In a post, Why The War Won't End, AS finds a study that concludes:
So long as a substantial minority (say 40%+) support the current policy (or at least oppose a rapid withdrawal) then Republicans can count on a public that is too divided on the issue to pose the certainty of electoral catastrophe.
Here's a supplemental theory. Assuming that no one knows whether continuing on is "worth it", the war won't end because of commitment bias.
If you split that minority above apart and find that it is overwhelmingly men, then that's more evidence that the theory is probably true.
Therefore, it's also a theory about how the 2008 elections may - may - go ...
