# Latest conservative grievance - polls "oversample" Democrats **URL:** \ **Category:** Politics & Elections\ **Created:** [September 24, 2012, 1:31am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647 "2012-09-24T01:31:14Z")\ **Posts on this page:** 20\ **Page:** 8
**Author:** ![Truman\_Burbank](https://avatars.discourse-cdn.com/v4/letter/t/ecc23a/32.png) [@Truman\_Burbank](https://boards.straightdope.com/u/Truman_Burbank)\ **Post date:** [September 27, 2012, 6:40am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/141 "2012-09-27T06:40:32Z")
“Higher as a proportion” I’ve been reading nothing lately, it seems, other than pundits discussing how the demographic tide is turning against the Repulicans. “Not enough angry white men” etc. I still don’t get why you think this is a ‘man bites dog’ story… ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 6:53am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/142 "2012-09-27T06:53:43Z")
Problem is, the demographic tide is not really turning against Republicans. If anything, it should be going for Republicans, at least in 2012. Not only was minority turnout much higher than normal in 2008, white turnout was down compared to 2004 The pollsters are assuming that 2008’s low level of white turnout remains, while minority turnout increases even more. Which could be the actual source of the oversampling. > **[Battle Ground Watch](https://www.battlegroundwatch.com/)** > > Well, we all know how the 2012 election turned out. > However at the time this was a blog site that many people followed. Content is from the site's archived pages offering a brief glimpse of what this site was all about. Thank you, Keith Backer. And it was more white men who stayed home in 2008. The pollsters are assuming these white dudes won’t be there in 2012 either. BTW, demographic projections have whites remaining above 70% of the population in 2050. Granted, many of those are Hispanic whites, but I don’t think anyone can predict confidently that in 2050 there will be major cultural differences between Hispanic whites and non-Hispanic whites. It’s hard to argue that such differences exist now, especially given the treatment of George Zimmerman in the media. ---
**Author:** ![Truman\_Burbank](https://avatars.discourse-cdn.com/v4/letter/t/ecc23a/32.png) [@Truman\_Burbank](https://boards.straightdope.com/u/Truman_Burbank)\ **Post date:** [September 27, 2012, 7:12am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/143 "2012-09-27T07:12:03Z")
well, I guess you just don’t have a problem at all, then. You’ll be dancing in the streets as the (D)s try to figure out what went wrong, and Nate shuts down his blog. Or, you might not be ‘getting it’ because you have to ignore data that won’t make it through your partisan blinders. It will be very interesting to see which, won’t it? ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 7:15am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/144 "2012-09-27T07:15:54Z")
I’m looking at the data. The data of the media pollsters says minorities will turn out in greater numbers than 2008. If they truly believe that, they should report it. But the one pollster who was asked about Democratic turnout advantages disavowed it. Why would he do that? ---
**Author:** ![Truman\_Burbank](https://avatars.discourse-cdn.com/v4/letter/t/ecc23a/32.png) [@Truman\_Burbank](https://boards.straightdope.com/u/Truman_Burbank)\ **Post date:** [September 27, 2012, 7:20am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/145 "2012-09-27T07:20:09Z")
Saying “I’m not sure it will hold” is not really the same thing as 'disavowing it", is it? Now, I’ve only got one semester of stats under my belt, a very long time ago. I’m not anything close to being a polling geek. So, as with any area where I am not an expert, I need to defer to those who understand the intricacies better than I. Do you think you see an obvious truth in the data that Nate Silver, for one example, is missing? ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 7:26am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/146 "2012-09-27T07:26:50Z")
nate Silver believes, quite reasonably, that adjusting for party ID is not scientific. Given what we know about polling, that’s true. But because there just haven’t been that many Presidential elections, every election breaks precedent or differs from expectations in some ways. Conventional wisdom is that Democrats will not match 2008 turnout levels. So either the CW is wrong, in which case it should be clearly stated, or the CW is right and the polling is wrong. As of yet, the pollsters getting these results aren’t willing to say definitively what they think. Neither is Nate Silver. Yet if all the polls but Gallup and Rasmussen are showing this huge surge in turnout, why not go along with it? Because no one really believes it. ---
**Author:** ![Truman\_Burbank](https://avatars.discourse-cdn.com/v4/letter/t/ecc23a/32.png) [@Truman\_Burbank](https://boards.straightdope.com/u/Truman_Burbank)\ **Post date:** [September 27, 2012, 7:38am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/147 "2012-09-27T07:38:16Z")
So I read your link in Post #142, and it sounds like wishful thinking. I’ll stick with Nate. Good night. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 7:48am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/148 "2012-09-27T07:48:50Z")
If it is wishful thinking, then it shouldn’t be hard to predict that Democratic turnout will be +8 or more nationally. I’ll bet even Silver would balk at taking that bet. Yet his model is predicated in large part on that. ---
