|
KMID : 0379920000250010027
|
|
Journal of The Korea Socity of Health Informatics and Statistics
2000 Volume.25 No. 1 p.27 ~ p.35
|
|
|
Independence and Conditional Independence Tests in the Repeated Randomized Response Models
|
|
|
|
|
Abstract
|
|
|
Krishnamoorphy and Raghavargo(1993) invented exact binomial and asymptotically normal test procedures for truthful answering in the repeated randomized response models under the assumption that two repeated response measures are independent.
Under the same assumption, Lakshmi and Raghavargo(1992) suggested asymptotic chi-square test for respondents´ truthful answering in the same models. Meanwhile, Lee(1999) suggested asymptotic test for detecting consistent answering in the repeated randomized models without the assumption of independence between two repeated response variables.
In this article we numerically detect the factors and the conditions of them, with which two response variables might be independent. And we find a weaker condition under which the test procedures suggested by Krishnamoorphy and Raghavarao(1993) can be applicable. That is, under the assumption of conditional independence between two response variables rather than that of independence we have the exactly same joint probability distribution of two variables and the same test statistics therein. Hence we can apply the same logical statements on deriving the tests for truthful answering in the repeated randomized response models as in Krishnamoorphy and Raghavarao(1993). Both chi-square test and Cochran-Mantel-Haeszel test are used to test independence and conditional independence between two respondence variables, respectively.
|
|
|
KEYWORD
|
|
|
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|