Mail Surveys: A Closer Look at Nonresponse Rates

The topics of response rates, corresponding nonresponse rates and the associated nonresponse bias are, or should be, of interest to all researchers using survey instruments (Struebbe, Kernan, and Grogan 1986). Given that ad hoc (i.e., one-shot) mail surveys, in comparison to other survey methods including mail panels, have traditionally suffered from much higher nonresponse rates (Visser et al. 1996), both practitioners and academicians conducting mail surveys of this type are generally very concerned with nonresponse rates and normally report these rates in their research output (Yammarino, Skinner, and Childers 1991). These researchers also generally state whether they were able to detect nonresponse bias. That is, they acknowledge that it is important to determine the error that results from a systematic difference between those who responded to a mail survey and those who did not respond, because such a systematic difference raises serious doubts concerning the accuracy of the results (Lambert and Harrington 1990).