**Author:** ![iiandyiiii](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/iiandyiiii/32/7924_2.png) [@iiandyiiii](https://boards.straightdope.com/u/iiandyiiii)\ **Post date:** [September 27, 2012, 9:12am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/149 "2012-09-27T09:12:14Z")
> [@adaher](#): > > If it is wishful thinking, then it shouldn’t be hard to predict that Democratic turnout will be +8 or more nationally. I’ll bet even Silver would balk at taking that bet. Yet his model is predicated in large part on that. It’s been said over and over again, but _higher party ID for Democrats does not mean higher turnout for Democrats_. They call a thousand or so people. 37% or whatever self-identify as Democrats (or perhaps 37% of those the pollsters deem as likely voters). If that’s higher than 2008, it may well be because _more voters self identify as Democrats now then they did in 2008_. Not higher Democratic turnout, but higher Democratic self-identification. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 11:56am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/150 "2012-09-27T11:56:28Z")
That’s still higher turnout. In 2008, the electorate was D+8. Is there anyone willing to predict that the electorate will be D+8 or higher on election day? ---
**Author:** ![iiandyiiii](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/iiandyiiii/32/7924_2.png) [@iiandyiiii](https://boards.straightdope.com/u/iiandyiiii)\ **Post date:** [September 27, 2012, 12:00pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/151 "2012-09-27T12:00:10Z")
> [@adaher](#): > > That’s still higher turnout. In 2008, the electorate was D+8. Is there anyone willing to predict that the electorate will be D+8 or higher on election day? That may be what the current data predicts. In any field, one can’t just dismiss data because it doesn’t fit your expectations. This data might be wrong, but not just because it “feels wrong”. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [September 27, 2012, 12:29pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/152 "2012-09-27T12:29:39Z")
**adahar** now you are starting to sound like **omg**. And seriously, that is not a compliment. Since you put all your eggs in the basket of hope upon hope that Gallup has it right let me pass today’s [Gallup’s numbers](http://www.gallup.com/poll/157709/obama-approval-vote-support-reach-better.aspx) and take on to you: > [@](#): > > Obama’s job approval rating has ranged from 43% to 52%, levels historically associated with either near-certain defeat or near-certain re-election for an incumbent. Obama is currently at the high end of that range, and has had more good days than bad this month, in terms of achieving job approval ratings of 49% or better. Looking at their [breakdown](http://www.gallup.com/poll/154559/US-Presidential-Election-Center.aspx) by demographic page (note after their somewhat discussed method of identification of race and correction methodology) we see Obama winning an amazing variety of groups. By age only losing the 65 plus group. Winning almost all educational levels including post graduate educated (losing only college educated only). Basically tied for the Independent vote. Winning Moderates by a wide margin. Yes Gallup says there is still time for things to change … heck Nate does too. But really? ---
**Author:** ![Jack\_Batty](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/jack_batty/32/3201_2.png) [@Jack\_Batty](https://boards.straightdope.com/u/Jack_Batty)\ **Post date:** [September 27, 2012, 1:02pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/153 "2012-09-27T13:02:09Z")
All I know is I’m warming up my molotov throwing arm for the riots. ---
**Author:** ![Jas09](https://avatars.discourse-cdn.com/v4/letter/j/d07c76/32.png) [@Jas09](https://boards.straightdope.com/u/Jas09)\ **Post date:** [September 27, 2012, 1:31pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/154 "2012-09-27T13:31:38Z")
> [@adaher](#): > > That’s still higher turnout. In 2008, the electorate was D+8. Is there anyone willing to predict that the electorate will be D+8 or higher on election day? I think that D+8 is unlikely, but D+6 is certainly plausible. I’ll explain why. Right now you’re seeing the D percentage roughly the same as 2008, with the R depressed and I way up. Ahead of elections this happens, but folks tend to “firm up” in their exit poll response - after you just voted for an R you’re more likely to call yourself an R. You may notice even the polls with big Obama leads are showing Romney winning I’s - that’s because that cohort includes a lot of Rs calling themselves Is. I think on election day they’ll admit to being Rs again, and you’ll see Obama up among the “true” independents remaining. So, based on current data, I think D+6 is very plausible, and would place a bet on that. It might even match the D+8 from 2008 just due to demographic changes, but I wouldn’t make an even-money bet on that. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 1:31pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/155 "2012-09-27T13:31:58Z")
