• 응용통계연구
• /
• 제30권1호
• /
• pp.69-82
• /
• 2017

R-지수(representativeness indicator)는 무응답이 발생했을 때 표본의 대표성을 나타내주는 지표이다. 표본의 대표성은 모수 추정의 정확성(accuracy)과 관계가 있으며 정확성은 편향(bias)와 관계가 있다. 따라서 표본의 대표성을 나타내는 R-지수가 높으면 대표성이 높아 편향이 없고 정확성이 높은 결과를 얻을 수 있다. R-지수는 일반화선형모형의 로짓 또는 프로빗 모형을 적합한 후 얻어진 경향 점수(propensity score)에 의해 계산된다. 본 논문에서는 R-지수와 이질적인 층별 응답률과의 관련성을 연구하였으며 편향, 제곱근 RMSE 등과 같은 비교통계량이 무응답에 얼마나 민감한지 등을 모의실험을 통하여 살펴보았다. 또한 변형된 2010년 경제총조사 자료를 이용하여 실제 자료분석도 실시하였다.

• 대한원격탐사학회지
• /
• 제25권4호
• /
• pp.321-338
• /
• 2009

이 연구의 목적은 비대칭 분포를 가지는 현장 조사 자료로부터 GIS 기반 주제도를 생성하기 위한 공간 내삽 방법으로 단변량 크리깅 기법을 비교하는데 있다. 기존 정규 크리깅과 비선형 자료 변환에 기반을 둔 로그 정규 크리깅, 다중 가우시안 크리깅과 지시자 크리깅을 지화학 원소 비소와 납에 대해 사례 연구를 통해 비교하였다. 예측 능력의 비교 분석을 위해 leave-one-out 기반 교차 검증을 통한 오차 분석을 수행하였으며, 샘플링 밀도의 차이에 따른 오차의 변화 양상도 분석하였다. 비교 분석 결과, 지시자 크리깅이 전반적으로 가장 높은 예측 능력을 나타내었으며, 작은 값과 높은 값의 예측 능력도 우수한 것으로 나타났다. 정규 크리깅에 비해 비선형 자료 변환 기반 크리깅 기법들이 우수한 예측 능력을 나타내었지만, 기존에 많이 적용된 로그 정규 크리깅은 샘플링 밀도와 상관없이 편향 정도가 가장 크게 나타내었다. 이 연구를 통해 얻어지는 정량적 검증 결과는 비대칭 분포를 가지는 현장 조사 자료의 내삽을 위한 크기깅 기법의 선정에 유용하게 이용될 수 있을 것으로 기대된다.

• 보건교육건강증진학회지
• /
• 제24권4호
• /
• pp.41-64
• /
• 2007

Objectives: The purpose of this study was to verify the validity of the evaluation indicators on safety and health education programs provided to Korean industrial workers, by developing further. Methods: To develop the early evaluation indicator, the secondary analysis was made on 'Survey on Industrial safety & health trends', the national wide survey data product by Korea Occupational Safety & Health Agency, as well as the various examinations on literatures. To validate such developed evaluation indicator, 13 safety and health managers in Focus Group discussion to prove the reliability of the contents were composed. To confirm the face validity, the interview with 6 industrial safety and health education experts was held to study the appropriateness of the content. With the cooperation from Korea Occupational Safety & Health Agency, the structured survey was performed with the safety and health managers from 588 businesses who participated in the education program for the later half of 2006. In this study, the number of responses was randomly divided into 1:1 two groups, in which one group was for the Exploratory Factor Analysis and the other group was for the Confirmatory Factor Analysis. Results: The results were as follows. First, for the Structure category, 18 evaluation indicators were developed into 4 evaluation categories such as a preliminary examination, education staff and organization, education environment and rules/ regulation. Secondly, the Activity category was comprised of 10 evaluation indicators with respect to 2 areas of curriculum satisfaction and program evaluation. Finally, for the Output category, 9 evaluation indicators were developed into 2 evaluation areas such as documentation and information share and education effects. Conclusions: The evaluation indicators developed through this study will possibly develop and be applied to evaluation tools on safety and health education program, which may further become the standardized indicators to better operate the industrial safety and health education programs.

• 산업경영시스템학회지
• /
• 제30권4호
• /
• pp.118-126
• /
• 2007

This study presents the intellectual capital (IC) indicator selection criteria based on the review of qualitative characteristics of financial accounting information and data quality of information systems. The AHP survey has been performed to verify the weight difference of the criteria to select effective IC indicators among industry. The results shows that there exists the weight differences in relevance and comparability factor by industry. To management IC effectively, different method considering industry environment should be required.

