As a rule of thumb, you should prefer rates (“X commit Y crime 3x more often than normal”) and proportions (“1 in 10,000 X commit Y crime”). Be wary of quantities (“100 X have committed Y crime this year”) and examples (“look at all these stories of X doing Y”).
Does anyone have tips on Distinguishing the fallacy from reality.
For example, if I wanted to know if the recent sexual problems in the acting community are over represented in that community, how would I find out?
How do I find the number of people in that industry, the number of events (where guilt is known) and the number outside of hollywood.
I’m sure methods must exists for getting this information, but I don’t know them.
As a rule of thumb, you should prefer rates (“X commit Y crime 3x more often than normal”) and proportions (“1 in 10,000 X commit Y crime”). Be wary of quantities (“100 X have committed Y crime this year”) and examples (“look at all these stories of X doing Y”).
See also: https://www.gwern.net/Littlewood