• 응용통계연구
• /
• 제22권6호
• /
• pp.1249-1263
• /
• 2009

표본수 결정에서 요구되는 검정력 함수는 연구가설에 상응하는 가장 적절한 검정방법에 의한 것이어야 한다. 의학연구의 논문에 자주 나타나는 순위자료 또는 범주형 빈도자료의 분석에는 비모수적 방법이 적절하며, 본 논문에서는 단변량 및 이변량 순위변수에 대한 윌콕슨-만-휘트니(Wilcoxon-Mann-Whitney; WMW) 검정법에 의한 표본수 결정방법을 제시한다. 단변량 순위변수의 윌콕슨 검정에서는 귀무가설과 대립가설 하의 분산을 이용한 표본수 공식이 귀무가설 하의 분산만 이용한 표본수 공식보다 정확하지만, 대립가설 하의 분산식에 나타나는 확률값이 일반적으로 알려져 있지 않으므로 이 확률값의 추정이 문제가 된다. 모의실험으로 두 방법에 대한 장, 단점을 알아본다. 효능과 안전성의 이변량 순위변수에서는 이변량 WMW 검정법에 의한 표본수 결정방법이 모수적 검정법에 의한 표본수 결정방법보다 더욱 바람직하다.

• 치과방사선
• /
• 제20권2호
• /
• pp.289-297
• /
• 1990

The purpose of this study was to assess the resolution and focal spot size of dental X-ray machines. Fifty dental X-ray machines were selected for measuring resolution and focal spot size. These machines were used in general dental clinics. The time of installation of the X-ray machine varies from 1 year to 10 years. The resolution of these machines was measured with the test pattern. The focal spot size of these machines was measured with the star test pattern. The following results were obtained: 1. The resolution of dental intraoral X-ray machines was not significantly changed in ten years. 2. The focal spot size of dental intraoral X-ray machines was not significantly increased in ten years. The statistical analysis between the mean focal spot size and nominal focal spot size was significant at the 0.05 level about the more than 3 years used machines.

• 한국안광학회지
• /
• 제9권1호
• /
• pp.29-34
• /
• 2004

본 연구는 타각적 검사장비의 개발을 위한 연구로 PC를 기반으로 Foucault Test (Knife edge Test)를 이용하여 굴절상태의 측정을 목표로 하였다. Knife edge Test를 이용하여 입사광선이 망막에서 반사된 출사광선의 이미지를 CCD 카메라를 사용하여 촬영하고, 그 이미지를 분석하여 반사상의 지름을 측정하여 동공의 크기를 측정하였다. 또한 반사 이미지의 방향으로 굴절상태를 판정하고 반사 이미지의 높이를 측정하여 굴절이상도를 확인하였다. 구면수차는 반사상과 반대방향에 나타나는 부반사 이미지의 명도를 비교하여 측정하였고, 그 크기를 통해 수차값을 계산할 수 있었다.

• 한국지하수토양환경학회지:지하수토양환경
• /
• 제15권2호
• /
• pp.1-9
• /
• 2010

Estimations of porosity and bulk density, particle size analysis of soli samples, tracer test and slug test were performed in a petroleum contaminated area of Kangwon for understanding characteristics of the aquifer. Porosities of the samples were estimated 0.158~0.257, and bulk densities were estimated as $1.73\sim2.10\;g/cm^3$. Majority proportion of the soil samples was 0.5~1.0 mm size. In the soil texture triangle, all samples were distributed at sand area. Uniformity coefficients were estimated as 7.71~10.39, and thus all samples were poorly-sorted. In the tracer test, Darcy velocity was estimated to $4.8\times10^{-6}$ cm/day, effective porosity was 0.175, and longitudinal dispersivity was 0.1 m. According to the slug test, hydraulic conductivities of the test wells were estimated as $2.243\times10^{-2}\sim1.634\times10^{-2}$ cm/sec. These hydrogeologic parameters can be used for efficient remediation design of the petroleum contaminated area.

• International Journal of Railway
• /
• 제6권3호
• /
• pp.112-119
• /
• 2013

A mid-size RC test apparatus (MRCA) equipped with a program is developed that can test samples up to D=10 cm diameter and H=20 cm height which are larger than usual samples used in practice. Using the developed RC test apparatus, two types of crushed trackbed foundation materials were tested in order to get the shear modulus reduction curves of the materials with changing of shear strain levels. For comparison purpose, large repetitive triaxial compression tests (LRT) with samples of height H=60cm and diameter D=30 cm were performed also. Resilient modulus obtained from the LRT was converted to shear modulus by considering elastic theory and strain level conversion and were compared to shear modulus values from the MRCA. It is found from this study that the MRCA can be used to test the trackbed foundation materials properly. It is found also that strain levels of $E_{v2}$ mostly used in the field should be verified considering the shear modulus reduction curves and proper values of $E_{v2}$ of trackbed foundation must be used considering the strain level verified.

