Comments on FiveThirtyEight: Politics Done Right: Likely Voters More Republican?

酒店經紀人, 菲梵酒店經紀, 酒店經紀, 禮服酒店上班, 酒店小姐兼職, 便服酒店經紀, ...

2009-09-11T08:54:08.102-04:00

酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店經紀,
酒店打工經紀,
制服酒店工作,
專業酒店經紀,
合法酒店經紀,
酒店暑假打工,
酒店寒假打工,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店工作,
酒店打工經紀,
制服酒店經紀,
專業酒店經紀,
合法酒店經紀,
酒店暑假打工,
酒店寒假打工,
酒店經紀人,
菲梵酒店經紀,
酒店經紀,
禮服酒店上班,
酒店小姐兼職,
便服酒店工作,
酒店打工經紀,
制服酒店經紀,
酒店經紀,
菲
梵,

艾葳酒店經紀公司提供專業的酒店經紀, 酒店上班小姐,八大行業,酒店兼職,傳播妹,或者想要到打工兼差、...

2009-09-11T02:29:45.102-04:00

艾葳酒店經紀公司提供專業的酒店經紀, 酒店上班小姐,八大行業,酒店兼職,傳播妹,或者想要到打工兼差、打工,兼差,或者八大行業,酒店兼職,想去酒店上班, 日式酒店,制服酒店,ktv酒店,禮服店,整天穿得水水漂漂的,還是想去制服店當上班小姐,水水們如果想要擁有打工工作、晚上兼差工作、兼差打工、假日兼職、兼職工作、酒店兼差、兼差、打工兼差、日領工作、晚上兼差工作、酒店工作、酒店上班、酒店打工、兼職、兼差、兼差工作、酒店上班等,想了解酒店相關工作和特種行業內容,想兼職工作日領、假日兼職、兼差打工、或晚班兼職想擁有快速賺錢又有保障的工作嗎???又可以現領請找專業又有保障的艾葳酒店經紀公司!

艾葳酒店經紀是合法的公司工作環境高雅時尚，無業績壓力，無脫秀無喝酒壓力，高層次會員制客源，工作輕鬆，可日領、現領。
一般的酒店經紀只會在水水們第一次上班和領薪水時出現而已，對水水們的上班安全一點保障都沒有！艾葳酒店經紀公司的水水們上班時全程媽咪作陪，不需擔心！只提供最優質的酒店上班,酒店上班,酒店打工環境、上班條件給水水們。心動嗎!? 趕快來填寫你的酒店上班履歷表

水水們妳有缺現領、有兼職、缺錢卡奴的煩腦嗎?想到日本留學缺錢嗎?妳是傳播妹??想要擁有高時薪又輕鬆的夜間兼職工作,打工機會和,假日打工,假日兼職賺錢的機會嗎??想實現夢想卻又缺錢沒錢嗎!??
艾葳酒店台北酒店經紀招兵買馬!!徵專業的酒店打工,想要去酒店的水水,想要短期日領,酒店日領,禮服酒店,制服店,酒店經紀,ktv酒店,便服店,酒店工作,禮服店,酒店小姐,酒店經紀人,
等相關服務幫您快速的實現您的夢想~!!

^^ nice blog!! thanks a lot! ^^徵信, 徵信社,徵信, 徵信社, 徵信...

2009-03-11T21:58:00.001-04:00

^^ nice blog!! ^@^徵信, 徵信, 徵信, 徵信社, 徵信社, 徵信社...

2009-03-11T21:58:00.000-04:00

情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,...

2009-02-14T01:53:00.000-05:00

徵信社, 感情挽回, 挽回感情, 徵信, 徵信社, 徵信, 捉姦, 徵信公司, ...

2009-02-08T20:39:00.000-05:00

Looks like everyone discovered the secret to a goo...

2008-07-19T00:09:00.000-04:00

Looks like everyone discovered the secret to a good thread while I was at Netroots Nation: Continue to discuss something you're interested in older post. Interesting.

Modeler: Back for a chill-out from the front threa...

2008-07-18T21:53:00.000-04:00

Modeler: Back for a chill-out from the front threads.

