Estimating Nonresponse Bias with Temporary Refusals

The accuracy of survey results are always endangered by the presence of nonresponse bias. Although we have a precise measure of the extent of nonresponse, we seldom have any information on how the nonrespondents differ from the respondents. In this article we attempt to estimate the characteristics of nonrespondents and thereby estimate nonresponse bias by comparing cooperative respondents with converted respondents who are known as temporary refusals.