Transportation planning relies heavily on the collection of data on travel behaviour by households, traditionally collected by recall surveys administered by household interviews. Although almost every major transportation study has relied on this method of data collection, there has been no systematic study of the quality of either the methodology or the resulting data. Yet these determine the validity of the data and, hence the conclusions, policies, and models developed from them. This paper deals with the design of a survey instrument; with a systematic procedure aimed at improving both the quality of the data and methodology, and also in reducing non-responses. The instrument is a diary, designed for an individual to record all travel and related activities for a 24-hour period. The paper describes, first, the process of design of the instrument through a series of tests and developmental surveys in West Germany. Second, the instrument was adapted for use in the United States of America, and the adaptations and improvements made are described. The methods used to field test the survey instrument are described, together with the resulting response rates and characteristics. The most significant conclusion of the paper is that the resulting instruments are applicable not only in the nation and culture in which they were designed originally, but also in other nations, languages, and cultures, adding a major dimension to the assessment of validity of this survey instrument.(a) for the covering abstract of the conference see IRRD 286978.