Modelling the impact of neighbourhood effects and psychosocial factors on commuting travel behaviour: an application in the metropolitan city of Cagliari

In recent years, studies have found that individuals’ choice of transportation mode is influenced not only by the level of service and socioeconomic variables but also by spatial and social interactions effects, also known as neighbourhood effect in which the individual’s behaviour is influenced by the relational effects of the people living in his neighbourhood. However, past research has only focused on topics such as children's travel behaviour, bicycle use, choice of residence place, and have not considered the presence of a neighbourhood effect when analysing commuting travel behaviour. In addition, the majority of previous works have investigated neighbourhood effects only for the place of residence, while it would be important to understand the presence of these effects also for the place of destination. Another issue of most of the papers who have analysed the effect of individuals’ spatial interactions, is that they neglected the impact of psycho-attitudinal variables which, in the last 20 years, have been proven to play a fundamental role in individuals’ travel behaviour. Given the above background, the object of the current thesis is the analysis of the spatial and social interactions and psychosocial factors that may affect individuals’ decision process when choosing a transport mode for their home - work and home - study trips. To this end, it is specified and estimated a spatial probit model, modelling individuals’ choice probability to use a sustainable means of transport. The utility function of the estimated model include: 1) the effect of the level of service variables and household characteristics; 2) the effect of psychosocial variables, whose values were computed through the computation of factor scores; 3) spatial interactions effects. The spatial effect is computed both for the place of origin (residence) and the place of destination (work or study address). The model is applied to a sample of students and workers (3251 individuals), who participated in 2019 to a survey conducted by the Interuniversity Centre for Economic Research and Mobility (CIREM) of the University of Cagliari (Sardinia) in the metropolitan city of Cagliari. Model results shows that spatial (social) effects are statistically relevant in explaining commuters’ travel behaviour, both when considering travel behaviour of individuals living close to the place of residence and working/studying close to the place of destination. Another important finding is that individual decision-making process in transport mode choice is influenced not only by socioeconomic characteristics and spatial interactions, but also by psychosocial effects. In particular, it is found that the intentions to adopt a sustainable travel behaviour of transport and the perceived behavioural control of using a sustainable means of transport positively influence the choice to use the active mobility and transit service. Finally, the results suggest that ignoring these spatial effects may lead to biased estimates of the model parameter values, and consequently, to the underestimation of the impacts of policies and strategies aimed at favouring sustainable mobility.