통계 교육과 상담: 현황과 과거 10년의 변천

송혜향,안윤기,이용구,조신섭,허명회 한국통계학회 1991 Journal of the Korean Statistical Society Vol.20 No.-

본 논문에서는 우리나라 대학교에 있어서의 통계학과의 현황과 대학교육에서의 통계학과의 발전과정을 논의하고자 한다. 통계학과 설립 및 석.박사과정 개설이 지난 10년동안 두드러지게 많았으며 그 발전된 현황을 파악하기 위해 설문조사를 실시하였고 본 논문에서 그 결과를 알리고 1970년대의 상황과도 비교하고자 한다. 1980년대에 들어서서 대학의 통계 교과과정에 대한 연구가 최지훈 외(1981)와 이용구(1989)에 의해 두차례 조사되었다.

Competing Risk 분석에서의 상호작용에 관한 고찰

송혜향 한국보건통계학회 1981 한국보건정보통계학회지 Vol.6 No.1

호스피스 간호에 대한 연구방법론

송혜향 가톨릭대학교 가톨릭대학교 간호대학 호스피스 교육연구소 1997 호스피스논집 Vol.2 No.-

Statistical Problems in the Determination of Normal Manges from Laboratory Data

송혜향,Song, Hae-Hiang The Korean Society for Preventive Medicine 1984 Journal of Preventive Medicine and Public Health Vol.17 No.1

Normal range has use mainly in the first phase of the diagnostic process, that is, to screen or to raise ideas about possible pathology. The traditional method of determining it is based on the probability paper or on the mean plus or minus two standard deviations. These methods are often turned out to be vague and impractical. The percentile method is adequate and flexible, though. The appropriate limit of lower and upper points should be chosen by considering medical aspects above all things and also the reliability of the range determined by the standard error. The results of normal range are interesting, strictly speaking, only for the hospital concerned. Differences exist between the normal ranges reported by various sources (Bezemer et al, 1983). It would be best to establish the normal range based on a population comparable to a group of individuals to whom the normal range is to serve as a norm. 질병의 진단이나 예후를 판정함에 도움이 되고자 의학분야에서 정상 범위를 정의해 왔으며, 이 정의에 도입된 전통적인 통계적 방법으로는 정규 확률지를 사용하여 정상범위를 추정함이었다. 이 방법은 정상군과 질병군의 검사 성적 분포가 각각 정규 분포한다는 가정에서 시작되며, 이 가정이 보편적일 수 없음을 여러 연구(Elveback et al, 1970; Feinstein, 1977)에서 밝히고 있다. 또한 통계적 방법론중 일상 사용되고 있는 평균치로부터 두배의 표준 편차 거리로 나타 내어지는 구간을 정상 범위로 추정함은 비대칭 분포의 경우 합당치 못한 경우가 생기게 된다. 대칭과 비대칭의 분포에 합당하게 사용될 수 있는 방법은 백분율에 근거한 방법으로 특정한 의학적인 요구가 없는 한 균등한 하단과 상단의 백분율 구간을 정함이 상례였다고 보겠다. 그러나 높은 칼슘의 수치를 나타내 보이는 환자에게 주의가 필요함과 같이 특수한 의학적인 촉구에 의해 하단과 상단의 불균등한 백분율 지점을 정상범위로 선정할 수도 있으나 백분율의 표준 오차에 근거해서 정상 구간이 추정되어야 하겠다.

Microarray 자료의 통계적 표준화 방법의 비교

송혜향,윤상헌,김제석 한국보건통계학회 2003 보건정보통계학회지 Vol.28 No.1

Microarrays can provide detailed information of the expression patterns of thousands of genes in tumours, but due to systematic variations such as dyes, hybridization and processing precedures, and technical attefacts, true biological differences in gene expression levels cannot easily be detected. If the vast quantity of these microarray data can be suitably normalized, the impact of this genomic expression data on the understanding of diagnosis and treatment of disease will be profound. Statistical normalization approaches to microarray data are not as yet fully explored and need to be continuously evaluated. We compared the currently much used normalization methods and recently proposed statiostical normalization methods of ANOVA with oligonucleotide microarray data of hepatic nocular lesions indluding hepatocellular carcinoma. This article proposes combined normalization methods of ANOVA after a robust local lowess, intensity and print- tip dependent, regression.

Identifying Copy Number Variants under Selection in Geographically Structured Populations Based on F-statistics

송혜향,정연준,허혜진,석인혜 한국유전체학회 2012 Genomics & informatics Vol.10 No.2

Large-scale copy number variants (CNVs) in the human provide the raw material for delineating population differences, as natural selection may have affected at least some of the CNVs thus far discovered. Although the examination of relatively large numbers of specific ethnic groups has recently started in regard to inter-ethnic group differences in CNVs, identifying and understanding particular instances of natural selection have not been performed. The traditional FST measure, obtained from differences in allele frequencies between populations, has been used to identify CNVs loci subject to geographically varying selection. Here, we review advances and the application of multinomial-Dirichlet likelihood methods of inference for identifying genome regions that have been subject to natural selection with the FST estimates. The contents of presentation are not new; however, this review clarifies how the application of the methods to CNV data, which remains largely unexplored, is possible. A hierarchical Bayesian method, which is implemented via Markov Chain Monte Carlo, estimates locus-specific FST and can identify outlying CNVs loci with large values of FST. By applying this Bayesian method to the publicly available CNV data, we identified the CNV loci that show signals of natural selection, which may elucidate the genetic basis of human disease and diversity.

