# 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:** 12
**Author:** ![wonky](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/wonky/32/393_2.png) [@wonky](https://boards.straightdope.com/u/wonky)\ **Post date:** [September 30, 2012, 4:42am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/221 "2012-09-30T04:42:35Z")
> [@adaher](#): > > That is what’s being reported. The problem is, no one is willing to stand by the internals, or even report the internals, because no one actually believes the internals. I know that’s what’s being reported. It’s because the rest of us, the ones who view the race differently, don’t think your point is very interesting. You seem to think that Democratic turnout numbers are like cold fusion, and anyone who isn’t focused on them isn’t focused on them because they are hiding something, while I think that people are focused on the real story, which is who is leading. From my perspective, looking for a conspiracy here is really weird and awfully silly. But whatever rocks your world. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 30, 2012, 5:24am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/222 "2012-09-30T05:24:54Z")
There’s no conspiracy. Just a discrepancy between conventional wisdom about November turnout and what the polls are saying. Only one can be right. ---
**Author:** ![Untoward\_Parable](https://avatars.discourse-cdn.com/v4/letter/u/51bf81/32.png) [@Untoward\_Parable](https://boards.straightdope.com/u/Untoward_Parable)\ **Post date:** [September 30, 2012, 6:10am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/223 "2012-09-30T06:10:32Z")
538’s numbers are a meta-analysis of the polls. The dems don’t want their people celebrating early and losing steam and the repubes don’t want their people to get depressed and the news media would love to have a closer race as people would keep watching their shows. Lots of parties have good reason to deny reality here. (Not that that reality can’t change, Obama could have a stroke or something) ---
**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 30, 2012, 6:28am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/224 "2012-09-30T06:28:27Z")
When the choice is “conventional wisdom” (shit we believe just because) vs data observed by a wide variety of sources … don’t bet on the shit we believe just because. Really this is not a huge story. The problem you are having is that you are stuck with this concept of voter party ID as being a trait like eye color or religion and that the Presidential vote follows having that trait. The converse is actually the case for many. Their choice of candidate drives what answer they give for party ID. During Reagan’s landslide election I suspect many previously Independent would have identified as Republicans and many Democrats began to refer to themselves as Independents … _because_ they were voting for Reagan. Or at least unhappy with Mondale as an alternative choice. They did not vote for Reagan primarily because they thought of themselves as Republican. It was the other way around. A president with decent job approval ratings against an uninspiring competitor and a competing party that is not articulating any consistent theme that resonates with the middle … what a surprise that more are choosing to not identify with the party of Romney and the obstructionist no compromise Congress. Flip it and have a president with poor approval and a theme that resonates with the middle for the GOP and the numbers would change in a flash. This is not the same thing as voter registration or even exit polling data. Democratic party preference lead does not result in the Obama lead, it results from it. ---
**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 30, 2012, 6:34am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/225 "2012-09-30T06:34:07Z")
[Fivethirtyeight finally weighs in!](http://fivethirtyeight.blogs.nytimes.com/2012/09/29/poll-averages-have-no-history-of-consistent-partisan-bias/) > [@](#): > > party identification is not a hard-and-fast demographic characteristic like race, age or gender. Instead, it can change in reaction to news and political events from the party conventions to the Sept. 11 attacks. Since changes in public opinion are precisely what polls are trying to measure, it would defeat the purpose of conducting a survey if pollsters insisted that they knew what it was ahead of time. … > > … On the whole, it is reasonably impressive how unbiased the polls have been. In both presidential and Senate races, the bias has been less than a full percentage point over the long run, and it has run in opposite directions. > > That does not mean the pollsters will necessarily get this particular election right. Years like 1980 suggest that there are sometimes errors in the polls that are much larger than can be explained through sampling error alone. The probability estimates you see attached to the FiveThirtyEight forecasts are based on how the polls have performed historically in practice, and not how well they claim to do in theory. > > But if there is such an error, the historical evidence suggests that it is about equally likely to run in either direction. > > Nor is there any suggestion that polls have become more biased toward Democratic candidates over time. Out of the past seven election cycles, the polls had a very slight Republican bias in 2010, and a more noticeable Republican bias in 1998, 2000 and 2006. > > They had a Democratic bias only in 2004, and it was very modest. > > Still, 2004 went to show that accusations of skewed polling are often rooted in wishful thinking. Nothing about why Gallup was so different than the others yet but still nice to see is take is the same as many others w=here had been taking. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 30, 2012, 7:12am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/226 "2012-09-30T07:12:35Z")
But if people all of a sudden want to call themselves Democrats, how come they aren’t registering as Democrats? Democratic registration is way down from 2008, according to media reports. > **[Drop in Ohio voter registration, especially in Dem strongholds, mirrors...](https://www.foxnews.com/politics/drop-in-ohio-voter-registration-especially-in-dem-strongholds-mirrors-nationwide-trend)** > > Voter registration in the Buckeye State is down by 490,000 people from 4 years ago. Of that reduction,. 44 percent is in Cleveland and surrounding Cuyahoga County, where Democrats outnumber Republicans more than two to one. As scientific as polls are, they don’t trump plain reality. Somethign weird is going on if the polls say one thing, but actual voter registration numbers say the opposite. ---
**Author:** ![wonky](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/wonky/32/393_2.png) [@wonky](https://boards.straightdope.com/u/wonky)\ **Post date:** [September 30, 2012, 7:52am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/227 "2012-09-30T07:52:35Z")
> [@adaher](#): > > But if people all of a sudden want to call themselves Democrats, how come they aren’t registering as Democrats? Democratic registration is way down from 2008, according to media reports. You know what? I bet you can think of a few answers to that before you ask anyone else. And you tell me there’s no conspiracy, yet how could all of these varied polling outfits be making the same dread error and hiding or ignoring it if there were no conspiracy? You’re frustrated that people aren’t reacting to what you think is the big story, and that’s your evidence that the big story isn’t happening. Again, I bet you can think of a few reasons why that might be. I know you want Romney to win, and I’ve got no beef with that. (I’ll be voting against him, but I moved from Ohio to CA, so it doesn’t really matter.) He might win it, too. But just take a step back, apply Occam’s Razor if you can, and think about what has to be true for the polls to be almost universally skewed the way you think they are. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 30, 2012, 8:34am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/228 "2012-09-30T08:34:13Z")
I’ve acknowledged that the polls could be right. If they are, then that means Democrats are poised to turn out in bigger numbers than in 2012. If you don’t believe the latter, then you believe the polls are wrong. ---
**Author:** ![wonky](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/wonky/32/393_2.png) [@wonky](https://boards.straightdope.com/u/wonky)\ **Post date:** [September 30, 2012, 8:50am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/229 "2012-09-30T08:50:27Z")
> [@adaher](#): > > I’ve acknowledged that the polls could be right. If they are, then that means Democrats are poised to turn out in bigger numbers than in 2012. > > If you don’t believe the latter, then you believe the polls are wrong. Name the mechanism and explain why it’s happening. Until you do that, it’s a conspiracy theory. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 30, 2012, 9:38am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/230 "2012-09-30T09:38:48Z")
How about this hypothesis? Democrats are going to turn out in bigger numbers than 2008, which is what the polls are showing. ---
**Author:** ![Lord\_Feldon](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/lord_feldon/32/341_2.png) [@Lord\_Feldon](https://boards.straightdope.com/u/Lord_Feldon)\ **Post date:** [September 30, 2012, 10:26am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/231 "2012-09-30T10:26:33Z")
> [@adaher](#): > > But if people all of a sudden want to call themselves Democrats, how come they aren’t registering as Democrats? Democratic registration is way down from 2008, according to media reports. How sudden is “all of a sudden”? The last chance an Ohio voter had to change party registration was in early March (you can only become a registered member of a party by voting in that party’s primary). It is literally impossible to change your party registration here between March 2012 and April 2013 (and if there no primaries for your party on your ballot in 2013, then you have to wait until 2014). Also, the Republicans had a high-profile presidential primary this year and the Democrats didn’t. People who were only paying attention to the presidential race had no reason to register as Democrats this year. ---
**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 30, 2012, 11:14am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/232 "2012-09-30T11:14:51Z")
> [@adaher](#): > > How about this hypothesis? Democrats are going to turn out in bigger numbers than 2008, which is what the polls are showing. Imagine if, suddenly, a significant portion of 2008 R’s now call themselves I’s, and a portion of 2008 I’s now call themselves D’s. But the people vote the exact same way as they did in 2008. Then the election has the same result, even though the “Democratic turnout” is way higher than 2008. It’s not that unusual for self-identification to change significantly from year to year. And that’s what the data is showing. ---
**Author:** ![adaher](https://avatars.discourse-cdn.com/v4/letter/a/dec6dc/32.png) [@adaher](https://boards.straightdope.com/u/adaher)\ **Post date:** [September 30, 2012, 11:17am UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/233 "2012-09-30T11:17:51Z")
Here’s my big problem with Ohio polling: the pollsters say there are more Democrats than Republicans who will vote. Except Republicans outnumber Democrats in Ohio, and furthermore Romney leads among independents in those very same polls: > **[Media Ignore Independents' Swing Toward Romney](https://www.breitbart.com/the-media/2012/09/27/romney-winning-swing-state-independents-media-ignores/)** > > Curiously, Independents seem to be mostly ignored this election cycle, and their presidential vote preference is almost unanimously ignored by the MSM. Instead, So in order for the polls to be right, we have to assume in Ohio that Democrats will best Republicans in turnout even though there are fewer registered Democrats, and do so enough to overcome Romney’s advantage among independent voters. Could it be Democrats are just unusually likely to say they will vote and not bother to register? Maybe they assume they can register at the polls or something? ---
**Author:** ![zut](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/zut/32/2875_2.png) [@zut](https://boards.straightdope.com/u/zut)\ **Post date:** [September 30, 2012, 12:10pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/234 "2012-09-30T12:10:33Z")
As an interesting historical example, read [this article on 2004 and 2008 voting in North Carolina](http://www.wfae.org/post/do-historic-voting-patterns-forecast-2012-election). Both in 2004 and 2008, the percentages of ballots actually cast closely matched the percentages of official party registration. However, the exit polling - in other words, the self-identification of the voters _at the time of the election_ - was significantly different than either. In addition, a glance at the 2004 North Carolina numbers shows the glaring error in the Breitbart article above. In 2004, at the time of the election, 47% of the NC voters were registered Democratic, 19% registered independent and only 34% Republican. Exit polling showed Kerry getting about 40% of the independent vote, which, added to the 47% Democratic vote, should add up to a Kerry win with over 54% of the vote. But Kerry actually _lost_ North Carolina something like 56-44. Correcting for partisans crossing over to vote for the other candidate does result in a 51-49 Bush win, but predicts a 2-point squeaker rather than the actual 12-point blowout. ---
**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 30, 2012, 1:11pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/235 "2012-09-30T13:11:38Z")
The simple truth is that the recent voter Democratic party ID numbers ARE NOT UNUSUAL (other than to some Conservative media hacks desperate to try to create a narrative). [Here’s Pews data over time.](http://www.people-press.org/2012/08/23/a-closer-look-at-the-parties-in-2012/) Democratic voter ID (as polled) is not hugely big compared to historic norms and is less than in 2008, which was crazy big for the polled Democratic voter ID advantage even a year before the election, thanks to Bush’s dismal approval ratings. The reason that the pollsters are not trumpeting that Democratic turn out is going to be much more than it was in 2008 is simply _because the data does not show that_. I think what confuses you (besides your source material and your huge desire to believe that Romney’s chances are better than they are) is the faulty assumption that there is some one to one connection between party identification as polled and in registration numbers or exit polling party ID. Again, [HufPo’s graphic](http://elections.huffingtonpost.com/pollster/party-identification#!estimate=custom). To the degree that polled party ID is _correlated with_ how people are going to turn out and vote we see that compared to 2008 Democratic voter ID is down. The spread between Democratic voter ID and GOP voter ID is up slightly, from a very rough 5% to about 8% and the numbers who identify as Independent are up. As to voter registration. People can vote without being registered as being pat of one party or the other. In Ohio you can even vote in the primary of one party while being a registered member of another. Voter registration by party does not mean very much (nothing at all really) as means of tracking fluid changes in party ID. ---
**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 30, 2012, 2:08pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/236 "2012-09-30T14:08:04Z")
> [@adaher](#): > > Here’s my big problem with Ohio polling: the pollsters say there are more Democrats than Republicans who will vote. Except Republicans outnumber Democrats in Ohio, and furthermore Romney leads among independents in those very same polls: > > [http://www.breitbart.com/Big-Journalism/2012/09/27/Romney-Winning-Swing-State-Independents-Media-Ignores](http://www.breitbart.com/Big-Journalism/2012/09/27/Romney-Winning-Swing-State-Independents-Media-Ignores) > > So in order for the polls to be right, we have to assume in Ohio that Democrats will best Republicans in turnout even though there are fewer registered Democrats, and do so enough to overcome Romney’s advantage among independent voters. > > Could it be Democrats are just unusually likely to say they will vote and not bother to register? Maybe they assume they can register at the polls or something? No, you’re still not getting it- the data says that many of those _registered_ Republicans _now call themselves independent_. Just now, at this very moment (or at the moment of the polling). Tomorrow they may call themselves Republicans again. But in the polls, many of them are calling themselves Independents. ---
**Author:** ![RickJay](https://avatars.discourse-cdn.com/v4/letter/r/bb73d2/32.png) [@RickJay](https://boards.straightdope.com/u/RickJay)\ **Post date:** [September 30, 2012, 2:11pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/237 "2012-09-30T14:11:11Z")
538 addresses the issue: [http://fivethirtyeight.blogs.nytimes.com/2012/09/29/poll-averages-have-no-history-of-consistent-partisan-bias/#more-35258](http://fivethirtyeight.blogs.nytimes.com/2012/09/29/poll-averages-have-no-history-of-consistent-partisan-bias/#more-35258) ---
**Author:** ![Sitnam](https://avatars.discourse-cdn.com/v4/letter/s/6a8cbe/32.png) [@Sitnam](https://boards.straightdope.com/u/Sitnam)\ **Post date:** [September 30, 2012, 2:21pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/238 "2012-09-30T14:21:13Z")
> [@jsgoddess](#): > > He’s up to 2% with Black Protestants now! Those victims must have decided to give up their handouts and take some responsibility for their lives. ---
**Author:** ![Really\_Not\_All\_That\_Bright](https://avatars.discourse-cdn.com/v4/letter/r/e8c25b/32.png) [@Really\_Not\_All\_That\_Bright](https://boards.straightdope.com/u/Really_Not_All_That_Bright)\ **Post date:** [September 30, 2012, 2:22pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/239 "2012-09-30T14:22:33Z")
> [@adaher](#): > > But if people all of a sudden want to call themselves Democrats, how come they aren’t registering as Democrats? Democratic registration is way down from 2008, according to media reports. Because most people live in states with open primaries, so there’s no point in registering. I know people (as you probably do) who have voted straight ticket for decades while still registered for the other side. Even in states with closed primaries, fewer people are going to bother registering Democratic because there _was_ no Democratic presidential primary. ---
**Author:** ![Gorsnak](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/gorsnak/32/7587_2.png) [@Gorsnak](https://boards.straightdope.com/u/Gorsnak)\ **Post date:** [September 30, 2012, 3:43pm UTC](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647/240 "2012-09-30T15:43:07Z")
[Suggested reading](http://www.mysterypollster.com/main/2004/09/weighting_by_pa.html) for **adaher**. After each paragraph, scroll up and reread the publication date. [Previous page](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647.md?page=11) [Next page](https://boards.straightdope.com/t/latest-conservative-grievance-polls-oversample-democrats/635647.md?page=13)