The Case of Less Rule-Based Problems: Inferring a Solution Strategy Versus Well-Justified Self-Explanations

• In Event: AERA Poster Session 9
  In Poster Session: Supporting Learning Through Fundamental Cognitive Processes

Abstract

A previously published study investigated how worked examples and problem-solving fosters application-oriented knowledge in the domain of statistics (Schwaighofer, Bühner, & Fischer, 2016). Notably, this study used less rule-based problems that have no algorithmic solution procedure. We developed a rubric to assess self-explanation quality in such a context and explored whether self-explanation quality mediates the worked example effect on learning gains reported in the previous publication. We operationalized self-explanation quality as the degree to which learners justified their problem solutions. To jointly investigate mediation and moderating factors we used conditional process analysis (Hayes, 2018). Our results suggest that in the context of less rule-based problems, it is more important to form a problem-solving schema than to justify solutions well.

Authors

• Sarah Bichler, University of California - Berkeley

• Frank Fischer, Ludwig-Maximilians-Universität München