The 'Bradley' effect almost certainly didn't happen in NH

[dsc]

I was one of the people who initially worried it might have been that which lead the polls to be so wrong. I now know, it wasn't. Why? Because the polls did correctly gage his level of support, what they got wrong was her level of support. The Bradley effect would have shown up in his support being too low, not hers being too high. I don't know for sure what did happen, but this is powerful evidence against the Bradley effect.

[Ian David]

1. I think it's because Repubublican operatives gave free pot to College Democrats.

But we'll get 'em back by giving free crystal meth to College Republicans.

[no name no slogan]

3. Oooooh... methed-up College Republicans

remind me to bring my video camera

[Ian David]

5. Methed-up Republicans is redundant. n/t

[no name no slogan]

2. Polls describe the past, they don't predict the future

All polls do is provide a snapshot at a particular point in time IN THE PAST. They MAY possibly predict future outcomes, but they are by no means foolproof.

[dsc]

4. of course they aren't

NH showed that. But the why matters. This time it appears they underestimated female support for Clinton, not overestimated white support for Obama.

[no name no slogan]

7. Don't quote me, but didn't polls taken closer to the election

show the numbers to be a lot closer than they had been even a few days before? Plus, it is possible for some last-minute shift to happen on election day, especially if there's a lot of undecideds.

[hayu_lol]

8. Just offhand, I'd say that the pollsters just plain didn't have the time...

to work out the last minute polls. Wasn't much time to finish with Iowa, fly to NH and then get their polls right. 2 out of 3 ain't bad...even if they were 'slightly' off.

Wonder what odds the Las Vegas bookies are offering?

[Pirate Smile]

6. I agree and that is good news.

Info to this effect was posted the next day which is why I never dove into the threads about that issue.