• 보건의료산업학회지
• /
• 제13권3호
• /
• pp.1-12
• /
• 2019

Objectives: The purpose of this study was to provide hospital management diagnosis status and basic data required for the future development of hospital management diagnosis program. Methods: We conducted a questionnaire survey on administrative staff of manager level or over of medical institutions in B metropolitan city. Results: As a result of analyzing the relative influence of the needs by hospital management diagnosis indicator, the effect of financial analysis indicator, patient treatment record indicator and medical revenues indicator were high in the medical institutions with number of beds of 100 beds or over and general hospital level or over both on a hospital level and on an individual level. Conclusions: Since the existing laws or systems are centered on large major hospitals, the management environment is very unfavorable for small and medium hospitals as can be seen from the results of this study. Therefore, the government should improve the transparency and rationality of the hospital management environment in Korea through regulation and system reforms that can be applied to all medical institutions.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2015년도 제51차 동계학술대회논문집 23권1호
• /
• pp.299-302
• /
• 2015

사회복지 지표조사는 주민들의 사회복지에 관한 다양한 상태를 전체적으로 파악할 수 있는 조사로 다양한 복지정책 개발에 있어 지역 주민들의 욕구를 반영할 수 있는 장점이 있다. 따라서 사회복지 지표조사는 복지 욕구를 알 수 있는 중요한 척도의 기준이라 할 수 있어 많은 지자체 및 정부기관에서 예산과 시간을 들여 조사를 실시하고 있다. 그러나 조사에 대한 분석 결과가 기초적인 통계 분석 위주로 되어 있어 실제 자료를 제대로 할용하지 못하고 있는 것이 현실이다. 이에 본 논문에서는 데이터 마이닝 기법을 이용한 분석 방법론을 제시한다. 또한 이를 충청북도의 중장기 사회복지 정책을 위한 지표조사에 적용하였다.

• Journal of the Korean Data and Information Science Society
• /
• 제17권1호
• /
• pp.1-11
• /
• 2006

Data mining technique is used to find hidden knowledge by massive data, unexpectedly pattern, relation to new rule. The methods of data mining are decision tree, association rules, clustering, neural network and so on. Clustering is the process of grouping the data into clusters so that objects within a cluster have high similarity in comparison to one another. It has been widely used in many applications, such that pattern analysis or recognition, data analysis, image processing, market research on off-line or on-line and so on. We analyze Gyeongnam social indicator survey data by 2001 using twostep clustering technique for environment information. The twostep clustering is classified as a partitional clustering method. We can apply these twostep clustering outputs to environmental preservation and improvement.

• Journal of the Korean Data and Information Science Society
• /
• 제16권1호
• /
• pp.59-69
• /
• 2005

Data mining is the method to find useful information for large amounts of data in database It is used to find hidden knowledge by massive data, unexpectedly pattern, relation to new rule. The methods of data mining are decision tree, association rules, clustering, neural network and so on. We analyze Gyeongnam social indicator survey data by 2001 using association rule technique for environment information. Association rule mining searches for interesting relationships among items in a given large data set. Association rules are frequently used by retail stores to assist in marketing, advertising, floor placement, and inventory control. There are three primary quality measures for association rule, support and confidence and lift. We can use to environmental preservation and environmental improvement by association rule outputs

• 동의신경정신과학회지
• /
• 제28권4호
• /
• pp.373-379
• /
• 2017

Objectives: The purpose of this research is to study the psychological characteristics of Korean medical students with a focus on the Myers-Briggs Type Indicator (MBTI) and the Minnesota Multiphasic Personality Inventory-2 (MMPI-2). Methods: The survey was conducted on 101 Korean medical students to investigate their psychological characteristics with a focus on Myers-Briggs Type Indicator and the Minnesota Multiphasic Personality Inventory-2. Results: Among the 16 MBTI personality types, ISTJ was the most common type with the prevalence of 24.8%. According to gender-based classification, there were more men in the J-type category and more women in the P-type category. In the MMPI-2 scales, males showed high scores in Pd, AGGR, DISC, MAC-R, and GM, and females showed high scores in Mf and GF. Conclusions: The personality test for male and female Korean medical students revealed few differences on some scales of the Myers-Briggs Type Indicator and the Minnesota Multiphasic Personality Inventory-2.

• Environmental Analysis Health and Toxicology
• /
• 제30권sup호
• /
• pp.3.1-3.11
• /
• 2015

Objectives This study presents the indicators and proxy variables for the quantitative assessment of green chemistry technologies and evaluates the relative importance of each assessment element by consulting experts from the fields of ecology, chemistry, safety, and public health. Methods The results collected were subjected to an analytic hierarchy process to obtain the weights of the indicators and the proxy variables. Results These weights may prove useful in avoiding having to resort to qualitative means in absence of weights between indicators when integrating the results of quantitative assessment by indicator. Conclusions This study points to the limitations of current quantitative assessment techniques for green chemistry technologies and seeks to present the future direction for quantitative assessment of green chemistry technologies.