Gaze collaboration patterns of successful and unsuccessful programming pairs using cross-recurrence quantification analysis

Abstract

A dual eye tracking experiment was performed on pairs of novice programmers as they traced and debugged fragments of code. These programming pairs were categorized into successful and unsuccessful pairs based on their debugging scores. Cross-recurrence quantification analysis (CRQA), an analysis using cross-recurrence plots (CRP), was used to determine whether there are significant differences in the gaze collaboration patterns between these pair categories. Results showed that successful and unsuccessful pairs can be characterized distinctively based on their CRPs and CRQA metrics. This study also attempted to interpret the CRQA metrics in relation to how the pairs collaborated in order to provide a somewhat clear picture of their relevance and meaning. The analysis results could serve as a precursor in helping us understand what makes a programming pair more successful over other pairs and what behaviors exhibited by unsuccessful pairs that should be avoided.