Abstract: | To study the accuracy with which nonresponse bias in mail questionnaires can be extrapolated from the trend derived from the several coordinate values of initial and follow-up response groups for any given variable, data were obtained on 7 variables for 3 groups. Of the 7 biographical data variables considered, 5 were found to be consistent with the hypothesis that the response-nonresponse probability function indicates the presence and direction of non-response bias. (PsycINFO Database Record (c) 2010 APA, all rights reserved) |