Abstract
본 연구에 사용된 국민건강영양조사 데이터는 복합표본설계방법인 2단계층화확률추출법에 의하여 표본추출되었으며 층화, 집락, 가중치 등의 요소를 반영한 데이터이다. 이와 같은 데이터를 단순임의표본추출로 간주하여 분석할 경우 분산 추정치에서 편향된 결과를 얻을 수 있다. 따라서 표본의 대표성과 부정확한 분산 추정을 고려하여 결측 자료의 처리와 복합표본설계의 3요소인 가중치, 층, 집락 요소를 반영하여 분석해야 한다. 단순임의표본분석과 복합표본분석에 사용되는 통계기법에는 차이가 있으며, 복합표본설계에 의한 범주형 자료의 경우 피어슨(Pearson) 카이제곱검정에 필요한 조건을 만족하지 못하여 검정력을 증가시키므로 라오-스콧(Rao-Scott) 카이제곱검정을 하여야 한다. 따라서 복합표본추출법(complex sampling)과 단순임의표본추출법(simple random sampling), 표본 가중치(sample weights)를 적용한 단순임의표본추출법의 세 가지 방법에 대한 차이를 파악하고자 한다.The data of the Korean national health and nutrition examination survey (KNHANES) used in this research is sampled by the two-stage sampling; complex sampling, and it reflects some elements such as stratification, clustering and weights. If we analyze such data as simple random sampling, we can obtain the biased result in the variance estimates. Therefore, considering representativeness of sample and inaccurate variance estimate, we should analyze it reflecting the processing of missing value, and weights, stratification and clustering which are three elements of complex sample design. There are some differences in statistics methods which are used in simple random sampling analysis and complex sampling analysis, and in case of categorial data that is of complex sample design, it should conduct Rao-Scott test due to the fact that the data does not satisfy required condition for Pearson chi-square test and it rather increases power. For this reason, we look at the differences about three methods; complex sampling, simple random sampling, and simple random sampling which applies sample weight.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.