Browsing by Author "Williamson DB"

Now showing 1 - 2 of 2

Influencing transport behaviour: A Bayesian modelling approach for segmentation of social surveys
(Elsevier B.V., 2018-06-07) Dawkins L; Williamson DB; Barr SW; Lampkin SR
Current approaches for understanding and influencing transport behaviour often involve creating fixed, homogenous groups of similar surveyed individuals in order to explore specific behavioural profiles, an approach known as segmentation. Most commonly, segmentation is not based on a formal statistical model, but either a simple ‘a priori’ defined group classification narrative, failing to capture the complexity of varying group characteristics, or a ‘post hoc’ heuristic cluster analysis, applied to multidimensional behavioural variables, creating complex descriptive group narratives. Here, we present an alternative, Bayesian finite mixture-modelling approach. A clear group narrative is created by constraining the Bayesian prior to group survey respondents based on the predominance of a single apposing transport behaviour, while a detailed insight into the behavioural complexity of each group is achieved using regression on multiple additional survey questions. Rather than assuming within group homogeneity, this creates a dynamic group structure, representing individual level probabilities of group membership and within group apposing travel behaviours. This approach also allows for numerical and graphical representation of the characteristics of these dynamic, clearly defined groups, providing detailed quantitative insight that would be unachievable using existing segmentation approaches. We present an application of this methodology to a large online commuting behaviour survey undertaken in the city of Exeter, UK. Survey respondents are grouped based on which transport mode type they predominantly commute by, and the key drivers of these predominant behaviours are modelled to inform the design of behavioural interventions to reduce commuter congestion in Exeter. Our approach allows us to prioritise the most effective intervention themes, and quantify their potential effect on motor vehicle usage. For example, we identify that individuals that predominantly commute by public transport, but also sometimes motor vehicle, do so on average up to one day per week less often, if they are strongly concerned about the environment, demonstrating how an intervention to promote environmental awareness could greatly reduce motor vehicle usage within this group.
‘What drives commuter behaviour?': a Bayesian clustering approach for understanding opposing behaviours in social surveys
(Oxford University Press, 2019-08-23) Dawkins L; Williamson DB; Barr S; Lampkin SR
The city of Exeter, UK, is experiencing unprecedented growth, putting pressure on traffic infrastructure. As well as traffic network management, understanding and influencing commuter behaviour is important for reducing congestion. Information about current commuter behaviour has been gathered through a large on‐line survey, and similar individuals have been grouped to explore distinct behaviour profiles to inform intervention design to reduce commuter congestion. Statistical analysis within societal applications benefit from incorporating available social scientist expert knowledge. Current clustering approaches for the analysis of social surveys assume that the number of groups and the within‐group narratives are unknown a priori. Here, however, informed by valuable expert knowledge, we develop a novel Bayesian approach for creating a clear opposing transport mode group narrative within survey respondents, simplifying communication with project partners and the general public. Our methodology establishes groups characterizing opposing behaviours based on a key multinomial survey question by constraining parts of our prior judgement within a Bayesian finite mixture model. Drivers of group membership and within‐group behavioural differences are modelled hierarchically by using further information from the survey. In applying the methodology we demonstrate how it can be used to understand the key drivers of opposing behaviours in any wider application.