Time spent on social media and its impact on travel behaviours and activity participation: Preliminary results from a comparative study based on association analysis and discrete choice models

An empirical analysis is proposed to study the effects that the use of computers for leisure and relaxing, that includes the time spent on social media, has on travel patterns, with special emphasis on some discretionary trips (especially leisure trips). Data from 2013 American Time Use Survey is used in the study. Results obtained using a data mining technique, namely association analysis (AA), and those derived from a discrete-continuous model of leisure engagement (DCM) are proposed. Association analysis is implemented first to explore the intertwined relationships between socio-economic characteristics, social media use and travel behaviours, and then inform the specification of the econometric model. Then the estimated models are applied to predict changes in the decisions of activity participation and time usage according to changed socio-demographic profile and time usage on other related activities. The analysis reveals the socio-demographic profile of those engaged in various types of leisure activities and the effect of travel on the time spent on leisure. High educated individuals are more likely to be involved in leisure associated to computer use, while the presence of children in the household and a full time job have the opposite effect. Travel time to different activities have little effect on the total leisure time.