An analysis of the role of residential location on the relationships between time spent online and non-mandatory activity-travel time use over time

Although the associations between the use of information and communications technologies (ICT) and individuals' daily travel and activity patterns have been extensively investigated for several decades, few studies have examined the amount of time spent using ICT, its implications on activity-travel behavior, and how such ICT-travel relationships may vary over time and according to residential location. This study takes a quasi-longitudinal perspective to explore how the amount of time individuals spend on the Internet for personal or non-work purposes correlates to their activity-travel for non-mandatory maintenance and leisure purposes, as well as how such associations evolve over a decade. More importantly, it examines how the role of people's residential locations in determining the associations between time spent on the Internet and travel has changed over time. Our approach utilizes two datasets from two major cross-sectional surveys in Scotland: the 2005/06 Scottish Household Survey (SHS) and the 2015 Integrated Multimedia City Data (iMCD) Survey, which were similarly structured and developed. To accommodate the multiple discreteness characterizing activity-travel choice and duration, the multiple discrete-continuous extreme value (MDCEV) model was employed to capture the Internet–travel relationships for the full sample and the urban, town, and rural sub-samples in both 2005/06 and 2015. Our findings suggest that use of the Internet for personal purposes increasingly tends to discourage rather than facilitate physical activity and travel for non-mandatory purposes over time, especially for those who spend high levels of time on the Internet (over ten hours per week). However, such Internet–travel relationships are generally weaker among people living in remote areas than those living in urban areas. While the relationships regarding maintenance activity purposes are significant for almost all levels of Internet users among the urbanites in both years, they were not significantly found at all among rural residents in either 2005/06 or 2015.