Developing a questionnaire: some methodological issues.

This paper discusses methodological issues that arose during the development of a questionnaire to be used in a survey of people who have non-insulin dependent diabetes and people who do not have the disease. The results of a pre-test and a pilot test undertaken to establish the validity, reliability and stability of the questionnaire are described. The preliminary study indicated that respondents' metabolic control was less than optimal and that they associated the seriousness of diabetes with modes of treatment rather than with control of blood glucose levels. It was concluded from the results that a full scale study was warranted and that a survey using the finalized questionnaire would yield reliable data which could be used to improve diabetes-related education and health promotion.