Department of Statistics and Operational Research, School of Mathematics and Physics, The Queen's University of Belfast, Belfast, Northern Ireland, UK.
Eye (Lond). 2009 Jul;23(7):1589-97. doi: 10.1038/eye.2009.105. Epub 2009 May 15.
The aim of this paper is to provide an overview of sample size estimations for the most frequent type of group studies that result in continuous, binary and ordered categorical outcomes.
The theory behind power and sample size calculations is explained using the basic probability concepts that underpin the most frequently used statistical significance tests.
Simple formulae and tables are presented for the estimation of sample sizes necessary for efficient and effective clinical and epidemiological trials. These may be used without recourse to sophisticated and complex computer software packages. Mathematical complexity is kept to a minimum. Examples and applications from the vision sciences are specifically highlighted.
The paper highlights, with practical examples, the concepts and computations necessary to make sample size estimations accessible to all eye professionals involved in research, diagnostic and statutory work.
本文旨在提供最常见的群组研究类型的样本量估计概述，这些研究类型会产生连续、二分类和有序分类结局。
使用基础概率概念解释了效能和样本量计算的理论，这些概念是最常用的统计显著性检验的基础。
本文提出了用于高效和有效临床及流行病学试验的样本量估计的简单公式和表格。无需使用复杂和复杂的计算机软件包即可使用这些公式和表格。数学复杂性保持在最低水平。特别强调了来自视觉科学的示例和应用。
本文通过实际示例强调了使所有参与研究、诊断和法定工作的眼科专业人员都能够进行样本量估计所需的概念和计算。
