Abstract:
|
Unit nonresponse in face-to-face household surveys using probability samples can affect the quality of survey estimates. Although researchers can use a variety of tools for reducing unit nonresponse rates, most efforts in this area have focused only on the design features to improve survey participation such as number of calls, prenotification, incentives, or mode switch. However, those tools can be ineffective when controlled access situations or facilities (e.g., high-rise apartment buildings with locked central entrances or gatekeepers) are encountered. In these situations a different strategy or effort will be required to bring the sampled person into the respondent. In this paper, we present how using a special strategy based on an administrative system or process can increase response rates in a rare population survey on the use of illicit drugs in South Korea, one of the societies becoming increasingly security- and privacy-minded. In addition, we show the results on survey quality improvement in using the strategy.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.