Academic Resilience and Behavioral Patterns in a Digital Problem-Solving Task: Evidence from Process Data

Digital problem-solving assessments provide process data that can reveal how students behave beyond final achievement. Previous research has shown substantial individual differences in problem-solving behavior. Academic resilience commonly refers to strong academic performance despite socioeconomic disadvantage, but little is known about how academically resilient students behave during problem solving. This study examined whether socioeconomically disadvantaged students classified as academically resilient showed distinctive behavioral patterns in the Programme for International Student Assessment (PISA) 2012 TICKET task. Academic resilience was defined using socioeconomic disadvantage and mathematics achievement, and propensity-score matching (PSM) was used to improve comparability on selected background characteristics. Process data were analyzed as action sequences to identify recurring behavioral profiles. Results showed that resilient students tended to follow longer action sequences and had higher backtracking proportions, whereas differences in timing indicators were not statistically significant. These findings suggest that mathematics-based academic resilience is associated with differences in the organization of task behavior.