One of the oldest findings in psychological research(Thorndike, 1931/1968) is that feedback facilitates learning. Without feedback about conceptual errors or an inefficient backstroke, the learner is likely to persist in making the same mistakes. In an extensive meta-analysis of 131 controlled studies, Kluger and DeNisi(1996) reported an average effect size or gain due to feedback of .40. They also acknowledge significant study variation with roughly one-third od studies showing negative effects. In attempting to identify characteristics of feedback most associated with positive effects, Kluger and DeNisi found that learning is more likely to be fostered when feedback focuses on features to the task and emphasizes learning goals. This important finding from the feedback literature is consistent with our previous argument for rubrics that allow performance to be judged in relation to well-defined criteria (rather than in comparison to other students), and it jibes with findings from the motivation al literature discussed later in the context of grading practices.