24-h smartphone usage patterns in university students using high-granularity tracking.

Excessive smartphone use among university students can significantly impact academic performance, social interactions, and overall wellbeing, yet most studies rely on self-reported screen time, which lacks accuracy and granularity. A micro-longitudinal study was conducted among 100 undergraduates in Singapore using objective, high-granularity smartphone tracking of smartphone screen time (SST), frequency, and app usage. Using zero-inflated gaussian generalized linear mixed models (ZIG GLMMs) for SST and zero-inflated negative binomial generalized linear mixed models (ZINB GLMMs) for use frequency, 24-h temporal patterns were characterized based on a 7-day analysis, with differences examined by sex and day type (weekday/weekend). Average SST was 6.1 ± 2.8 h/day, with sustained use across waking hours and a peak at night. Sex differences in screen time and use frequency varied across a day. Also, weekdays exhibited a higher smartphone use frequency, likely linked to academic or social demands, while weekends featured prolonged leisure use. Self-reported and objective SST showed weak agreement (Cohen's Kappa ~ 0). This study demonstrates the feasibility of scalable and objective smartphone tracking to capture granular digital behaviors with minimal participant burden. Findings offer valuable insights into temporal patterns of smartphone use and provide a methodological foundation for future digital health research and interventions development.