It is remarkable that in the interpretation of scientific evidence, be it as reviewer for a medical journal or as reader, clinicians appear to be the ones most impressed by p-values and statistical significance, or the absence thereof. They attach in their reviews and presentations more importance to it than epidemiologists and statisticians. In fact, the latter groups often actively discourage the use, and certainly the overinterpretation of significance testing, as reflected also in reporting guidelines such as STROBE, which see the presentation of p-values in a scientific article as an option, not an obligation.
