An Adjustment for Nonresponse in Sample Surveys

One of the more serious problems encountered in conducting sample surveys is that of nonresponse. Over the years, a variety of techniques for dealing with the problem have been proposed, but none have been universally satisfactory. The present study investigates the use of Bayes's theorem in adjusting for nonresponse bias. To test the use of Bayes's theorem to adjust for nonresponse, 600 hospitals were used in a simulated sample survey. On the basis of known information on five variables Bayes's formula correctly predicted the status of 92 of the 100 "non-respondents" relative to a sixth variable.