'Statistical significance' is one of the most widely misunderstood phrases in science, according to a 2013 Scientific American article.
It's a controversial topic. Probability values (p-values) have been used as a way to measure the significance of research studies since the 1920s, with thousands of researchers relying on them since. With this reliance, though, comes misunderstanding and, therefore, misuse.
This misunderstanding is what the latest episode of the How Researchers Changed the World podcast explores, in conversation with statistician Ron Wasserstein.
In particular, the podcast focuses on Ron's research into the misuse of p-values as a measure of statistical significance, which culminated in his 2016 paper: 'The American Statistical Association's statement on P-values: Context, Process and Purpose.'
Significance tests and p-values are widely used, according to Ron, to remove 'uncertainty' from scientific research. But uncertainty exists everywhere, and scientific research is no exception. For Ron, uncertainty in research should be embraced and accepted.
"Significance tests and dichotomised p-values ... have turned many researchers into what I'll call 'scientific snowbirds', trying to avoid dealing with uncertainty by escaping to a happier place." - Ron Wasserstein
With increasing use, and misuse, of p-values, statistics as a whole was starting to get a bad name. Some journals even banned the use of p-values and other statistical methods. So, Ron was tasked with leading the creation of a framework outlining how p-values should be used in research, which would be published as a statement by the American Statistical Association , a leading authority in the statistics world.
"We were challenged to do the ASA statement on p-values because of these attacks on statistics as a whole field of research." - Ron Wasserstein
It wasn't a simple task, but although the debate regarding p-values continues, the statement has had an impact on the research world beyond what Ron could ever have imagined...