A study of drive-alone/car-pool choice has been carried out using a disaggregate binomial logit model and data from a travel-to-work survey of staff at the atomic energy research establishment at Harwell. The work represented the first stage of a more general investigation of travel-to-work behaviour. Modelling format followed that of earlier drive-alone/car-pool choice studies, being based on the analysis of a data sub-set relating to individuals who either drove or car pooled. The work clearly showed the way in which various factors affect propensity to pool including distance to work, distance from potential pooling partners (there were indications that these had to be of a similar level in the job-grade hierarchy, and not necessarily nearest neighbours), sex and demands for cars from other household members. While the model that was developed was able to predict the behaviour of markedly different groups of Harwell workers well, there were some discrepancies from the findings of the earlier drive-alone/car-pool studies: possible reasons are discussed. More fundamentally, however, limitations of the modelling structure, which are likely to lead to overestimation of the effect of increased petrol price on the level of car pooling, are indicated. (Author/publisher)
Samenvatting