2 ms·
> Many scientists receive only cursory training in statistics, This seems to be true (at least I have many younger colleagues who are woefully ill informed ab
by kwhitefoot 10y ago
> Many scientists receive only cursory training in statistics,
This seems to be true (at least I have many younger colleagues who are woefully ill informed about even basic stats.) but it's very strange.
Most of what I learned of statistics, including tests for significance and regression, I learned in senior high school at the age of 17.
Learning not to search of correlations like this is pretty much equivalent to ensuring that you do not over fit your data set which is surely such conventional wisdom that no one getting a scientific education in the last sixty or seventy years should have missed it.
I'm aware that there is a lot of highly sophisticated statistical analysis that I don't know about but things like p value are not among them.