zbMATH — the first resource for mathematics

How randomized response techniques need not be confined to simple random sampling but liberally applicable to general sampling schemes. (English) Zbl 1365.62027
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, 17-27 (2016).
Summary: S. L. Warner [J. Am. Stat. Assoc. 60, No. 309, 63–69 (1965; Zbl 1298.62024)] the pioneer in research on randomized responses (RRs) to gather and analyze sensitive and stigmatizing data, started specifically with samples selected exclusively by the simplest scheme of simple random sampling (SRS) with replacement (SRSWR) and most of his successors followed it. But subsequent study remarkably in 2001 and onward has decisively shown that this restriction is uncalled for. Noteworthy aspects of RRs are unbiased and sometimes approximately unbiased estimation of the proportions of people bearing undesirable features in human communities and also totals/means of real values of sensitive variables of social interest. How to protect privacy in respondents’ disclosures and also how to adjust data analysis with options allowed to respond directly if no sensitivity is perceived are important points of note covering qualitative as well as quantitative data. Indirect but nonrandomized data gathering is also attended to. Bayesian analysis is also permitted. In caring for such diverse aspects sampling procedures are freely permitted. How it is achieved is a story briefly unfolded here.
For the entire collection see [Zbl 1349.62001].
62D05 Sampling theory, sample surveys
Full Text: DOI