5 ms·

.

• daveFNbuck 10y ago
  They don't explain the same conclusion. One predicts that police departments with more vets will tend to be less militarized, the other predicts that police departments with more vets will tend to be more militarized. These could be easily distinguished by gathering data, and it would be very helpful to know which, if either, is true.
• coldtea 10y ago
  Their predictive power depends on the credibility of the premise (are actually vets less violent, non-vets more prone to playing solider-cop), which in both cases is taken for granted.
• nerdponx 10y ago
  That's not what predictive power means, and that's now how causal inference works.