Professionals as responders: variations in and effects of response rates to questionnaires, 1961-77.

Abstract

Over the past 16 years the Institute for Social Studies in Medical Care (London) approached 19 samples of British professional groups and asked them to participate in surveys (interviews or posted questionnaires). This paper considers the factors influencing response rates and in what ways those responding may be unrespresentative of the sample as a whole. Factors that appeared to influence response rates are type of professional (nurses respond more often than doctors -- never below 90% and general practioners respond more often than consultants); length of questionnaire (a comparative study resulted in a shorter questionnaire out-drawing a longer one by 17 percentage points); and subject matter of the survey. Later surveys got lower response rates possibly a reflection of subject matter or the result of increasing apathy or antagonism toward health service research associated with the government. General conclusions about the representativeness of the respondents when compared with basic data on those who did not respond are that younger better qualified doctors were more likely to respond and among general practioners those who worked in groups were more likely to respond. Men and women doctors responded at about the same rates. Variations occurred on some surveys indicating that some subject matter may be of greater interest to certain groups. Studies that linked data from patients and professionals (comparing attitudes of patients toward their general practioners to discover differences among those general practioners who responded to specific surveys and those who did not) were also made but no general conclusions were drawn.