Winkler and Franklin (1979) describe a Bayesian approach in the context of Warner’s (1965) RR model. For a dichotomous population with π _A as the proportion of persons bearing a sensitive character A, let, following Warner (1965), an RR be employed with P as the probability that a subject answers “yes” or “no” about bearing A and with probability (1 —P) about its complement. Let λ be the probability of a “yes” response. For an SRSWR of size n the likelihood is 6.1 L = L ( λ ) = ( n n 1 ) λ n 1 ( 1 − λ ) n − n 1 https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9780203741290/1f1f18f7-589b-4255-95c2-3f15b9d73704/content/math6_1_B.tif"/> where n ₁ is the frequency of “yes” replies and λ = P π A + ( 1 − P ) ( 1 − π A ) https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9780203741290/1f1f18f7-589b-4255-95c2-3f15b9d73704/content/math6_99_1_B.tif"/>