Abstract
FREQUENTLY in scientific papers one finds the terms ‘insignificant’, ‘significant’ and ‘highly significant’ used in discussing the difference between control and experimental, with their ranges usually established by P = 0.05 and P = 0.01. Statistics text-books provide little or no guidance as to their proper usage because statisticians apparently disapprove of the convention. Yet the universality of the convention attests to the need for clearly defined, commonly accepted terms which properly weight the results of the experiment. The present convention is commonly accepted among non-statisticians, but, as will be shown here, is not clearly defined. Therefore an improved convention is proposed for the consideration of those who would benefit from it—the non-statisticians in research. Perhaps by improving the convention, it can be made more palatable to statisticians, as well as more useful to non-statisticians.
Similar content being viewed by others
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
MILLER, D. ‘Significant’ and ‘Highly Significant’. Nature 210, 1190 (1966). https://doi.org/10.1038/2101190a0
Issue Date:
DOI: https://doi.org/10.1038/2101190a0
This article is cited by
-
“Significant” and “Highly Significant”
Nature (1966)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.