Veep Veep

by Andrew Gelman on September 1, 2008 · 4 comments

I’ve said it before, and I’ll say it again: there’s no evidence I know of that VP nominees have much effect on presidential vote.

4 comments

{ 4 comments }

Shag from Brookline September 2, 2008 at 6:37 am: Just suppose that if after Agnew was nominated in 1968 but before the election it had been revealed that he had been involved in the hanky-panky that eventually caused him to resign as VP. Might that have had an effect on that election? I realize supposition is not evidence. But with the Internet, times have changed. Look at what has been disclosed about Palin so quickly after McCain blessed her. As was the case with Eagleton, there is the opportunity to drop Palin; but I doubt the GOP convention will drop her. Perhaps AFTER the convention, she will be dropped – late on a Friday afternoon – possibly replaced with Dick Cheney!
anewnadir September 2, 2008 at 4:50 pm: one word: Eagleton.

There’s a reason he was taken off of the ticket, and that’s because of the common perception that undergoing electroshock treatment showed an inability to govern.

I don’t think Sarah Palin will last much longer. If she does, there will be gaffes, and Democratic attacks on her integrity will have more and more bite.

In the end, you may not notice any discernible effect that she had on poll numbers, but I have a hard time believing that this woman will not have an effect on the electability of a man who you just got done saying has a 40% chance of dying in his first term of office.
j fearon September 3, 2008 at 11:16 am: In line with the above, if candidates generally make good choices for VP, then we might not observe any effect in the data. But this would not mean that quality of VP pick has no influence on presidential vote. I’m not saying that the Palin pick will turn out to be the “off path” choice that shows this to be the case, but in principle …
Andrew September 3, 2008 at 6:53 pm: Yes, exactly. But I did look pretty carefully at the polling on Bush/Quayle in 1988 and, even there, you only had 2% of the voters saying they would vote for Bush but not for Bush-Quayle. And I suspect that 2% is an overestimate.

But, yeah, 1% or 2% is still a big deal! Just not the huge effects people are talking about: “game-changer,” etc.