zbMATH — the first resource for mathematics

The classical randomized response techniques: reading Warner (1965) and Greenberg et al. (1969) 50 years later. (English) Zbl 1365.62028
Chaudhuri, Arijit (ed.) et al., Data gathering, analysis and protection of privacy through randomized response techniques: qualitative and quantitative human traits. Amsterdam: Elsevier/North Holland (ISBN 978-0-444-63570-9/hbk; 978-0-444-63571-6/ebook). Handbook of Statistics 34, 29-41 (2016).
Summary: In surveys of human populations, very often, a major source of nonsampling error is the presence of nonresponse. For various reasons, people refuse to respond. In case the issue under investigation is sensitive or stigmatizing, nonresponse seriously jeopardizes the validity of conclusions. It is very likely that some of the people agreeing to participate provide untruthful or misleading answers in order to reveal as little information as possible and thus to protect their privacy. S. L. Warner’s [J. Am. Stat. Assoc. 60, No. 309, 63–69 (1965; Zbl 1298.62024)] innovative randomized response technique opened new ways in surveys of human populations and created a new area of research. In this review paper, we present the two classical randomized response models, Warner’s pioneering technique and the unrelated question model as presented in B. G. Greenberg et al. [“The unrelated question randomized response model: theoretical framework”, J. Am. Stat. Assoc. 64, 520–539 (1969; doi:10.1080/01621459.1969.10500991)] and discuss very briefly some technical aspects of the two techniques. However, our emphasis in this chapter is not on the mathematics of the two classical techniques. Reading the two papers again, 50 years later, we offer some thoughts and stress the importance of these two influential publications to survey methodology.
For the entire collection see [Zbl 1349.62001].
62D05 Sampling theory, sample surveys
94A50 Theory of questionnaires
Full Text: DOI