English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

There is no doubt that significance tests have been abused in epidemiology and in other disciplines: statistically significant associations or differences have, in error, been automatically equated to substantively important ones, and statistically nonsignificant associations or differences have, in error, been automatically equated to ones that are zero. The proper inference from a statistically significant result is that a nonzero association or difference has been established; it is not necessarily strong, sizable or important, just different from zero. Likewise, the proper conclusion from a statistically nonsignificant result is that the data have failed to establish the reality of the effect under investigation. Only if the study had adequate power would the conclusion be valid that no practically important effect exists. Otherwise, the cautious "not proven" is as far as one ought to go. In part as a reaction to abuses due to misinterpretation, a movement is under way in epidemiology to cleanse its literature of significance tests and p-values. A movement that aimed at improving the interpretation of tests of significance would meet with the approbation of most of my statistical colleagues, but something more disturbing is involved. The author of one submission to a journal that publishes articles on epidemiology was asked by an editor that "all (emphasis mine) references to statistical hypothesis testing and statistical significance be removed from the paper." The author of another submission to that journal was told, "We are trying to discourage use of the concept of 'statistical significance,' which is, in our view, outmoded and potentially misleading. Thus, I would ask that you delete references to p-values and 'statistical significance.' " The author of a submission to another such journal was informed by one of its editors that "if you do not agree with my standards (concerning the inappropriateness of significance tests), you should feel free to argue to the point, or simply ignore what you may consider to be my misguided view by publishing elsewhere. As an editor, however, I can hardly be expected to accept papers that violate the scientific principle that I espouse." An insidious message is being sent to researchers in epidemiology that tests of significance are invalid and have no place in their research, and that their use in articles submitted for publication will result in rejection or a demand for radical revision. When read in the context of the movement I've just described, Dr. Alexander Walker's apparently innocuous comments about p-values' assume a different meaning. Dr. Walker cites the assessment of "higher order terms in statistical models of rich data sets" as an example of the usefulness of p-values (in fact, it is his only example). The obvious inference is that significance tests are possibly useful only in investigating tangential issues and less important features of the data, and are otherwise out of place. It is my opinion that, on the contrary, tests of significance have a valid role to play at each important step in the analysis of epidemiologic data. Data analysts who choose to employ them, and who interpret the resulting p-values in an

Significance tests have a role in epidemiologic research: reactions to A. M. Walker.