This study is concerned with the problems of valuation of time and the estimation of models of modal choice. Following a survey of the value of time literature, a behavioural hypothesis is developed which forms the basis of the modal choice model to be estimated. This model is then set in the context of the theory of consumer demand. The data used in the estimation of this model is derived from the Edinburgh Glasgow Area Modal Split Study; the objectives, operations and results of this survey effort are described, and the development of the data base is outlined. It is a feature of this study that some novel collection methods were used, notably in the questionnaire. The estimation procedures which may be used to estimate binary choice models are compared statistically and logit analysis is selected for estimating the model. The content of the model was found to differ greatly from the models estimated for commuters, and the income stratification was discovered to modify the models considerably. A value of time was derived, but only from a sub-optimum model. This value is higher than the values derived from commuting studies. No value of time could be derived for the separate income groups. The major conclusion is that the use of models built for commuters in other situations is not a justifiable procedure.