[ExI] Why so much published 'science' is wrong.
anders at aleph.se
Sun Jul 12 10:01:06 UTC 2015
Eyeballing data is a good method, although there are many hidden gotchas. I am somewhat surprised how rarely people just look at the raw data and plot it: a smart colleague of mine struggled mightily doing something useful with a dataset, until I just plotted it and immediately saw that it contained -999 entries to mark missing data. No wonder his statistical method was useless when that was treated as real data (that should be positive).
William is totally right about the docility of numbers. It is when they form a crowd of points they start finding a voice.
Anders Sandberg, Future of Humanity Institute Philosophy Faculty of Oxford University
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the extropy-chat