Using an investigaton of the dose-reponse relationship for road traffic noise as an exanrple, this paper examines some of the methodological issues involved in studies which attempt to relate human response to some aspects, of the environment. It is intended to highlight some of the difficulties involved and to provide background to policy makers and practitioners who wish to apply the results of such studies. It is suggested that studies of an environmental factor in isolation may give misleading results unless that factor is shown to be of immediate concern to the respondents. The paper examines the difficulty in defining and measuring human response and suggests that frequency of official complaints is likely to be a very poor measure of community response. It shows that psychological and physical factors, independent of the environmental factorof interest, have an influence on the response to that environmental factor; and that this influence may be greater than that of the environmental 'dose'. This means that one is not able to predict an individual 's response to a given dose, andinstead must rely on predicting the pooled response of groups of people. It is suggested that the range and distribution of the initial data should be examined critically before attempting to apply any results from a study of a dose-response relationship. Finally, the concepts of habituation, suppression and mobility of affected people are introduced as areas requiring further investigation.