An exploration of the characteristics of excess travel within commuting

Abstract

Travel behaviour research aims to inform and provide evidence for sound transport policy. Excess travel, where individuals demonstrate excessive use of for example time or distance, challenges assumptions underpinning fundamental beliefs of travel behaviour research where travel should be minimised in order to get to the destination. This thesis explores the phenomenon of excess travel and the characteristics of people exhibiting excess travel within a commuting context, using Tyne and Wear as a case study. Building on existing definitions of excess commuting, which include time and distance, this study gradually adds additional parameters of cost, effort, and many other parameters (e.g. value of time, weights for walking and waiting) in the generalised cost formula, and the final sample is analysed to identify similarities and differences between excess commuters (EC) and not excess commuters (NEC). The methodology uses a GIS technique for sampling and a questionnaire approach for data collection. The final sample includes origin-based (home) commuters who completed a questionnaire delivered to their home addresses, and destination-based (work) commuters who completed an online version of the same questionnaire. Analytical methods are used to identify EC and NEC based on self-reported (‘pure’) values of the four key parameters of time, cost, distance and effort while commuting and using a generalised cost approach. For the parameters of time and cost as well as for the generalised cost results seven saving options are considered, where 5% savings is the lowest option and 50% or more savings is the highest option. An analysis of various attributes and their differences in medians together with a series of socio-economic characteristics are used to distinguish between EC and NEC within the four groups in total (time, cost, effort, generalised cost). The results show that within the collected sample EC make up between 32% (in the cost group) and 78% (in the effort group) of the total sample (depending on the parameter/group considered), and that there are some statistically significant differences at the 95% level between EC and NEC within the groups. The fact that the number of EC varies between the groups is to be expected, as the literature review suggested that taking different parameters into account produces different results. Generally, EC seem to behave in a similar manner to the rest of the sample, in terms of most of the factors tested, when making choices about commuting, but for example 41% of the respondents ii drive to work and within this driving group there are more EC than NEC (for example 44% of EC versus 37% of NEC within the time group or 52% of EC versus 36% of NEC within the cost group). More importantly, the median values for the four key parameters of travel to work (actual commute time, ideal one-way commute time, commute cost, commute distance) are higher in majority of the cases for EC than for NEC within the four groups. Attitudes and preferences also play a role, demonstrating that the most frequent trip purpose, the commute, can provide some benefit to travellers. The results also show that in terms of the activities such as listening to music/radio, reading book/newspapers, exercising or concentrating on the road a majority of statistically significant differences between EC and NEC occur within the cost and the effort groups only. The demand for more direct routes and cheaper fares on public transport is emphasised by the majority of the sample. The respondents tend to be well informed about their travel to work alternative transport modes and different transport planning tools available, and the Internet stands out as a primary source of information employed by majority of both EC and NEC. In exploring the characteristics of EC and NEC in more depth, recommendations are identified for public transport providers to improve their services and encourage more commuters to transfer travel time into activity time.