Date

12-18-2015

Document Type

Thesis

Abstract

In order to understand and make informed decision on sensitive topics such as domestic violence and drug use, interviews have been used to collect data. However it is difficult to assess how truthful respondents are since they may not feel at ease revealing the truth to an interviewer. Surveyors of sensitive issues face the problem that respondents may be reluctant to answer truthfully since the respondent may feel pressured socially or may fear the repercussions of their truthful answer. Processes known as random response techniques have been introduced to allow interviewers the ability to extract information they need for a population, while preserving privacy of individual respondents by introducing randomness to the surveying process. The focus of this project was to critically assess the accuracy of two such techniques and the level of privacy protection afforded to the respondent by the techniques. Using the method of moment matching, estimators of the population proportion were created for both techniques. Once the estimators were created, the accuracy of the estimators could be assessed by studying the variance. Finally we measured the level of privacy protection the techniques afforded to respondents by calculating mutual information and entropy.

Department

Mathematics

Thesis Comittee

Dr. Kevin Rion (Thesis Director)

Dr. Irina Seceleanu

Dr. Mahmoud El Hashash

Copyright and Permissions

Original document was submitted as an Honors Program requirement. Copyright is held by the author.

Download

Included in

Mathematics Commons

Share

COinS