Online Program

Moving to a World Beyond P<0.05 (304162)

Keywords: p-value, statistical significance, inference, estimation, confidence intervals

For nearly a hundred years the concept of “statistical significance” has been fundamental to statistics and to science. And for nearly that long, it has been controversial and misused as well. In a 2019 editorial in a special issue of The American Statistician, we argue that it is time to stop using statistical significance as a means of determining the worth of scientific. I'll say why this change is so hard to make, but why it is essential to do so, and suggest alternate strategies for thinking about and presenting results.