• 대한간호학회지
• /
• 제45권5호
• /
• pp.641-649
• /
• 2015

Purpose: The purpose of this study was to introduce the main concepts of statistical testing and effect size and to provide researchers in nursing science with guidance on how to calculate the effect size for the statistical analysis methods mainly used in nursing. Methods: For t-test, analysis of variance, correlation analysis, regression analysis which are used frequently in nursing research, the generally accepted definitions of the effect size were explained. Results: Some formulae for calculating the effect size are described with several examples in nursing research. Furthermore, the authors present the required minimum sample size for each example utilizing G*Power 3 software that is the most widely used program for calculating sample size. Conclusion: It is noted that statistical significance testing and effect size measurement serve different purposes, and the reliance on only one side may be misleading. Some practical guidelines are recommended for combining statistical significance testing and effect size measure in order to make more balanced decisions in quantitative analyses.

• 대한치과의사협회지
• /
• 제52권9호
• /
• pp.558-569
• /
• 2014

Sample size determination is critical, but not easy to do. Sample size defined as the number of observations in a sample to be studied should be big enough to have a high likelihood of detecting a true difference between groups. Practical procedure for determining sample size, using $G^*$power and previous dental articles, was shown in this study. Examples involving independent t-test, paired t-test, one-way analysis of variance(ANOVA), and one-way repeated-measures(RM) ANOVA were used. The purpose of this study is to enable researchers with non-statistical backgrounds to use in practice freely available statistical software G*power to determine sample size and power.

• KIEE International Transactions on Electrophysics and Applications
• /
• 제4C권2호
• /
• pp.68-72
• /
• 2004

The performance of copper-tungsten for electrodes used in an ultra high voltage interruption system was evaluated by means of an interruption test, which requires a large-scale apparatus and high cost. In this study, prior to the interruption test, the characteristics of a Cu-W electrode were estimated through the DC arc test, which is a simple, low cost procedure. The DC arc characteristics of a 20wt%Cu-80wt%W electrode were investigated with the change of tungsten powder size distribution and the addition of nickel. In specimens containing a high volume fraction of large sized tungsten particles, the relative density and hardness of sintered Cu-W electrodes increased while the electrical conductivity and the DC arc resistance decreased. Furthermore, the relative density became enhanced with the increase of the amount of nickel while the hardness and electrical conductivity diminished and the DC arc resistance worsened.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2005년도 춘계학술대회 논문집
• /
• pp.615-620
• /
• 2005

The purpose of this study was to provide a criterion of ballast aggregate size distribution for paved track. The parameters considered in this study were different types of particle size distribution. Then the analysis for analyzing compressive and flexural strength, physical characteristics of aggregate, and the ballast box test were performed. In the test result of the physical characteristic performed by the ballast box test, the bearing capacity was measured lower than the expected value because of its boundary conditions. Among four types of ballast aggregate, type B was selected as one of best candidate distribution because of its bearing capacity, strength, development and economics.

• Journal of the Korean Data and Information Science Society
• /
• 제24권6호
• /
• pp.1349-1357
• /
• 2013

임상시험에서 적당한 크기의 표본 수 결정은 통계적으로 유의한 연구결과의 도출과 연구수행의 효율적인 비용을 산출하기 위해서 중요한 사항 중의 하나이다. 기존의 비열등성 시험에서 표본 수 계산방법에는 t 검정법을 이용한 모수적 방법이 있고, Wilcoxon 순위합 검정을 이용하여 Wang 등 (2003)이 제안한 표본 수 계산방법을 Kim과 Kim (2007)이 비열등성 시험에 확장시켜 적용하여 제시한 비모수적 방법이 있다. 본 논문에서는 Orban과 Wolfe (1982)가 제안한 선형위치통계량의 검정법에 Kim (1994)이 계산한 검정력의 결과를 이용한 표본 수 계산 방법을 제안하고, 그에 따른 표본 수 계산결과를 기존에 제시된 표본 수 계산 결과와 비교하였다. 그 결과 기존의 방법들보다 본 논문에서 제안하는 방법으로 계산한 경우의 표본 수가 가장 작게 나왔다. 따라서 모집단에 대해서 구체적인 분포함수를 가정하기 힘든 경우 모수적 방법을 이용하게 되면 검정력이 떨어지거나 유의수준을 제어하지 못하는 문제점을 보완하고, 모수적 방법에 비해 표본 수가 크게 나와 시간이나 비용 면에서 효율적이지 않았던 Wilcoxon 순위합 검정을 이용한 방법의 단점을 보완할 수 있는 방법으로 본 논문에서 제시한 위치 (placement)를 이용한 표본 수 계산이 이용될 수 있다.