On the basis of the results from the pilot study, a questionnaire comprising three sections was devised. The first section covered travel and demographic details; the second was a semantic differential scale (see Figs 6 & 7), the third a checklist of 18 improvements which could be made to railway carriages. The questionnaires were distributed to a random sample of 1500 passengers on 30 different trains, controlling for the time of day, day of the week, route travelled, the distribution of passengers in the carriage, and carriage type.