• Communications for Statistical Applications and Methods
• /
• 제6권1호
• /
• pp.159-168
• /
• 1999

We introduce a new conditional signed-rank test for the tree alternatives comparing several treatments with a control in the randomized block design. We demonstrate its performance by comparing with 3 classes of signed-rank tests proposed by Park et al.(1991) in some general situations. In most cases the proposed procedure is simpler to compute and has better power than others.

• 대한수학교육학회지:학교수학
• /
• 제3권1호
• /
• pp.101-108
• /
• 2001

Classroom exercise using the randomized response technique may be used to summarize a high school unit of instruction on probability. In this paper, we show that the derivation of the formula for this technique illustrates basic concepts in probability, such events being mutually disjoint and independent.

• 한국통신학회:학술대회논문집
• /
• 한국통신학회 2004년도 추계종합학술대회 논문 초록집
• /
• pp.326-326
• /
• 2004

• 한국콘텐츠학회논문지
• /
• 제17권10호
• /
• pp.65-71
• /
• 2017

통계조사 또는 실태조사는 '어떤 조사자가?', '어떤 목적으로?', 그리고 '어떻게?' 작성되었는가에 따라 왜곡의 가능성이 있는 것이 사실이다. 심지어는 통계결과는 '거짓말' 더 나아가서 범죄 또는 비행 등에 관한 통계는 '새빨간 거짓말'로 불리기도 한다. 범죄나 비행에 관한 통계를 신뢰하지 못하는 이유는 여러 가지를 들 수 있겠지만 그 중 대표적인 원인중의 하나가 숨은 범죄(Hidden Crime) 또는 형사사법기관에 보고되지 않는 범죄(Unreported Crime)가 존재한다는 것이다. 이러한 숨은 범죄문제를 보완하기 위해 피해자조사 또는 자기보고식 조사 등의 방법이 사용되고 있으나, 이 또한 범죄의 유형에 따라 과소보고 또는 과대보고의 문제가 있다. 범죄와 비행 그리고 일탈행동에 대한 조사는 매우 민감한 사항들이기 때문에, 조사대상자들은 심리적 부담을 가지게 된다. 이처럼 조사대상자들의 경험을 밝히는 것이 부담스러운 민감한 내용에 대해 진실한 답변을 유도할 수 있는 방법으로 통계학 분야에서 확률화응답모형(randomized response model)이 개발되어 사용되어왔다. 이 기법은 피해자조사 또는 자기보고식 조사의 문제점을 해결할 수 있는 매우 유용한 방법임에도 불구하고, 우리나라에서 범죄학 분야의 조사에서 사용된 경우는 매우 적다. 따라서 이 연구에서는 범죄학 분야연구에 있어 확률화응답모형의 적용가능성을 타진하기 위하여, 확률화응답모형을 활용하여 대학생들을 대상으로 성매매에 대한 내용을 실제로 측정해 보고, 확률화응답모형의 유용성을 확인해 보았다.

• 한방재활의학과학회지
• /
• 제20권4호
• /
• pp.91-103
• /
• 2010

Objectives : The purpose of this study is to perform quantity and quality assessment of randomized controlled trials(RCTs) in the journal of oriental rehabilitation medicine(JORM). Methods : Upon extracting randomized controlled trial from all the articles published in the JORM from 1991 to 2009, assessments were made on the adequacy of the consolidated standards of reporting trials(CONSORT) checklist. The assessment were performed independently by two researchers, and adjustment of the differences were done by discussions. Results : Among the 824 articles from 1981 to 1999, 33 RCTs were selected. 4 RCTs were published from 1991 to 2002, and 29 RCTs were published from 2003 to 2009. Adequacy of the CONSORT checklist was 29.4%(27.3% from 1999 to 2005, 31.6% from 2006 to 2009). Conclusions : RCTs of the JORM consistently increased in quantity and quality. But researchers should make an effort to reporting more accurate.

• 응용통계연구
• /
• 제27권1호
• /
• pp.61-70
• /
• 2014

랜덤화 블록 계획법은 동질적인 실험단위를 묶어 여러 개의 블록으로 나눈 후, 각 블록의 실험단위에 처리를 적용하는 방법이다. 랜덤화 블록 계획법에서 Jonkckheere (1964)와 Terpstra (1952), Page (1963) 그리고 Hollander (1967) 등이 순서대립가설의 다양한 방법을 제안하였다. 특히, 블록 내 순위합의 가중치를 주는 방법으로 Page (1963) 검정법이 있다. 본 논문에서는 Page 검정을 확장하여 순서대립가설에 새로운 비모수적 방법론을 제안하였다. 또한, 몬테카를로 모의시험 연구를 통해 제안된 방법과 이전의 방법들의 검정력을 비교하였다.

