Remove, rather than redefine, statistical significance

@article{Amrhein2017RemoveRT,
  title={Remove, rather than redefine, statistical significance},
  author={V. Amrhein and S. Greenland},
  journal={Nature Human Behaviour},
  year={2017},
  volume={2},
  pages={4}
}
  • V. Amrhein, S. Greenland
  • Published 2017
  • Psychology, Medicine
  • Nature Human Behaviour
  • To the Editor — Benjamin et al.1 propose to redefine statistical significance with a trichotomy: what was once ‘highly significant’ (P < 0.005) becomes ‘significant’, what was once significant (P < 0.05) becomes ‘suggestive’, and what was ‘nonsignificant’ (P > 0.05) remains nonsignificant. Trichotomization is better than dichotomization, and we agree that P values around 0.05 convey only limited evidence against the tested hypothesis (which is usually a ‘null’ hypothesis of no effect)2. We also… CONTINUE READING
    67 Citations
    Why "Redefining Statistical Significance" Will Not Improve Reproducibility and Could Make the Replication Crisis Worse
    • 4
    • PDF
    Abandon Statistical Significance
    • 253
    • PDF
    The p value wars (again)
    • U. Dirnagl
    • Psychology, Medicine
    • European Journal of Nuclear Medicine and Molecular Imaging
    • 2019
    • 2
    • PDF
    Abandon Statistical Signi fi cance
    • Blakeley B. McShanea, David Galb, Andrew Gelmanc, Christian Robertd, Jennifer L. Tackette
    • 2019
    • 4
    • PDF
    The Impact of P-hacking on “Redefine Statistical Significance”
    • 2
    • Highly Influenced
    Manipulating the Alpha Level Cannot Cure Significance Testing
    • 51
    • PDF
    Three Recommendations for Improving the Use of p-Values
    • 39

    References

    SHOWING 1-10 OF 10 REFERENCES
    Statistical tests, P values, confidence intervals, and power: a guide to misinterpretations
    • 850
    • PDF
    Invited Commentary: The Need for Cognitive Science in Methodology
    • S. Greenland
    • Computer Science, Medicine
    • American journal of epidemiology
    • 2017
    • 56
    • PDF
    The Long Way From α-Error Control to Validity Proper
    • K. Fiedler, F. Kutzner, J. Krueger
    • Psychology, Medicine
    • Perspectives on psychological science : a journal of the Association for Psychological Science
    • 2012
    • 127
    Power failure: why small sample size undermines the reliability of neuroscience
    • 3,965
    • PDF
    Competing interests The authors declare no competing interests. Nature HumaN BeHaviour | VOL 2 | JANUARY 2018 | 4 | www.nature.com/nathumbehav
    • Stat. 70,
    • 2016
    The statistical crisis in science
    • 378
    • PDF