Activity Number:
|
314
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Survey Research Methods*
|
Abstract - #300417 |
Title:
|
A New Multinomial Distribution Approach To a Quantitative Randomized Response Technique
|
Author(s):
|
Jong-Min Kim*+ and William Warde
|
Affiliation(s):
|
University of Minnesota and Oklahoma State University
|
Address:
|
Division of Statistics, Morris, Minnesota, 56267, USA
|
Keywords:
|
Multinomial Distribution ; Response Design Matrix ; Multiple contrast method ; Product-Moment Correlation
|
Abstract:
|
A new quantitative randomized response technique is presented in this paper. The proposed technique will use a Hopkins' randomizing device to derive a multinomial distribution for sensitive categories. After obtaining the observed estimates for sensitive category proportions, which also include the random responses from the Hopkins' randomizing device, we derive the true estimates of the proportions for the sensitive categories in a situation where a model accounting for the respondent to lie is used. For contingency tables, we derive a Pearson product-moment correlation between two different sensitive questions.
|