Hi Krzysztof, thanks for reading!
I think most people can follow the logic in that simple arithmetic in the full context:
Harmonic mean of 30 and 10 = ...
Arithmetic mean of reciprocals = 1/30 + 1/10 = 4/30 ÷ 2 = 4/60 = 1/15
Reciprocal of arithmetic mean = 1 ÷ 1/15 = 15/1 = 15
Nice post! But I’d push for a more radical overhaul of wordclouds’ arbitrary aesthetic mappings, which I call a “chatterplot”: https://towardsdatascience.com/rip-wordclouds-long-live-chatterplots-e76a76896098
Agree with everything you say. The height / weight vs income examples are always great ways of demonstrating robustness to outliers in medians vs means etc!
Such adjustments & transformations are not uncommon, I think. But as always, judgement should be used on a case-by-case basis. Depending on the distribution of data, magnitude of the range, etc, distortions could be introduced by such adjustments.
Thanks for reading & responding Asad!
The ideas & links you provide are interesting & novel ideas (to me), I’ll definitely have to spend some time digesting them!
Appreciate you linking or incorporating this into your course as well.
This article was actually already translated to Chinese here: http://mp.weixin.qq.com/s/QNvtw4sO-FPGtrpL1JEjOg
And unfortunately, though I asked them to link back to my post & credit me, they for some reason attributed it to a completely different person (altho with my job title):
Thanks for the catch Mladen!
You are, of course, correct.
Not sure how I convinced myself of this other than the fact that it is approximately the case in my trivial example.
Will make a correction & credit you asap.
Thanks again for reading & responding.