The degree of knowledge gain metric: Evaluating how searchers acquire new information while searching online

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

This study presents the degree of knowledge gain, a behavioral metric that estimates how much searchers learn during Web searches. Grounded in information theory and Shannon entropy, the metric quantifies reductions in uncertainty by analyzing query reformulations, clicks, and document positions, avoiding reliance on pre/post-testing or self-reports. The metric was validated with a cohort of legal professionals who completed three domain-specific search tasks on a national legal platform. Behavioral coding of think-aloud sessions and statistical tests (analysis of variance; Spearman rank correlation) demonstrated strong positive correlations between the degree of knowledge gain and conventional transfer of learning scores, with structured tasks producing the highest gains. This approach offers an automated, scalable indicator of learning, enriching the Searching as Learning paradigm and enabling real-time adaptation in personalized search, intelligent tutoring, and conversational systems.