create new tag
, view all tags, tagging instructions

Use of “Significant” Alone in Publications

Do not use the word “significant” alone in scientific publications that include statistical analysis. This term is ambiguous, with the following two distinct possible meanings:

  • Use “statistically significant” if the intention is to note that the P-value was <0.05.
  • Use “important” or “substantial” if the intention is to note that the observed effect or association is large enough to be important (clinically or scientifically).
Using the term “significant” alone promotes conflation of these two different meanings, which is a serious problem in the interpretation of clinical and translational research. A result can have P<0.05 while still being trivially small, and more commonly, an estimated effect can be large enough to be important even if P>0.05.

The same considerations apply to using “significantly” alone.
Topic revision: r2 - 05 Oct 2014 - 15:35:54 - PeterBacchetti

Copyright & by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding CTSPedia? Send feedback