Unique Methods: Door Knocking
Achieving high response rates is a critical component of data collection, especially when working with a longitudinal study such as Project EAT, which has spanned over a decade.
After exhausting more typical methods such as telephone, online, and mail, OMS took the data collection efforts to the streets in one final attempt to boost the response rates for Project EAT. OMS focused the door knocking efforts on cohort participants whose mail had never been returned to sender and who had not completed the online or paper version of the survey.
OMS sent research staff in groups of two with the goals of delivering survey packets and verifying participant address information. When participants were home, OMS provided them with a survey packet containing the paper survey and a link to an online version. Participants were given the option of mailing their completed survey to OMS or having OMS pick it up at a later time. These home visits were supplemented with reminder phone calls occurring a few days after the visit. When participants were not home, a survey packet was left with other individuals residing at the address or at the doorstep.
The door knocking efforts are ongoing, but they have already resulted in 38 new completed surveys. While not large in the grand scheme of things, each of these responses is invaluable to the project. Another finding from the door knocking method was that approximately 20% of participant addresses were invalid, despite the post office never returning any of our mail sent to those participants.
This data collection effort is built off of research indicating that multi-mode efforts increase response rates and sequential efforts in particular can be effective. (Improving Response To Web and Mixed-Mode Surveys, Morgan Millar and Don Dillman, Public Opinion Quarterly, Vol. 75 No. 2, Summer 2011, pp. 249-269, Response Rate and Measurement Differences in Mixed Mode Surveys: Using Mail, Telephone, Interactive Voice Response and the Internet, Don A. Dillman, Glenn Phelps, Robert Tortora, Karen Swift, Julie Kohrell, and Jodi Berck, Draft paper, financial support from the Gallup Organization and the Department of Rural Sociology and the Social and Economic Sciences Research Center at Washington State University).