Looking for Log-Based Traces of Mathematical Reasoning

Abstract

We explore the potential of log-based analysis to capture students’ mathematical reasoning. Twelve secondary school students worked with a GeoGebra applet on identifying reflective symmetry, while their justifications and actions recorded. Ten types of reasoning were identified through qualitative analysis; two of which were operationalized via log data, however with only low-moderate alignment with students’ explanations. Reasoning-independent logged data was not predictive of any of the reasoning types. Findings highlight the need for context-dependent, pedagogically informed log variables for assessing and supporting mathematical reasoning.