> [@iiandyiiii](#): > > That may be what the current data predicts. > > In any field, one can’t just dismiss data because it doesn’t fit your expectations. This data might be wrong, but not just because it “feels wrong”. Absolutely. However, if the polls are all showing that, how come the prognosticators aren’t willing to predict that what the polls are saying is what will happen? They are perfectly willing to say Obama is ahead by X points. Not so willing to say that Democratic turnout is looking to exceed 2008. That’s all I want to know. If the polls are right, then stand by the internals. ---
**Author:** ![Richard\_Parker](https://avatars.discourse-cdn.com/v4/letter/r/35a633/32.png) [@Richard\_Parker](https://boards.straightdope.com/u/Richard_Parker)\ **Post date:** [September 27, 2012, 1:37pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/156 "2012-09-27T13:37:40Z")
Your argument only makes sense if pollsters are saying that their individual poll’s topline result will be the result on Nov. 6. Can you give a cite for anyone doing that? If not, then why should they be stating that their internals will match the result on Nov. 6? ---
**Author:** ![iiandyiiii](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/iiandyiiii/32/7924_2.png) [@iiandyiiii](https://boards.straightdope.com/u/iiandyiiii)\ **Post date:** [September 27, 2012, 1:38pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/157 "2012-09-27T13:38:48Z")
> [@adaher](#): > > Absolutely. However, if the polls are all showing that, how come the prognosticators aren’t willing to predict that what the polls are saying is what will happen? > > They are perfectly willing to say Obama is ahead by X points. Not so willing to say that Democratic turnout is looking to exceed 2008. That’s all I want to know. If the polls are right, then stand by the internals. Saying that “Democratic turnout is looking to exceed 2008” implies that some significant number of Democrats who did not vote in 2008 will vote in 2012. The data does not necessarily indicate this is so- it may just be that some independent voters in 2008 now consider themselves Democratic voters. It’s just self-identification- which tends to fluctuate. ---
**Author:** ![Truman\_Burbank](https://avatars.discourse-cdn.com/v4/letter/t/ecc23a/32.png) [@Truman\_Burbank](https://boards.straightdope.com/u/Truman_Burbank)\ **Post date:** [September 27, 2012, 1:39pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/158 "2012-09-27T13:39:15Z")
Adaher, I looked a bit further into “Battlegroundwatch” or whatever that website was you linked to last page, and immediately found this sentence: “Maybe Pennsylvania is in play after all. It’s hard to know what to believe these days. Support for Obama has been characterized as “rotten underneath” meaning it will crumble should Romney gain any momentum.” (That’s an example of what I meant by ‘wishful thinking’). You appear to be taking your talking points from sources that are far from dispassionate about the election outcome, and massage the data to try to support their desired conclusions. Ironically, they then accuse all of the major pollsters of doing the same thing. When you have to twist your narrative to get Fox’s polling into the same pigsty as 538, Rasmussen, and Gallop, your narrative is exceedingly strained. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 1:41pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/159 "2012-09-27T13:41:59Z")
> [@Jas09](#): > > I > > So, based on current data, I think D+6 is very plausible, and would place a bet on that. It might even match the D+8 from 2008 just due to demographic changes, but I wouldn’t make an even-money bet on that. I agree that D+6 is plausible, but at D+6, Obama doesn’t have the big lead the polls say he does. That reduces his lead by about 3 points, since the average of polls is around D+9. Swing state polls are also showing Democrats exceeding 2008. Now the Gallup poll is interesting. Let’s take a closer look at the data: McCain beat Obama among white voters by 12 points. Romney leads white voters by 14. Obama beat McCain among nonwhites by 80 points. He’s beating Romney by 64. So Obama is significantly underperforming with both groups compared to 2008, yet still winning. That can only be explained by Gallup assuming MORE nonwhite turnout than in 2008. A very shaky assumption. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 27, 2012, 1:43pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/160 "2012-09-27T13:43:21Z")
> [@Richard\_Parker](#): > > Your argument only makes sense if pollsters are saying that their individual poll’s topline result will be the result on Nov. 6. Can you give a cite for anyone doing that? > > If not, then why should they be stating that their internals will match the result on Nov. 6? Given that the media and many dopers are essentially calling the race over, I’d assume they think the polls will hold up. If this is just a temporary bounce for Obama based on higher September enthusiasm for Democrats, then we should expect the race to go back to a near tie. I’m not just quoting talking points here, I love polling and the polls are almost always right. But the concerns Republicans have expressed sound legit to me. The polls are showing something that pundits are uncomfortable predicting will hold up in November, a huge Democratic identification advantage. I just don’t see how pundits can trumpet a big lead and call the race all but over, yet be unwilling to stand by the pollsters’ internals. If you don’t think the internals are likely to hold up until Nov. 6, then you don’t think this race is over. [Previous page](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647.md?page=7) [Next page](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647.md?page=9)