Response Rates in Random-Digit-Dialed Telephone Surveys: Estimation vs. Measurement.

The efficacy of the random digit dialing method in telephone surveys was examined. Random digit dialing (RDD) generates a pure random sample and provides the advantage of including unlisted phone numbers, as well as numbers which are too new to be listed. Its disadvantage is that it generates a major proportion of nonworking and business numbers--about four in five. Although a number of RDD sampling strategies have been proposed, they all lack a means of classifying those numbers at which there is never an answer. They could be outside the sampling range (businesses or nonworking numbers) or households whose members are rarely home. The computation of response rates is a major problem, however, especially since many surveyers want a minimum response of 50 percent. This study examined the validity of the approach which distributes never-answered numbers in the same proportion as those answered. Five surveys were conducted in Sacramento and Santa Cruz (California) during evening and weekend hours, and then a followup was made during business hours. Results indicated that nonworking telephone numbers were generally identified during the regular (evening and weekend) calling hours. Business numbers were difficult to identify, and were probably understated. This proposed method was not recommended. (GDC)