One thought that has been on my mind for some times is that contraception has increased the probability of children not just being 'produced', but actually being wanted. This should have improved (and, according to my observation, actually has improved) parent-child relations, which in turn leads to less generational conflict and younger generations assuming more responsibility at an earlier point in time. Not across all social strata, of course (I am aware of the AA single mother phenomenon in the USA, and in parts of the Turkish community in Germany, as presumably as well in parts of the Latino community in the USA, parent-children relations are still different), but at least in the middle-class mainstream. This might be another motivational factor that explains younger voters being more knowledgeable. [Asides, one might argue that 'knowing more' than your parents is an intelligent way to transform the generational conflict into a new, more civilized quality].

Moreover, 'baby boomers' have such a demographic power that the following generation has hardly a chance to influence the cultural and political mainstream, and is almost inevitably forced into a sub-culture. This is probably holding even more true in Germany than in the USA. Due to low birth rates during and directly after WW II, the German baby-boomers (my generation - born 1959-1968, a few years later than the American ones) make up for almost one quarter of the total German population, and are clearly determining the cultural-political mainstream. With the baby-boomers ageing, and the next generation being sub-cultural or apathetic due to crowding-out, there is suddenly a new 'window of opportunities' opening up for the following generation. I am seeing it here in Germany with young colleagues that have been born in the early 1980s, and I could imagine a similar pattern in the USA. You might call it another facet of the 'opportunity' part of the equation.

So, I think, you are right. Ultimately it is about "displeasure with the direction of the country [which] also motivates those who are going to inherit it". The neccessary (and implicit) pre-conditions are that those who are going to inherit it are assuming responsibility for their inheritance, and see an opportunity to influence the course of action.

Frank,I'm still reading. :-)Good point about the ...

2008-07-18T20:34:00.000-04:00

Frank,

I'm still reading. :-)

Good point about the education differences, but the trend in improving education has been going on for a while now, and Pew says:

One pattern that differs from previous surveys of political knowledge is that younger voters are significantly more knowledgeable about the candidates' positions than are older voters.

I think the major factors that have changed are the role of the internet and, as you mentioned, the novelty of the candidates. I'd guess displeasure with the direction of the country also motivates those who are going to inherit it. I think your troika is reasonable.

I cannot begin to tell you how many problems I hav...

2008-07-18T20:03:00.000-04:00

I cannot begin to tell you how many problems I have with this poll. Start off with them withholding results that favored Obama to the inconsistency results for RV and LV. ABC especially has been trying to make this race look close.

Don't know if there is still anybody reading in th...

2008-07-18T18:44:00.000-04:00

Don't know if there is still anybody reading in this quiet corner outside the troll stream - amyway:

Cugel: "We know old people are going to vote because they've voted before." In fact, we even don't know that. I have made the observation that old people may stop voting when they start getting serious health problems, for feeling they don't have a right to decide on a future that probably is not their's anymore. However, I don't have any idea how common this sort of attitude is.

Modeler: "I wonder to what extent young people are more well-informed because they are more likely to fact-check online." This is probably one factor. Another one should be the increase in education levels from generation to generation. Finally, I feel Daile Petry made an extremely worthwhile point when noting that so far, for many, in particular young voters, "choices have been two rich old white men", which is not really stimulating young people's desire to become informed. This year's choice, however, is different. In short, it probably comes down to the troika of motivation (new type of candidate), ability (better educated) and opportunity (on-line
information access).

Daile Petry: Credit to you to have brought a number of my ideas to the point. I watched Kerry's speech at the 2004 convention, and even though the speech itself was pretty good, my immediate reaction was" He is not going to win. Too much 1968, and not enough 2004. Too much New England, no Latino and no mid-west appeal." The first time I saw Obama was in a CNN interview on the evening of his 2006 election to the Senate, and my immediate reaction, just as yours, was "He can make it as President, because he is fresh and new in style". I did, however, not expect Obama's time to come already in 2008 ..

Not true, according to this survey.ARG: New regist...

2008-07-18T17:40:00.000-04:00

Not true, according to this survey.

ARG: New registrants can be classified as a likely voter and it is also possible for someone stating that they intend to register on election day but are not registered when we talk to them.

Cue Cugel rant about ARG. :-) However, I could imagine including some unregistered voters who appear highly likely to vote. For instance, a voter could be considered if the voter is less than 22 years old and the state has same-day registration.