• 부모자녀건강학회지
• /
• 제21권1호
• /
• pp.50-60
• /
• 2018

Purpose: This study is a systematic literature review and meta-analysis study conducted to identify the effects of breastfeeding intervention program on premature infants by integrating and analyzing the results of randomized controlled trials and non-randomized trials. Methods: The literature review process was based on the PRISMA (Preference Reporting Items for Systematic Reviews and Meta-Analyzes) guideline. Data retrieval and collection were conducted from May 25 to May 30, 2018, and the articles analyzed were all domestic papers retrieved from the database. Two reviewers independently select the studies and assessed methodological risk of bias of studies using the Cochrane criteria. The topics of breastfeeding interventions were analyzed using descriptive analysis and the effects of intervention were meta-analyzed using the R program. Results: Finally, eight papers were included in the systematic review and meta-analysis. The breastfeeding intervention program for premature infants showed a significant increase in the self-efficacy and the amount of pumping and baby's hight. Conclusion: This study has some limitations due to the few randomized controlled trials and non-randomized trials comparing breastfeeding for premature babies in Korea. Therefore, it needs to be integrated with the research conducted in other countries.

• 응용통계연구
• /
• 제29권7호
• /
• pp.1411-1419
• /
• 2016

랜덤화 블록 계획법을 검정하는 비모수적 방법에는 일반적인 대립가설에서 Friedman (1937)이 제안한 검정법이 있다. 이 방법은 처리 효과의 차이를 알아보기 위한 검정법으로 블록 내 순위를 사용해 검정함으로써 블록 간 정보의 손실이 있을 수 있다. 본 논문에서는 Hodges와 Lehmann (1962)이 제안한 정렬방법을 이용하여 블록 간 정보 손실을 줄이고, Jo와 Kim (2013)이 제안한 랜덤화 블록 계획법의 결합위치 방법을 확장하여 결합위치에 점수함수를 적용한 새로운 비모수적 방법을 제시하였다. 또한 Monte carlo simulation을 통하여 기존의 검정 방법과 제안한 검정법의 검정력을 비교하였다.

• 한방안이비인후피부과학회지
• /
• 제21권2호
• /
• pp.94-101
• /
• 2008

Objectives : To discuss the types of control groups in randomized controlled trials (RCTs) of herbal medicine, and to provide suggestions for improving the design of control group in future clinical trials. Methods : We reviewed the 8 articles about clinical trial design of Chinese herbal preparation which were published from 2005 through 2008. We selected those articles from CNKI(中國知識基礎施設工程(http://www.cnki.net)). Results : It is necessary to have control group in randomized controlled trials(RCTs) of Korean herbal preparation. But there are problems in the selection of appropriate control group drug. This paper lists several problems about the choice of control drug and puts forward some proposals and countermeasures. There are problems such as ethics and manufacturing matching placebo and positive control herbal drug. Conclusion : To improve the quality of control group design, we introduce standard drug plus placebo drug method and add-on research for placebo control group design, double dummy technique, using negative control drug or composite control drug for active control group design.

• 대한간호학회지
• /
• 제44권4호
• /
• pp.446-457
• /
• 2014

Purpose: This study was done to evaluate the effects of psychosocial interventions on cortisol and immune response in adult patients with cancer. Methods: MEDLINE via PubMed, Cochrane Library CENTRAL, EMBASE, CINAHL and domestic electronic databases were searched. Twenty controlled trials (11 randomized and 9 non-randomized trials) met the inclusion criteria with a total of 862 participants. Methodological quality was assessed using the Cochrane's Risk of Bias for randomized studies and the Risk of Bias Assessment tool for non randomized studies. Data were analyzed using the RevMan 5.2.11 program of Cochrane library. Results: Overall, study quality was moderate to high. The weighted average effect size across studies was -0.32 (95% CI [-0.56, -0.07], p=.010, $I^2 $=45%) for cortisol concentration, -0.62 (95%CI [-0.96,-0.29], p<.001, $I^2 $=0%) for T lymphocyte (CD3) and -0.45 (95%CI [-0.74, -0.16], p=.003, $I^2 $=0%) for Th lymphocyte (CD4) numbers. Psychosocial interventions were not effective for Tc lymphocyte (CD4), NK cell, monocyte, and cytokine response. Conclusion: Although these results provide only small evidence of successful immune modulation, they support the conclusion that psychosocial interventions can assist cancer patients in reducing emotional distress and improving immune response.