If you are like me you love a good parody and hence this new Google product.
Matthew Hankins carried out an analysis of the wording researchers use when their statistical tests yield non-significant results (p > 0.05). Hilarious!
” The following list is culled from peer-reviewed journal articles in which (a) the authors set themselves the threshold of 0.05 for significance, (b) failed to achieve that threshold value for p and (c) described it in such a way as to make it seem more interesting.
here is the list in full:
(barely) not statistically significant (p=0.052)
a barely detectable statistically significant difference (p=0.073)
a borderline significant trend (p=0.09)
a certain trend toward significance (p=0.08)
a clear tendency to significance (p=0.052)
a clear trend (p<0.09)
a clear, strong trend (p=0.09)
[Matthew Hankins, via Justin Wolfers]
You know the saying about hammers and nails. I do believe being a domain expert not only means knowing how to use what tool to what end, but also knowing when the tools of your domain are not the right ones for a job:
(Under the Finds title I collect links that I have found interesting, relevant or inspiring for some reason, but that don’t squarely fit into the core themes of this blog. It’s only an experiment as of now.)