Activity spaces are a longstanding geographic concept used to describe the mobility patterns of an individual. The proliferation of disaggregate space-time data now enables a sophisticated network-orientated approach to measuring and visualising these spaces. The current study leverages individual Global Positioning System trajectories from 365 participants over a seven-day period to introduce ‘route repetition’, a novel network-based metric quantifying habitual path selection within street networks. Using piecewise structural equation modelling with spatial controls, we demonstrate two key findings: higher route repetition is linked to lower crime rates in neighbourhoods (β = −0.24, p < 0.01); higher route repetition is also associated with longer movement duration (i.e., people who repeat routes more tend to spend more time traveling for a trip) (β = 0.24, p < 0.01). The network-centric conceptualisation of activity spaces advances beyond traditional elliptical and kernel density approaches by capturing the topological constraints of urban infrastructure. Our route repetition measure offers methodological innovation allied with substantive insight into routine spatial behaviour, providing a nuanced framework for analysing neighbourhood mobility patterns and their complex relationships with urban environmental factors such as crime exposure, land use diversity, socio-demographic composition and street network configuration. • Individual GPS trajectories used to identify route repetition and activity spaces. • Mobility patterns are influenced by route repetition, stops, and duration. • Neighbourhood crime influences individual mobility patterns.
Lu et al. (Wed,) studied this question.