The fickle P value generates irreproducible results

被引：502
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
收藏
页码：179 / 185
页数：7
相关论文
共 42 条
  • [1] Journals unite for reproducibility
    不详
    [J]. NATURE, 2014, 515 (7525) : 7 - 7
  • [2] [Anonymous], 2013, ECONOMIST
  • [3] Replication effort provokes praise-and 'bullying' charges
    Bohannon, John
    [J]. SCIENCE, 2014, 344 (6186) : 788 - 789
  • [4] P-Value Precision and Reproducibility
    Boos, Dennis D.
    Stefanski, Leonard A.
    [J]. AMERICAN STATISTICIAN, 2011, 65 (04) : 213 - 221
  • [5] Power failure: why small sample size undermines the reliability of neuroscience
    Button, Katherine S.
    Ioannidis, John P. A.
    Mokrysz, Claire
    Nosek, Brian A.
    Flint, Jonathan
    Robinson, Emma S. J.
    Munafo, Marcus R.
    [J]. NATURE REVIEWS NEUROSCIENCE, 2013, 14 (05) : 365 - 376
  • [6] THE EARTH IS ROUND (P-LESS-THAN.05)
    COHEN, J
    [J]. AMERICAN PSYCHOLOGIST, 1994, 49 (12) : 997 - 1003
  • [7] The New Statistics: Why and How
    Cumming, Geoff
    [J]. PSYCHOLOGICAL SCIENCE, 2014, 25 (01) : 7 - 29
  • [8] Replication and p Intervals p Values Predict the Future Only Vaguely, but Confidence Intervals Do Much Better
    Cumming, Geoff
    [J]. PERSPECTIVES ON PSYCHOLOGICAL SCIENCE, 2008, 3 (04) : 286 - 300
  • [9] Explorations in statistics: confidence intervals
    Curran-Everett, Douglas
    [J]. ADVANCES IN PHYSIOLOGY EDUCATION, 2009, 33 (02) : 87 - 90
  • [10] Show the data, don't conceal them
    Drummond, G. B.
    Vowler, S. L.
    [J]. JOURNAL OF PHYSIOLOGY-LONDON, 2011, 589 (08): : 1861 - 1863
12345