Abstract
Collecting reliable responses to sensitive survey questions is challenging, since respondents may be more likely to refuse to respond or to provide biased responses. To address these challenges, Warner (J Am Stat Assoc 60:63–69, 1965) pioneered the randomized response (RR) technique to estimate proportions of individuals in a population with either of two possible attributes. The RR technique can overcome non-response and underreporting biases because it doesn’t reveal the respondent’s attribute, and a generalization of the random component of the response by Christofides (Metrika 57:195–200, 2003) improves estimation properties. In this study, we develop a new RR model to estimate proportions of individuals with each of multiple categories of an attribute using a single sensitive question by means of only one randomization device based on Christofides’s (Metrika 57:195–200, 2003) model. Under the proposed model, a respondent reports the absolute difference between an integer associated with his or her attribute and a random integer. In a part of this research, we conduct a simulation study of the relative efficiency of the proposed methods. The technique is illustrated using data from the 2012 Family and Gender Module of the Taiwan Social Change Survey to estimate the proportions of individuals of different sexual orientations, and the results are compared with the results of direct inquiry from the same survey.
Similar content being viewed by others
References
Abul-Ela ALA, Greenberg BG, Horvitz DG (1967) A multi-proportion randomized response model. J Am Stat Assoc 62:990–1008
Barabesi L, Franceschi S, Marcheselli M (2012) A randomized response procedure for multiple sensitive questions. Stat Pap 53:703–718
Bourke PD (1974) Multi-proportions randomized response using the unrelated question. Report No. 74 of the Errors in Survey Research Project, Institute of Statistics, University of Stockholm (Mimeo)
Bourke PD (1981) On the analysis of some multivariate randomized response designs for categorical data. J Stat Plan Inference 5:165–170
Bourke PD (1982) Randomized response multivariate designs for categorical data. Commun Stat Theory Methods 11(25):2889–2901
Bourke PD (1990) Estimating a distribution function for each category of a sensitive variable. Commun Stat Theory Methods 19(9):3233–3241
Chang HJ, Liang DH (1996) A two-stage unrelated randomized response procedure. Aust J Stat 38:43–51
Chen CC, Singh S (2009) The Franklin’s randomized response model for two sensitive attributes. Sect Surv Res Methods 2009:4171–4185
Christofides TC (2003) A generalized randomized response technique. Metrika 57:195–200
Christofides TC (2005) Randomized response technique for two sensitive characteristics at the same time. Metrika 62:53–63
Drane W (1976) On the theory of randomized response to two sensitive questions. Commun Stat Theory Methods 5(6):565–574
Foutz RV (1977) On the unique consistent solution to the likelihood equations. J Am Stat Assoc 72:147–148
Franklin LA (1989) A comparison of estimators for randomized response sampling with continuous distributions from a dichotomous population. Commun Stat Theory Methods 18(2):489–505
Gjestvang CR, Singh S (2006) A new randomized response model. J R Stat Soc Ser B 68:523–530
Greenberg BG, Abul-Ela A, Simmons WR, Horvitz DG (1969) The unrelated question randomized response model: theoretical framework. J Am Stat Assoc 64:520–539
Horvitz DG, Shah BV, Simmons WR (1967) The unrelated question randomised response model. In: Proceedings of the social statistics section, American Statistical Association, pp 65–72
Kim JI, Flueck JA (1978) An additive randomized response model. In: Proceedings of social statistics section, American Statistical Association, pp 351–355
Kuk AYC (1990) Asking sensitive questions indirectly. Biometrika 77:436–438
Lee CS, Sedory SA, Singh S (2013) Estimating at least seven measures of qualitative variables from a single sample using randomized response technique. Stat Probab Lett 83:399–409
Lensvelt-Mulders GJLM, Hox JJ, van der Heijden PGM, Maas CJM (2005) Meta-analysis of randomized response research: thirty-five years of validation. Sociol Methods Res 33:319–348
Mangat NS (1994) An improved randomized response strategy. J R Stat Soc Ser B 56:93–95
Mangat NS, Singh R (1990) An alternative randomized response procedure. Biometrika 77:439–442
Moshagen M, Musch J, Ostapczuk M, Zhao Z (2010) Reducing socially desirable responses in epidemiologic surveys. Epidemiology 21(3):379–382
Moshagen M, Musch J (2012) Surveying multiple sensitive attributes using an extension of the randomized response technique. Int J Public Opin Res 24:508–523
Mukherjee R (1981) Inference on confidential characters from survey data. Calcutta Stat Assoc Bull 30:77–88
Mukhopadhyay P (1980) On the estimation of some confidential characters from survey. Calcutta Stat Assoc Bull 29:77–88
Seil KS, Desai MM, Smith MV (2014) Sexual orientation, adult connectedness, substance use, and mental health outcomes among adolescents: findings from the 2009 New York city youth risk behavior survey. Am J Public Health 104(10):1950–1956
Silva LC (1983) On the generalized randomized response model with polychotomous variables. Rev Invest Operac 4(III):75–100
Tamhane AC (1981) Randomized response techniques for multiple attributes. J Am Stat Assoc 76:916–923
Ward BW, Dahlhamer JM, Galinsky AM, Joestl SS (2014) Sexual orientation and health among U.S. adults: national health interview survey, 2013. National Health Statistics Reports; No. 77. Hyattsville, MD: National Center for Health Statistics
Warner SL (1965) Randomized response: a survey technique for eliminating evasive answer bias. J Am Stat Assoc 60:63–69
Zou G (1997) Two-stage randomized response procedures as single stage procedures. Aust J Stat 39:235–236
Acknowledgments
The authors are grateful to an Associate Editor and two referees for their helpful comments that improved the presentation. The research of S.M. Lee and S.H. Hsieh were supported by the Ministry of Science and Technology (MOST) of Taiwan, ROC (103-2420-H-035-001-2R, 103-2118-M-035-003-MY2 and 102-2118-M-001-002-MY2, respectively). The authors would like to thank research fellows Ying-Hwa Chang for adding the new RR technique by the research project “Taiwan Social Change Survey: Year 6 of cycle 3”, which was sponsored by the MOST. The Survey Research Data Archive, Academia Sinica is responsible for the data distribution. The authors appreciate the assistance of the aforementioned institutes and individuals in providing all relevant data.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hsieh, SH., Lee, SM. & Tu, SH. Randomized response techniques for a multi-level attribute using a single sensitive question. Stat Papers 59, 291–306 (2018). https://doi.org/10.1007/s00362-016-0764-9
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-016-0764-9