This study aims to quantitatively model rather than to presuppose whether or not air pollution in Beijing (China) affects people’s activities of daily living (ADLs) based on an Internet of Behaviours (IoB), in which IoT sensor data can signal environmental events that can change human behaviour on mass. Peoples’ density distribution computed by call detail records (CDRs) and air quality data are used to build a fixed effect model (FEM) to analyse the influence of air pollution on four types of ADLs. The following four effects are discovered: Air pollution negatively impacts people going sightseeing in the afternoon; has a positive impact on people staying-in, in the morning and the middle of the day. Air pollution lowers people’s desire to go to restaurants for lunch, but far less so in the evening. As air quality worsens, people tend to decrease their walking and cycling and tend to travel more by bus or subway. We also find a monotonically decreasing nonlinear relationship between air quality index and the average CDR-based distance for each person of two citizen groups that go walking or cycling. Our key and novel contributions are that we first define IoB as a ubiquitous concept. Based on this, we propose a methodology to better understand the link between bad air pollution events and citizens’ activities of daily life. We applied this methodology in the first comprehensive study that provides quantitative evidence of the actual effect, not the presumed effect, that air pollution can significantly affect a wide range of citizens’ activities of daily living.

To study and understand the importance of Internet of Things-driven citizen science (IoT-CS) combined with data satisficing, we set up and undertook a citizen science experiment for air quality (AQ) in four Pakistan cities using twenty-one volunteers. We used quantitative methods to analyse the AQ data. Three research questions (RQ) were posed as follows: Which factors affect CS IoT-CS AQ data quality (RQ1)? How can we make science more inclusive by dealing with the lack of scientists, training and high-quality equipment (RQ2)? Can a lack of calibrated data readings be overcome to yield otherwise useful results for IoT-CS AQ data analysis (RQ3)? To address RQ1, an analysis of related work revealed that multiple causal factors exist. Good practice guidelines were adopted to promote higher data quality in CS studies. Additionally, we also proposed a classification of CS instruments to help better understand the data quality challenges. To answer RQ2, user engagement workshops were undertaken as an effective method to make CS more inclusive and also to train users to operate IoT-CS AQ devices more understandably. To address RQ3, it was proposed that a more feasible objective is that citizens leverage data satisficing such that AQ measurements can detect relevant local variations. Additionally, we proposed several recommendations. Our top recommendations are that: a deep (citizen) science approach should be fostered to support a more inclusive, knowledgeable application of science en masse for the greater good; It may not be useful or feasible to cross-check measurements from cheaper versus more expensive calibrated instrument sensors in situ. Hence, data satisficing may be more feasible; additional cross-checks that go beyond checking if co-located low-cost and calibrated AQ measurements correlate under equivalent conditions should be leveraged.

Conventional wisdom posits that smog suppresses outdoor activity while shifting peoples’ activities indoors. Using anonymized Mobile Phone Data Provider Records fused with Point-of-Interest (POI) data sourced from the Gaode (Amap) open database for Beijing (2–22 February 2015), we test this substitution hypothesis at an hourly resolution across 12 POI-defined activity categories. We estimate the adjusted population density (APD) from mobile phone data via usage-bias calibration, interpolate city-wide AQI (Air Quality Index) and PM2.5 fields, and identify associations with a two-way fixed-effects design (Voronoi polygon (VP), day × hour model. We also handle time-invariant POI activities, while factoring in weather and day types. We find a dual suppression of both outdoor and indoor physical activities: worsening air quality is associated with lower participation in most outdoor and indoor activities. Effects are heterogeneous across categories and hours; shopping shows all-day negative marginal effects, whereas a few categories (e.g., sightseeing) display positive correlations in select afternoon hours consistent with congestion-avoidance rather than health-driven indoor substitution. Quantitatively, a 100-point AQI increase is associated with an order of 1–5 persons/km2 decline at peak hours for most activities. A Comprehensive Impact Index (CII) summarizes the spatial heterogeneity across the city. POI venue operators should anticipate city-wide activity reduction both indoors and outdoors under heavy pollution, rather than plan solely for outdoor-to-indoor activity shifts.