Alternative Estimator in Dichotomous Randomized Response Technique

Abstract

Obtaining correct responses to sensitive questions in social and behavioural researches is an ancient problem in survey research with respondents misreporting on sensitive behaviours or giving false response to protect themselves. This paper develops an alternative unbiased estimator by modifying the dichotomous randomized response technique model to tackle this problem. The proposed estimator was compared numerically with conventional ones by considering different practicable and suitable design choices. Proposed model was also considered when sampling with unequal probabilities with or without replacement. It was observed that the proposed estimator performs efficiently than the conventional ones. As the proposed model captures progressively more people involved in the sensitive attribute, the model outperforms other models considered. Therefore, social and behavioural researchers can now obtain correct and valid responses from sensitive behavioural researches with ease in order to make informed and reliable decisions.

Keywords

Dichotomous randomized response technique / Efficiency / Sensitive behavioural attribute / Unequal probability

References

[1] Adebola FB, Johnson OO. An improved Warner’s randomized response model. Int. J. Stat. Appl.. 2015, 5 6 263-267

[2] Adebola FB, Johnson OO, Adegoke NA. A modified stratified randomized response technique. J. Math. Theory Model.. 2014, 4 13 29-42

[3] Adebola FB, Adediran AA, Ewemooje OS. Hybrid tripartite randomized response technique. Commun. Stat. Theory Methods. 2017, 46 23 11756-11763

[4] Chang HJ, Huang KC. Estimation of proportion and sensitivity of a qualitative character. Metrika. 2001, 53 269-280

[5] Chaudhuri A. Estimating sensitive proportions from unequal probability sample using randomized responses. Pak. J. Stat.. 2001, 17 3 259-270

[6] Chaudhuri A. Randomized Response and Indirect Questioning Techniques in Surveys. 2011 Boca Raton: Chapman & Hall/CRC

[7] Chaudhuri A, Christofides TC. Indirect Questioning in Sample Surveys. 2013 Berlin: Spinger

[8] Chaudhuri A, Bose M, Dihidar K. Estimating sensitive proportions by Warner’s randomized response technique using multiple randomized responses from distinct persons sampled. Stat. Pap.. 2011, 52 1 111-124

[9] Chaudhuri A, Stenger H. Survey Sampling: Theory and Methods. 2005 2 Boca Raton: Chapman & Hall/CRC

[10] Coutts E, Jann B. Sensitive questions in online surveys: experimental results for the randomized response technique (RRT) and the unmatched count technique (UCT). Sociol. Methods Res.. 2011, 40 1 169-193

[11] Ewemooje OS, Amahia GN. Improving the efficiency of randomized response technique for two sensitive characters. FUTA J. Res. Sci.. 2016, 12 1 65-72

[12] Ewemooje OS. Estimating two sensitive characters with equal probabilities of protection. Cogent Math.. 2017, 4 1-14

[13] Ewemooje OS, Amahia GN. Improved randomized response technique for two sensitive attributes. Afr. Stat.. 2015, 10 2 839-852

[14] Ewemooje OS, Amahia GN, Adebola FB. Estimating prevalence of induced abortion and multiple sexual partners using improved randomized response technique for two sensitive attributes. Commun. Stat. Case Stud. Data Anal. Appl.. 2018, 3 1–2 21-28

[15] Greenberg B, Abul-Ela A, Simmons W, Horvitz D. The unrelated question randomized response model: theoretical framework. J. Am. Stat. Assoc.. 1969, 64 529-539

[16] Huang KC. Survey technique for estimating the proportion and sensitivity in a dichotomous finite population. Stat. Neerl.. 2004, 58 1 75-82

[17] Hussain, Z., Shabbir, J.: Randomized use of Warner’s randomized response model. Interstat. http://interstat.statjournals.net/YEAR/2007/articles/0704007.pdf (2007)

[18] Kim J-M, Warde WD. A stratified Warner’s randomized response model. J. Stat. Plan. Inference. 2004, 120 1–2 155-165

[19] Krumpal I. Estimating the prevalence of xenophobia and anti-Semitism in Germany: a comparison of randomized response and direct questioning. Soc. Sci. Res.. 2012, 41 1387-1403

[20] Lee C-S, Sedory SA, Singh S. Estimating at least seven measures of qualitative variables from a single sample using randomized response technique. Stat. Prob. Lett.. 2013, 83 399-409

[21] Mangat NS. An improved randomized response strategy. J. R. Stat. Soc. Ser. B. 1994, 56 93-95

[22] Singh HP, Tarray TA. A sinuous stratified unrelated question randomized response model. Commun. Stat. Theory Methods. 2016, 45 11 3126-3137

[23] Tarray TA, Singh HP. New procedures of estimating proportion and sensitivity using randomized response in a dichotomous finite population. J. Modern Appl. Stat. Methods. 2016, 15 1 635-669

[24] Warner S. Randomized response: a survey technique for eliminating evasive answer bias. J. Am. Stat. Assoc.. 1965, 60 63-69