This research methods brief presents data from random digit dial (RDD) surveys in nine countries. We show that response rates to such surveys are typically below 60 percent and can be as low as 7 percent. We also show that most of the sample is lost at two points in the survey: non-contact, where respondents do not pick up the phone, and early refusal, where respondents terminate the interview before the survey begins. Beyond that point, cooperation is relatively high, with breakoffs during the interview ranging from effectively 0 to 10 percent across the nine countries. This evidence suggests that the most promising ways to increase response rates are strategies that increase pick-up rates and improve the first impression respondents have of the interviewer. While increasing contact and consent rates should logically improve response rates, it is not a guarantee. Future research would be needed to confirm whether respondents who are newly induced to answer and consent to the survey would be just as cooperative as those who would have consented without further intervention.

This paper was originally posted on the SSRN website.

Andrew Dillon, Steven GlazermanMichael Rosenbaum
Publication type: 
Phone Survey Methods Resource
April 29, 2021