Advances and Applications in Statistics
Volume 11, Issue 1, Pages 15 - 28
(February 2009)
|
|
RANDOMIZED RESPONSE WITHOUT A RANDOMIZATION DEVICE
Tasos C. Christofides (Cyprus)
|
Abstract: Randomized response techniques are methods designed to estimate the prevalence of a stigmatizing characteristic which cannot be estimated with conventional survey methodology procedures. Each participant using a randomization device provides a response from which it is not possible to infer whether he/she has the sensitive characteristic. Using the responses from all sampled persons the investigator can estimate the population proportion of people belonging to the stigmatizing group. However, all randomized response techniques make use of a randomization device. In many cases, using the device requires considerable amount of time, creates confusion to the participants and above all, raises suspicions as to whether the privacy of the individuals participating in the survey is really protected. Finally, the use of the device makes the collection of information via telephone or the internet almost impossible and most of the time requires face to face interviews. In this paper, we present a method which can be used to collect information about a sensitive attribute without using a randomization device while at the same time the privacy of the participants is protected. The method can be used in telephone and internet interviewing and can also be incorporated in extensive questionnaires. |
Keywords and phrases: randomized response, protection of privacy, simple random sampling, stigmatizing characteristic. |
|
Number of Downloads: 345 | Number of Views: 955 |
|