with a second set of studies that tried to pull apart and better understand this result: is it more because of the better timeliness of computer homework, or is to more about getting high quality explanation?