Earticle

Categorical data collected based on complex sample design is not proper for the standard Pearson multinomial-based chi-squared test because the observations are not independent and identically distributed. This study investigates effects of bias of point estimator of population proportion and its variance estimator to the standard Pearson chi-squared test statistics when the sample is collected based on complex sampling scheme. This study examines the effect under two population homogeneity test. The standard Pearson test statistic can be partitioned into two parts; the first part is the weighted sum of ?? with eigenvalues of design matrix as their weights, and the additional second part which is added due to the biases of the point estimator and its variance estimator. Our empirical analysis shows that even though the bias of point estimator is small, Pearson test statistic is very much inflated due to underestimate the variance of point estimator. In the connection of design-based variance estimator and its design matrix, the bigger the average of eigenvalues of design matrix is, the larger relative size of which the first component part to Pearson test statistic is taking.

Measurement error is one of main source of error in survey. It is generally defined as the difference between an observed value and an underlying true value. An observed value with error may be expressed as a function of the true value plus error term. In some cases, the measurement error variance may be also a function of the unknown true value. The error variance function can be rewritten as a function of true value multiplied by a scale factor. This research explore methods for estimation of the measurement error variance based on the data from complex sampling design. We consider the case in which the variance of mesurement error is a linear function of unknown true value, and the error variance scale factor is small. We applied our results to the U.S. Third National Health and Nutrition Examination Survey (the U.S. NHANES Ⅲ) data for empirical analyses, which has replicate measurements for relatively small subset of initial respondents's group.

본 연구는 신체활동수준, 주관적 건강상태, 코로나19 염려의 관계를 분석하는 것에 목적이 있다. 이를 위하여 2020년 지역사회건강조사를 바탕으로 전국 만 19세 이상 성인 229,269명의 설문데이터 자료를 분석하였다. 복합표본설계로 가중치, 층화변수, 집락변수를 지정하여 분석하였다. SPSS 통계분석 프로그램을 이용하여 복합표본 빈도분석, 복합표본 교차분석, 복합표본 회귀분석을 실시하고, 다음과 같은 결과를 얻었다. 첫째, 신체활동수준이 높은 그룹이 신체활동수준이 낮은 그룹에 비하여 주관적 건강상태가 높은 것으로 나타났다. 둘째, 신체활동수준이 높은 그룹이 신체활동수준이 낮은 그룹에 비하여 코로나19 염려가 낮은 것으로 나타났다. 셋째, 주관적 건강수준이 높은 그룹이 낮은 그룹에 비하여 코로나19 염려가 낮은 것으로 나타났다. 다만 본 연구에서는 실제 코로나19 감염여부는 분석에 포함하지 않았다는 제한점이 있다.

This study tried to analyze the relationship among physical activity level, subjective health status, COVID-19 Fear. This study used the 2020 Community Health Survey that includes 229,269 survey data from adults over 19 years old. The complex sampling design was applied including weight, stratification, cluster variables. Through the SPSS statistics program with complex sampling frequency analysis, complex sampling Chi-square and complex sampling regression, this study found followings. First, the group with high level of physical activity showed higher level of subjective health status than the group with low level of physical activity. Second, the group with high level of physical activity showed lower level of COVID-19 fear than the group with low level of physical activity. Third, the group with high level of subjective health status showed lower level of COVID-19 fear than the group with low level of subjective health status. However, this study has the limitation that this study did not check whether participant is diagnosed with Covid-19 or not.

This study was motivated by the National Assessment of Educational Achievement (NAEA) involving a complex survey design for sampling. The goal of the study was to explore the benefit of weighted analysis in multilevel modeling. Large-scale data such as the NAEA data are often analyzed in multilevel analyses to reflect the nested data structure. A simulation study was conducted incorporating different conditions. Multilevel analysis results were compared without weights, with basic design weights, and with a raked weight, which were generated in consideration of explicit and implicit stratification variables with auxiliary variables. Simulation also compared the analysis results by varying the number of classes per school in a stratified cluster sampling design. In summary, substantial benefits were not observed in weighted multilevel analysis compared to unweighted multilevel analysis, although the weighted descriptive statistics were close to the population means relative to the unweighted descriptive statistics. These findings may help provide scientific evidence for the use of sampling weights in analyses with large-scale educational data.

본 연구에서는 농촌 삶의 실태와 농산어촌 주민들의 만족도를 파악하기 위한 표본설계방안을 마련하고자 한다. 기존의 농촌통계조사의 표본설계 전 과정을 검토해서 현행 조사의 문제점 등을 개선하고, 과거 표본조사자료로부터 지역별, 항목별 특성을 분석하여 표본배정을 위한 변수와 층화변수 등을 찾아 표본설계에 반영하고자 한다. 그리고 2005년 인구주택총조사의 90% 모집단조사구를 조사모집단으로 설정하여 층화방법, 표본크기, 표본배정방법 등을 결정한다. 또한 가중치와 추정량, 그리고 추정량에 대한 오차공식 등을 제시하며, 표본조사구와 예비표본조사구를 추출하고 표본조사구내에서 표본가구의 추출방법과 표본대체 방법 등을 제시한다. 마지막으로 효율적인 조사관리 방법과 향후 조사의 개선방안을 다루고자 한다.

In this paper we propose the complex sampling design for survey of the life style and satisfaction of agricultural and fishery in Korea. We consider the survey design procedure in the previous agricultural and fishery survey, improve the problems in current survey and analyze the past data and determine the method of stratification, the size of sample, and the method of sample allocation. The sampling frame is considered 90% survey population from the census in 2005 conducted by National Statistical Office. In addition, we suggest the weight and estimator, the formula of sampling error, the sampling method of households, and the method of substitution. Finally we discuss with the method of survey management and improvement in future.

연구목적: 사회활동 유형에 따라 개인의 주관적 건강수준과 삶의 만족도에 차이가 있는지 알아보고자 한다. 연구방법: 2017년 지역사회건강조사 자료를 활용하여 225,476명의 자료로 주관적 건강수준과 삶의 만족도에 영향을 미치는 인구사회적 요인과 건강행동 요인을 통제하고 사회활동의 영향을 분석하였다. 다중회귀분석을 실시하였으며, 모든 분석은 복합표본분석을 실시하였다. 연구결과: 종교활동과 자원봉사 같은 의미를 추구하는 활동보다 친목활동과 여가/레저활동 같은 즐거움을 추구하는 활동에서 주관적 건강과 삶의 만족도 수준이 높았고, 두 사회활동 모두 주관적 건강과 삶의 만족도에 긍정적인 영향을 미치는 것으로 나타났다. 결론: 사회활동 유형에 따라 주관적 건강과 삶의 만족도에 차이가 있었다. 향후 다양한 사회활동을 포함하여 한국적 상황과 특성이 반영된 결과가 도출될 필요가 있다.

Objective: This study was performed to examine the difference in self-rated health and life satisfaction according to types of social activity. Methods: The data from 225,476 participants in the Korean Community Health Survey 2017. Multiple regressions were performed using SPSS statistical software (ver. 25.0), and the Complex Samples module to adjust for stratification, clustering, and weight. Results: When comparing the social activities related with self-rated health and life satisfaction after adjusted sociodemographic characteristics. In terms of hedonism seeking activity, self-rated health and life satisfaction were higher than eudaimonism seeking activity. Both the eudaimonism seeking activity and the hedonism seeking activity had positive effects on self-rated health and life satisfaction. Conclusions: This study showed that self-rated health and life satisfaction were significantly influenced hedonism and eudaimonism seeking activities.