Abstract
BACKGROUND: According to the World Health Organization, 27% of adults do not meet the recommended daily levels of physical activity (PA), making accurate PA measurement essential for informing evidence-based policies. This study explores ethical engagement with citizens through their ubiquitous digital tools (i.e., smartphones) to examine variations between retrospectively and prospectively reported PA behaviors within the same cohort. METHODS: This study is part of the Smart Platform, a digital citizen science initiative for ethical monitoring and real-time intervention. Data were collected from 118 adults who participated over eight consecutive days, including both weekdays and weekends. Prospective PA was assessed using time-triggered ecological assessments, while retrospective PA was measured using a modified, time-triggered, smartphone-based validated tool. Paired sample t-tests were used to compare retrospective and prospective PA. Linear regression models examined associations between socio-demographic and contextual factors and both types of PA reporting. Analyses were conducted for the overall sample and by gender (male vs. female). RESULTS: Participants consistently reported higher PA through retrospective measures compared to prospective ones (p < 0.001). In the overall sample, one significant association was found in the retrospective model, while three were identified in the prospective model. Among males, those who engaged in PA for fun or to maintain physical health reported higher retrospective PA, though this was not significant in the prospective model. In contrast, female participants who engaged in PA for fun reported higher PA in both retrospective and prospective models. CONCLUSIONS: Although exploratory, early findings suggest that repeated, prospective assessments via ubiquitous digital devices may enhance the validity and reliability of PA measurement. As citizen-owned digital tools become increasingly widespread, ethically leveraging big data through digital citizen science offers a promising approach to improve PA monitoring and support public health efforts.