비례승산비를 따르지 않는 순위형 자료의 메타분석법

송혜향,김현재 한국보건통계학회 2003 보건정보통계학회지 Vol.28 No.1

Researches are undertaken ever-increasingly to find innovative treatments, and thus published papers are accumulating at an enormous rate. Meta-analysis enable previously published study results to be reviewed and summarized, and the meta-analysis methodologies differ depending on continuous or discrete outcomes. Ordinal outcomes are often found in medicine a few examples are the number of injurious fall, the degree of dementia or degree of pain. The summary statistics of these ordinal data are Odds Ratio(OR), Relative Risk(RR) and Risk Difference(RD). In this paper the generalized odds ratio, generalized relative risk and generalized risk difference (Edwardes and Baltzan, 2000)are proposed as summary statistics of the 2×R contingency table data and meta-analysis method based on generalized association estimates is proposed. Recently, Whitehead et al.(2001) suggested logistic regression models for meta-analysis of 2×R contingency table data and these models require th proportionality of odds ratio. In this paper we compare meta-analysis methods based on the generalized estimates with logistic regression models, using the data of Whitehead et al.(2001, 1994) and also the fall data which were collected in order to evaluate the efficacy of exercise programs. The proportional odds assumption does not hold for some study data in the exercise intervention programs and thus the generalized estimates would be more suitable. Simulations were carried out based on the above 2×R contingency table data to evaluate association measures, tests of treatment efficacy and homogeneity. Even with these limited simulations it was shown that much used RD with arbitrarily dichotomizing ordinal data are not a recommended strategy and that logistic regression models present sometimes wrong estimates when the assumption of proportional odds ratio does not hold

임상시험 결과에 대한 통계적 분석

송혜향 한국역학회 1989 Epidemiology and Health Vol.11 No.1

자궁경부암 환자의 우울과 관련된 요인조사 : 자궁적출술 및 근치적 방사선치료를 받은 환자를 중심으로

송혜향,소향숙 연세대학교 간호정책연구소 1996 간호학탐구 Vol.5 No.1

In this descriptive correlational study, 45 newly diagnosed uterine cervical cancer patients were interviewed to determine the predictors of depression scores at the post-operative 3 month. The theoretical framework was derived basically from the Lazarus & Folkman's process model of stress-coping-adaptation and cognitive model of depression. A total of 45 subjects from one general hospital in Kwang-ju city were observed and interviewed during 7 months with the aid of 7 structured instruments: Mishel's Uncertainty in Illness scale, Folkman & Lazarus's W.C.C.L., Beck's Depression Inventory, Pollock's health- related hardiness scale, Tai's family support scale, Rosenberg's self-esteem scale, and records on the demographic and medical factors. The data obtained were analyzed using percentage, paired t-test, Stepwise Multiple Regression and Cronbach's Alpha Coefficients. The results were as follows: 1. A majority of patients was between 30 and 50 years of age and 80 0 of women had stage I cancer. 2. The mean score of the perceived uncertainty was 1.64 + 0.69. Those of problem-focused coping and emotion-focused coping were 1.44 + 0.59 and 1.23 + 0.49, respectively and the former was more used significantly(t= 2.54, p= .0147). 3. The score of depression on the post-operative 1 month was 19.96 + 10.43 and 15.47 + 9.58 on the post-operative 3 months. So the depression scores were significantly decreased from 1 month to 3 months post-operative(t= =3.77, p= .0005). 4. When uncertainty score was entered into the equation as the dependent variable, the stepwise procedure resulted in the inclusion of cancer staging, explaining 29.7%(B= 2.10, p= .034) of the variation in the uncertainty score. 5. When problem-focused coping score was entered into the equation as the dependent variable, the stepwise procedure resulted in the inclusion of the presence or absence of an ovary, explaining 11.9%(B= -.527, p= ,035) of the variation in problem-focused coping score, self-esteem, explaining 9.4%(B= .300, p= .033), uncertainty, explaining 3.2%(B= .287, p= .070). 6. When emotion-focused coping score was entered into the equation as the dependent variable, the stepwise procedure resulted in the inclusion of the education level, explaining 6.5%(B= .363, p= .001) of the variation in emotion-focused coping score, perceived family support, explaining 8.8%(B= .332, p= .008), uncertainty, explaining 8.4%(B= .340, p= .018), age, explaining 3.7%(B= .026, p= .049), the induced anti-cancer chemotherapy, explaining 7.5%(B= .306, p= .086). 7. When the depression score at the post-operative 3 month was entered into the equation as the dependent variable, the stepwise procedure resulted in the inclusion of uncertainty, explaining 45.2%(B= .255, p= .031) of the variation in the depression scores, self-esteem, explaining 5.3%(B= -.229, p= .034), cancer staging, explaining 3.7(B= .138, p= .036), and hardiness, explaining 3.3%(B= .427, p= .038), respectively. However, when depression at the post-operative 1 month was entered as another independent variable, this depression variable itself explained 58.8%(B= .515, p= .0002) of the variation, and neither uncertainly, self-esteem, nor cancer staging were significant predictors. In view of the results of the present study that cognitive variables explained depression scores better than coping scores, a longitudinal study design is much more powerful compared to the one-time-measurement study design for studying Lazarus & Folkman's process model of stress-coping-adaptation and cognitive model.

메타분석법

송혜향 연세대학교 간호정책연구소 1992 간호학탐구 Vol.1 No.1