Evans Richard B, O'Connor Annette
Veterinary Diagnostic and Production Animal Medicine, Iowa State University College of Veterinary Medicine, Ames, IA 50011, USA.
Vet Clin North Am Small Anim Pract. 2007 May;37(3):477-86. doi: 10.1016/j.cvsm.2007.01.006.
Evidence-based veterinary medicine relies critically on the scientific validity of research. A component of validity is the statistical design and subsequent analysis of data collected during the study. Correct statistical design reduces bias and improves generalizability, and correct analysis leads to appropriate inferences. Inference is the art and science of making correct decisions based on data. Because veterinarians are responsible for the medical care of their patents, it is also their responsibility to understand inferences about treatments presented in papers. This article is designed to assist veterinarians with the interpretation and understanding of statistics presented in papers.
循证兽医学严重依赖于研究的科学有效性。有效性的一个组成部分是研究期间收集的数据的统计设计及后续分析。正确的统计设计可减少偏差并提高普遍性,而正确的分析会得出恰当的推论。推论是基于数据做出正确决策的艺术与科学。由于兽医负责其患者的医疗护理,因此他们也有责任理解论文中关于治疗方法的推论。本文旨在帮助兽医解读和理解论文中呈现的统计数据。