The fickle P value generates irreproducible results

被引：521

作者：

Halsey, Lewis G. ^{[1
]}

Curran-Everett, Douglas ^{[2
,3
]}

Vowler, Sarah L. ^{[4
]}

Drummond, Gordon B. ^{[5
]}

机构：

[1] Univ Roehampton, Dept Life Sci, London, England

[2] Natl Jewish Hlth, Div Biostatist & Bioinformat, Denver, CO USA

[3] Univ Colorado, Dept Biostat & Informat, Colorado Sch Publ Hlth, Denver, CO 80202 USA

[4] Univ Cambridge, Canc Res UK Cambridge Inst, Cambridge, England

[5] Univ Edinburgh, Edinburgh, Midlothian, Scotland

来源：

NATURE METHODS | 2015年 / 12卷 / 03期

关键词：

STATISTICAL SIGNIFICANCE; CONFIDENCE-INTERVALS; POWER; SIZE; REPLICATION; POINTS; WORTH;

D O I：

10.1038/nmeth.3288

中图分类号：

Q5 [生物化学];

学科分类号：

071010 ; 081704 ;

摘要：

The reliability and reproducibility of science are under scrutiny. However, a major cause of this lack of repeatability is not being considered: the wide sample-to-sample variability in the P value. We explain why P is fickle to discourage the ill-informed practice of interpreting analyses based predominantly on this statistic.

引用

页码：179 / 185

页数：7

共 42 条

[1] Journals unite for reproducibility [J].