Replication Data for: Exercise Snacking and Sleep Onset Regularity in Chinese Adolescents: A Cluster Randomized Clinical Trial

Replication data for a 2-arm parallel-group cluster randomized clinical trial evaluating the effect of a school-based exercise snacking program (four daily 2-minute bouts of vigorous calisthenic activity) on objectively measured sleep onset regularity, sleep duration, mental health, and physical activity in Chinese adolescents. Trial registration: ClinicalTrials.gov NCT07338058. Setting: A senior high school in Guangzhou, China. Participants: 386 adolescents aged 14-19 years from 12 classes randomized 1:1 at the class level (6 intervention, 6 control). Design: 4-month active intervention with 2-month postintervention follow-up. Outcomes assessed at four time points (Baseline, Month 2, Month 4, Month 6) via continuous wrist-worn accelerometry (Xiaomi Smart Band 8) and self-report questionnaires (DASS-21, WHO-5). Dataset: 15,615 person-night observations across 386 participants. Variables include nightly accelerometer-derived sleep duration and sleep onset clock time, daily step count, and assessment-window-level psychological measures (DASS-21 total and subscales, WHO-5 well-being). Demographic and socioeconomic covariates are included at the participant level. Use this dataset to reproduce the published analyses or to conduct secondary analyses on adolescent sleep regularity, exercise interventions, or school-based cluster trials. Anonymization: All data have been de-identified prior to deposit. Direct identifiers (participant names, student IDs, class rosters, dates of birth, and any free-text fields containing identifying information) have been removed. Participants are referenced only by randomly assigned study IDs with no link key retained in the deposited files.