Students often perform symbolic mathematical procedures without understanding the concepts that underlie them. One way to support students in making sense of these procedures is to provide a second representation that triggers conceptual thinking, in a way that symbols generally do not. This work investigates grounded feedback, in which students’ inputs are in a representation they are intended to learn (often symbolic), while dynamic feedback is shown in a second, easier-to-reason-with representation (often more concrete). Grounded feedback is based on similar feedback designs with experimental support in adults (Mathan & Koedinger, 2005; Nathan, 1998). The proposed work will extend the prior work by investigating grounded feedback in children and by examining different features of grounded feedback to determine if the entire set of features is truly necessary.
Kenneth R. Koedinger (Chair)
Daniel L. Schwartz (Stanford Graduate School of Education)
Robert S. Siegler (Psychology)
Catherine Copetas, copetas [atsymbol] cs.cmu.edu