COMPLEX SURVEY DESIGNS for SIMULTANEOUS ESTIMATION of TOTALS of TWO SENSITIVE VARIABLES

• Activity Number: 616 - The Use of Randomized Response Technique (RRT) in Theory and Practice and Recent Developments - Second Round Toward Collecting Sensitive Data: Making Lies Naked!
• Type: Topic Contributed
• Date/Time: Thursday, August 3, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Social Statistics Section
• Author(s): Segun Ahmed* and Sarjinder Singh and Stephen A. Sedory
• Companies: Texas A & M University-Kingsville and Texas A&M University-Kingsville and Texas A&M University-Kingsville
• Keywords: TWO SENSITIVE VARIABLES ; ESTIMATION OF TOTALS ; VARIANCE ; PROTECTION OF RESPONDENTS

Abstract

COMPLEX SURVEY DESIGNS FOR SIMULTANEOUS ESTIMATION OF TOTALS OF TWO SENSITIVE VARIABLES

Segun Ahmed, Stephen Sedory and Sarjinder Singh Department of Mathematics Texas A&M University-Kingsville Kingsville, TX 78363

ABSTRACT

Ahmed, Sedory and Singh (2016) have introduced a pioneer idea problem of simultaneous estimation of means of two quantitative sensitive variables by using only one randomized response and another pseudo response from a respondent in a sample. They considered the use of simple random and stratified random sampling. In this paper, we have considered the problem of estimation of population totals of two sensitive variables by making use of complex survey designs. The bias and variance expression of the resultant estimators are derived. The performance of the new estimators has been investigated through an extensive simulation study using R programming.

Keywords: Two sensitive variables, estimation of totals, variance, protection of respondents.