Lost in Translation: Polls Show Conflicting Results

[Bullet1987]

It seems as though there are two outcomes in the polls leading up to Super Tuesday. Some polls show a post-Debate bump for Hillary Clinton, and some don't. Case in point being the brand new Zogby poll and Field Poll out of CA...and a handful of other post-debate polls from CA that show it as a statistical dead heat. Then there is the MSNBC poll that shows Clinton with a 9-point surge. Which one is closer to the truth? It's impossible to know...which is why I've always been a big critic of poll obsession because it's very easy to MAKE a poll show what you want it to. If you wanted to show a post-debate surge, then there are ways of doing that. Polls are too damn political, so can't be taken as seriously as some people on this site do (like the word of God). Case in point being the polling done post-Iowa. Very few firms factored in the obvious Clinton-leaning that took place because they wanted to continue to portray the story of Obama surging. Polls play so much into the MSM, therefore should be taken more critically...because at the end of the day all the media wants is a story. Polling companies can show a post-debate surge so then the MSM can come behind them and say..."hey, it looks like Clinton got a surge following the last debate." When really it could just be a flucuation of numbers.

I'm not saying she didn't get a boost, but her performance wasn't as lights out as some of the polling would suggest (especially a 9-point jump in CA when every other poll shows it much closer). She fumbled HARD on the IWR, and most post debate anaysis called it even. Polls should be taken with a grain of salt. Not a damn one showed Obama winning SC by 28 points...there are too many things that factor into winning that polls simply can't gauge.

[LadyVT]

1. pollster.com -- collects all the polls, every day

[Drunken Irishman]

2. One thing: The post-debate bump was only one day and didn't show a trend.

I said this in those posts about the polls and I'll say it again. Just because they show Clinton gaining in one day does not mean there is a surge or a bump. It just means, that day, she received the most support. These daily tracking polls aren't nearly as accurate as the polls we're seeing coming out today -- since those are done with a larger sample size and don't roll over. And while those polls are questionable, I've always felt more confident in them than tracking polls.

The problem with tracking polls is that the MOA could ultimately kick in one day and not be an issue the next. A bump could be a trend, or it could just be a statistical anomaly, you can't really tell until one or two days of more polling. If Clinton really did get a post-debate boost, she should go up 2 or more points tomorrow in Rasmussen's poll. If it was an anomaly, she'll either stay at her current numbers, or she'll drop again.

But it's also important to remember that the Rasmussen numbers show close to 20% of voters who are either supporting another candidate, or are undecided. That could change things.

[Dogmudgeon]

3. 8.7 million people watched the debate

That is not enough to create much of a bump.

A lot of people here are hanging on the polls obsessively, but the undecideds, the MOE, and the less-than-certain voters make up a huge group this time. I can almost guarantee that half the races on Tuesday will be surprises.

--p!

[geiger]

4. Nice reality check, thanks.

[FrenchieCat]

5. She didn't even win the debate.....in my opinion......

She let immigrants know that she would pit them against Black folks anytime required, that she likes mandates, and that she still can't seem to come clean with her war vote.....and I'll add that she dogged out Levin mischarecterizing his amendment (hope he comes out and endorses OBama for that shit). Sje still can't see that vote as being wrong, although it was.

And she supposed to get a bump from that performance. Getta out of town!

The only thing that she did right was to look at the audience a lot, and act like she was giving a lot of facts, when she was doing a whole lot more talking than was needed 1/2 of the time.

Meanwhile, Obama told us exactly how he would kick McCain's ass in a General election.....or Romneys....in a way that Hillary can't.

But ok...it was a tie! :eyes:

[CreekDog]

6. Field v. Zogby is not a case in point...those polls are both within MOE's

Zogby shows 2 pt Obama lead and Field shows 2 pt Clinton lead both leads are statistically insignificant, meaning that if both polls were redone, they could produce results that currently match the other poll.

I thought it was crazy that you said that these two polls were case in point. A case in point are two polls showing 10 point differences.