Quote:
Keith Larman wrote:
I'd like to second the thanks and also suggest I'd love to see a spreadsheet myself.
And as a niggling aside from a statistics geek, I'd be a little careful using the word correlation here. Correlation has a very well defined meaning and if I might suggest it would probably be better to say there is a very strong correspondence. Just a suggestion from the cheap seats.
Regardless, great article. And thank you again.
|
Ditto, and thanks. As a geek, but only an end-user of statistics, from experience it pays to note that making an argument from similarity is fraught with danger. Human DNA shows only a 1.6 % genetic divergence from Bonobo DNA. What is different is vastly more important in determining the result than what is similar. A difference of 18 % is an order of magnitude larger, roughly. a fifth whittled out. A study might be of value in focusing on what O Sensei left out, and more particularly -- WHY? in each case. That may be of more use than piling up more evidence of wholly expected similarities in cousin arts. Although likely you have to do the one to do the other