2 ms·
No, statistics aren't useless, but its usefulness cuts both ways: if you can add one or two relevant variables and almost entirely remove the observation, then
by mseebach 9y ago
No, statistics aren't useless, but its usefulness cuts both ways: if you can add one or two relevant variables and almost entirely remove the observation, then statistics tells you that the observation was only there due to omitted-variable bias.
- dwaltrip 9y agoSure, that's fine. That's part of using statistics to the best of our ability.