Is it significant? Is it relevant?

As outlined in the editorial by Stecker et al. (2017) in the October issue of Clinical Neurophysiology, there is little difference between the evidence for a p-value slightly less than 0.05 and another slightly larger than 0.05. However, based on a stringent cut-off of α = 0.05, a hypothesis for which p = 0.04 is conventionally accepted as true while, alternatively, with a p = 0.06 is rejected since it is "not significant". This is a common and uncomfortable "In-or-Out" scenario, but also every time a test of statistical significance is employed, researchers' interpretation of the study findings would greatly benefit from the valuable data conveyed by confidence intervals.

http://ift.tt/2DYstz1