Kühnast Corinna, Neuhäuser Markus
Institute for Medical Informatics, Biometry and Epidemiology, University of Duisburg-Essen, Essen, Germany.
Ger Med Sci. 2008 Apr 7;6:Doc02.
Although non-normal data are widespread in biomedical research, parametric tests unnecessarily predominate in statistical analyses.
We surveyed five biomedical journals and - for all studies which contain at least the unpaired t-test or the non-parametric Wilcoxon-Mann-Whitney test - investigated the relationship between the choice of a statistical test and other variables such as type of journal, sample size, randomization, sponsoring etc.
The non-parametric Wilcoxon-Mann-Whitney was used in 30% of the studies. In a multivariable logistic regression the type of journal, the test object, the scale of measurement and the statistical software were significant. The non-parametric test was more common in case of non-continuous data, in high-impact journals, in studies in humans, and when the statistical software is specified, in particular when SPSS was used.
尽管非正态数据在生物医学研究中广泛存在，但参数检验在统计分析中仍不必要地占据主导地位。
我们调查了五本生物医学期刊，并针对所有至少包含未配对t检验或非参数Wilcoxon-Mann-Whitney检验的研究，调查了统计检验的选择与其他变量（如期刊类型、样本量、随机化、资助等）之间的关系。
30%的研究使用了非参数Wilcoxon-Mann-Whitney检验。在多变量逻辑回归中，期刊类型、检验对象、测量尺度和统计软件具有显著性。非参数检验在非连续数据、高影响力期刊、人体研究以及指定统计软件（特别是使用SPSS时）的情况下更为常见。
