Possible factors affecting response to postal questionnaires: findings from a study of general practitioner services.

Poor response rates affect the validity of results from postal questionnaires. The effect of three controllable factors upon response rates was examined in a recent study of patients' experiences and views of general practitioner services. The factors studied were the institution from which questionnaires were dispatched, the length of the questionnaire and the inclusion of a potentially sensitive question about ethnic origin. Questionnaires sent out by local Family Practitioner Committees were more likely to be returned than those sent out by a London-based independent research unit, but there were no differences in the nature of replies obtained. Neither the length of the questionnaire nor the inclusion of the potentially sensitive question affected response rates. The results of the study indicate that Family Practitioner Committees wishing to carry out postal surveys of users of general practitioner services can obtain satisfactory response rates.