Roberta L. Sangster (2003) "Do Current Methods Used to Improve Response to Telephone Surveys Reduce Nonresponse Bias?"
Telephone surveys have taken the place of many field surveys over the past 25 years. As this transition occurred, the greatest concern was coverage bias that might occur with the shift to telephone interviews. Coverage is now less of a concern for most telephone surveys and new concerns about nonresponse have arisen to take center stage. There is general consensus that response rates are declining in RDD (list assisted, random digit dialed surveys) telephone surveys.Many researchers control the decline in response by increasing effort and thus the cost of data collection. The typical types of compensation include increasing the number of call attempts, lengthening the interview period, and making refusal conversion attempts. These methods tend to be costly, time consuming, and increase respondent burden and frustration. Other strategies to improve response targeting calls to the best time and days for making contact and using advance letters and monetary incentives into the survey process. This article reviews the effectiveness of the current methods used to reduce nonresponse and evaluates whether the method helps to reduce nonresponse bias. The review is organized under two general topics, sample management and methods used to encourage. Based on the review it proposes the development of a model that optimizes calling strategies with the express intent to reduce the potential for nonresponse bias rather than continue to focus on methods to just reduce nonresponse.It concludes with a discussion of future research.
Last Modified Date: July 19, 2008