Due to the issues and problems with the interpretation of ‘statistical significance‘, especially by the general public, it has been suggested that ‘statistically detectable‘ might be a preferable term.
Join my campaign to stop scientists using the term “statistically significant” and replace it with “statistically detectable”, so people don’t assume it always means *actually* significant https://t.co/lraPnJc4Cy
— Tom Chivers (@TomChivers) December 23, 2020
Tom Chivers did tweet about this, but later said that:
This was kind of a joke tweet, but I do think statistical significance is terribly badly named and “detectable” would at least be LESS misleading. But I guess ideally we’d stop using an arbitrary cutoff altogether
— Tom Chivers (@TomChivers) January 3, 2021
It was more of a point that he was trying to make about the public’s understanding of the word ‘significance’.
Related Topics:
Statistical Significance
Page last updated:
Comments are closed.