A Dataset of University Students' Stress and Anxiety Levels based on Questionnaires and Wearable Sensors

Mental health issues such as stress and anxiety are highly prevalent among university students, often affecting their academicperformance and overall well-being. Understanding these conditions through objective, real-world data is essential fordeveloping effective monitoring and intervention strategies. We present a multimodal dataset that captures students’ dailystress and anxiety levels through self-reports and wearable sensor data. The dataset was collected during one academicsemester (February–July 2025) from undergraduate volunteers at two Mexican universities. Participants provided daily ratings ofstress and anxiety using a mobile application, while Fitbit Inspire 3 devices continuously recorded physiological and behavioraldata including heart rate variability, sleep quality, oxygen saturation, stress score, physical activity, and step count. The datasetfeatures over 80% questionnaire compliance and validated Fitbit measurements. This dataset addresses the scarcity of public,ecologically valid datasets on student mental health and enables reproducible research and analyses in affective computing,wearable sensing, and machine learning for stress and anxiety monitoring. This work and all authors received support from SECTEI, IPN, and ITESM through the SMIEAE project (4618c24).