4 ms·
Not OP, but statistical significance has to do with how _sure_ you are that there is an effect. It doesn't really say anything about how _large_ the effect is.
by downerending 7y ago
Not OP, but statistical significance has to do with how _sure_ you are that there is an effect. It doesn't really say anything about how _large_ the effect is.
It's pretty obvious that gender can't be a top predictor of memory, intelligence, etc. The reason is that even if there is a definite difference where males are better than females, it'd be a modest delta between two bell curves. That would leave huge numbers of pairings in which a female had better memory than a male, etc.
- thaumasiotes 7y agoStipulating most of your comment, gender can still easily be a top predictor of memory / intelligence / etc., because "top" is an ordinal concept. That just says it has more of an effect than other predictors. But see also https://sci-hub.tw/https://amstat.tandfonline.com/doi/full/10.1080/2330443X.2015.1129918#.Xdndn5NKiL5 https://sci-hub.tw/https://amstat.tandfonline.com/doi/full/1... , finding that the top predictor (among those investigated) of a police officer shooting a civilian is "the officer is black", narrowly beating out "the officer is rapidly accumulating negative marks in his file". Are there a large number of pairings in which a white officer shoots someone and a black officer doesn't? Sure, but the effect of race is still enormous in this case. ("Black officers were 3.3 times more likely to shoot than other officers.") I would expect the effect of sex to dwarf the effect of race, but the study is limited to police officers that exist, of which more than 92% in the sample are male. So probably there's some restriction of range.
- downerending 7y agoSure. I had in mind something like "average parent IQ" as a top predictor. Next to that, probably almost everything else pales.