Individual differences in the effect of feedback on children's change in analogical reasoning

Various forms of feedback are used in formative assessment and interactive learning environments. The effects of different types of feedback are often examined at a group level. However, effective feedback may differ in learners with different characteristics or between learners at different stages in the learning process. In this paper explanatory item response theory (IRT) models are used to examine individual differences in feedback effects in children’s performance on a computerized pretest-training-posttest assessment of analogical reasoning. The role of working memory and strategy-use as well as interactions between these factors were examined in a sample of 1000 children who received either stepwise elaborated feedback, repeated simple feedback or no feedback during the training sessions. The results show that working memory efficiency significantly predicted initial ability and confirm that elaborate feedback is the most effective form of training in this particular interactive learning environment. Furthermore, children with initially less advanced strategy-use benefitted far more from each type of feedback than the children displaying more advanced strategies and this was unrelated to working memory efficiency. In children with advanced strategy-use working memory appears to moderate the effect of training. Explanatory IRT analyses appear useful in disentangling the effects of learner characteristics on performance and change during formative assessment and could possibly be used in optimizing feedback in computerized training and assessment environments.