Advance Letters as a Means of Improving Respondent Cooperation in Random Digit Dial Studies: A Multistate Experiment

Publisher Website
Google Scholar
Abstract
Advance letters can potentially reduce the degree of nonresponse in random digit dial (RDD) surveys; however, they can also have a heterogeneous impact on subgroups, disproportionately raising participation rates among certain segments of the population and thereby having a detrimental effect on nonresponse bias. This is, in part, because advance letters can only be used in RDD surveys with the subset of respondents for whom an address can be identified. It may also be related to who in a household sees the letter. We assess whether the use of advance letters can improve the level of participation in the Behavioral Risk Factor Surveillance System (BRFSS) without introducing other potential data biases. The data reported here corroborate previous findings, in terms of the positive impact that advance letters can have on overall response rates (approximately a 6-percentage-point gain). Moreover, the advance letters were cost-efficient in that the cost of obtaining a fixed number of completed surveys using advance letters was lower than the cost without letters. However, the positive impact of the advance letter on reducing nonresponse may have been offset to some extent in that the advance letter may have biased the sample of those who completed the interview toward older, white respondents and those of higher socioeconomic status, and away from younger, nonwhite individuals and persons with lower education and income levels. This latter group was underrepresented in the final sample and also less like to remember having received a letter.

This publication has 0 references indexed in Scilit: