« Step one is denial |
| New west »
Gallup, we hardly knew ye: A breakdown of pollster performance in the 2012 election.
Bad enough for the conservative entertainment complex to cook the data, downright weird for the Romney campaign to "unskew" its own polls.
For some reason I keep expecting to hear the phrase "It's not settled science."
Nov 12, 2012 at 09:24 AM | Permalink
TrackBack URL for this entry:http://www.typepad.com/services/trackback/6a00d8341cc33e53ef017c335ef492970b
Listed below are links to weblogs that reference GIGO:
You can follow this conversation by subscribing to the comment feed for this post.
Delicious. All of it. But we're heading towards the cliff, so let's hope the party of anti-intellectuals can appreciate what it's really facing in this case and not make the same mistake twice.
Nov 12, 2012 at 10:31 AM
"downright weird for the Romney campaign to 'unskew' its own polls"
Hey, if you believe people of color are too lazy to work, it's a short hop to believing they're too lazy to vote.
Nov 12, 2012 at 10:52 AM
Over at TPM, the meme I invoke with "not settled science" is discussed.
Also, readers tell Josh that of course the insiders knew a Romney loss was at the very least a good possibility -- they're just ass-covering for the folks who believe the entertainment machine. Whatever the merits of this argument, I have talked to enough people and read enough online to say that a lot of people were genuinely shocked by the outcome -- which means they were limiting their worldview to the CEC and/or wishful thinking.
Ed Cone |
Nov 12, 2012 at 06:25 PM
anyone who believes that scientific polling is scientific is practicing bad science. The bettors had it right again. They had Hillary or Obama over a weak candidate in
2008 and gave Romney less than a 20% chance right after the first debate. You could see the 2 point hiccup after the debate and the GOP's last chance for the Senate coincide with Aiken opening his piehole.
Intrade bettors don't have a dog in the ring but they can change sides in the middle of the fight. The larger the pool the more accurate the nonvoters are. They cast a ballot only in their best interest. They can go long, short or spread.
The problem is now that the method is so accurate that the reason to show up and cast a ballot for a slim/low odds candidate or bond issue does not exist. What are the enlightened lemmings going to do.
tk solomon |
Nov 12, 2012 at 06:35 PM
"The bettors had it right again."
The bettors had it right because the polls had it right. What do you think they based their bets on? Their own innate common sense and nothing else? No, they had to factor in information from the campaigns. Now, it appears that they factored in the best and most accurate information, which, as we've seen, came from averaging the polls. It's true that political markets did better than the pundits (who would have had us believe that Romney was in the lead after the first debate), but so did the poll aggregators.
Anyone who believes that scientific polling isn't scientific doesn't know what science is.
Andrew Brod |
Nov 12, 2012 at 07:26 PM
The comments to this entry are closed.