response rate


response rate

the proportion of individuals invited to participate in a study who actually do participate. In social surveys using SAMPLING the response rate is unlikely to be 100%; 90% or over would be considered very good, and over 70% is normally acceptable. The method of collecting data inevitably affects the response rate. Postal questionnaires generally have a low one, while personal INTERVIEWS achieve a higher rate. Reasons for NONRESPONSE include death, house removal, unwillingness to cooperate, not being available when the interviewer calls, etc. Because of this almost inevitable failure to achieve the selected sample, systematic error or BIAS is introduced and an assessment of the VALIDITY of extrapolating from the sample to the parent population has to be made. There are various statistical techniques used to strengthen the level of confidence that is put on sample data.