中国临床药理学杂志2012，Vol.28Issue(10)：776-778,3.
分析临床研究中等级资料统计存在的问题与对策
Analysis problems with solutions for ranked data in clinical trial
摘要
Abstract
Objective To discuss the statistical analysis method of ranked data in clinical research. Methods Been analysised with simulated data based on clinical research purposes and statistical analysis principle. Results The statistical analysis results may be different if Wilcox-on rank sum test and Chi - square test are used respectively of some ranked data. Of some repeated ranked data, the baseline before treatment was neglected if only analysis the data after treatment with Ridit test or Wilcoxon rank sum test. Conclusion It is recommended that based on the purpose of a research project, the statistical analysis approach: applicable approach for ranked data or based on Pearson's chi - square test. If ranked data consists of pre - and post - treatment measurement information, statistical analysis of post - treatment data with Ridit analysis or Wilcoxon rank - sum test may be inaccurate. It is advised to adopt an applicable statistical approach to analyze both pre - and post — treatment measurement information.关键词
等级资料/统计分析Key words
ranked data/ statistical analysis分类
医药卫生引用本文复制引用
毕京峰,陈素红,屠舒,朱真珍,魏振满..分析临床研究中等级资料统计存在的问题与对策[J].中国临床药理学杂志,2012,28(10):776-778,3.基金项目
国家十一五科技重大专项课题基金资助项目(2009ZX10005-017) （2009ZX10005-017）
中国人民解放军第三○二医院院内课题基金资助项目 （）
