Apriori-based Analysis of Learned Helplessness in Mathematics Tutoring: Behavioral Patterns by Level, Intervention, and Outcome

John Paul P. Miranda's study applies the Apriori algorithm—a classic association rule mining technique—to identify behavioral patterns of learned helplessness (LH) in mathematics tutoring systems. Analyzing interaction logs across three dimensions (LH level, system-based intervention, and problem-solving outcome), the research reveals that skipping problems without using hints is the single most frequent behavioral pattern associated with unsolved outcomes. In contrast, persistence behaviors like not skipping problems were less dominant overall but strongly tied to successful problem-solving.

When comparing by LH level, low-LH students exhibited stronger associations between problem-solving and not skipping, as well as positive links between hint usage and solved outcomes. High-LH students showed more avoidance patterns, with skipping strongly tied to unsolved outcomes. Notably, students without system intervention had the highest lift for persistence-success links, while the with-intervention group paradoxically showed stronger patterns of skipping leading to unsolved outcomes. Outcome-specific analysis confirmed that not skipping consistently predicted solved problems across all groups, while skipping without hints was a reliable predictor of failure.