What's wrong with ARG? or any other pollster for that matter, unless it's Zogby Interactive's online polls for obvious reasons.

I doubt any other pollsters use such a loose "likely voter" screen that would ever accept NON-REGISTERED voters Gallup and Rasmussen don't.

If somebody says they will register on election day, that doesn't mean they will do either. But, then someone saying they are "certain" to vote doesn't mean they will vote. But, it's a good bet that the unregistered won't do either.

After all, unless they've just moved or something they have ignored voting for presumably a long time. In the case of the young, they might have an excuse.

"But, I remember I went and registered immediately when came of age. Took the buggy down and registered, just like that! Why, I remember it like it was yesterday! The Civil War had been over for a while by then and we'd just come out of the great shoe crisis of '88! . . ."

WHOA! I was channeling John McCain there for a minute. Have to stop that. Hey! You kids get off my lawn!

; ^ )

Cugel said: Non-registered voters are CERTAINLY no...

2008-07-18T17:15:00.000-04:00

Cugel said:

Non-registered voters are CERTAINLY not going to make it past ANY "likely voter" screen!

Not true, according to this survey.

ARG: New registrants can be classified as a likely voter and it is also possible for someone stating that they intend to register on election day but are not registered when we talk to them.

Cue Cugel rant about ARG. :-) However, I could imagine including some unregistered voters who appear highly likely to vote. For instance, a voter could be considered if the voter is less than 22 years old and the state has same-day registration.

I thought this nugget from pollster.com was interesting:

True, media pollsters have spent decades developing likely voter "models" to identify the true electorate, but most of that research identifies characteristics that are proven to predict turnout a few weeks before the election

It's remarkable that there are not more long-term models available.

"Anonymous said... The problem with using a "re...

2008-07-18T16:22:00.000-04:00

"Anonymous said...

The problem with using a "registered voters" screen for polling this far from the election is that some states allow very late or even same-day registration. This is an important bias for projections since students, minorities, transients, the poor, and first-time voters are likely to register late so they won't be tracked in polls during this early time in the campaign. This is likely to undercount support for candidates with this kind of support."

Non-registered voters are CERTAINLY not going to make it past ANY "likely voter" screen! If you're not even registered, how likely are you to vote? Not very.

Q: "Are you registered to vote?"
A: "No, but I'm probably going to register before the election."
Q: "Thank you for your time. No further questions."

You can argue that many former non-voters may register and vote later in the year, but then a poll of registered voters will catch them.

But, there is simply NO way to logically include people who "may" register and "might" vote if they ever decide to register -- for whatever reason.

"Anonymous said... The problem with using a "re...

2008-07-18T15:53:00.000-04:00

"Anonymous said...

The problem with using a "registered voters" screen for polling this far from the election is that some states allow very late or even same-day registration. This is an important bias for projections since students, minorities, transients, the poor, and first-time voters are likely to register late so they won't be tracked in polls during this early time in the campaign. This is likely to undercount support for candidates with this kind of support.

Most years this isn't much of a problem because turnout among the unregistered is so low but it may be a factor in this election, particularly with so many states reducing their registration requirements.
July 18, 2008 1:56 PM "

good point, and conversely, in swing states like Indiana and Georgia with onerous voter ID requirements , many first-time and late-deciding voters (like those nuns in their 80s) could be turned away. Wonder how that gets factored into LV models?

The problem with using a "registered voters" scree...

2008-07-18T14:56:00.000-04:00

The problem with using a "registered voters" screen for polling this far from the election is that some states allow very late or even same-day registration. This is an important bias for projections since students, minorities, transients, the poor, and first-time voters are likely to register late so they won't be tracked in polls during this early time in the campaign. This is likely to undercount support for candidates with this kind of support.

Most years this isn't much of a problem because turnout among the unregistered is so low but it may be a factor in this election, particularly with so many states reducing their registration requirements.

WOWIf 12% more republicans have given this electio...

2008-07-18T14:22:00.000-04:00

WOW

If 12% more republicans have given this election more thought than 2004 and are 16% MORE dissatissfied than in 2004 then this election is over barring a MAJOR (10x Rev Wright) political earthquake.

"83% of Republicans and 66% of Democrats now claim...

2008-07-18T14:06:00.000-04:00

"83% of Republicans and 66% of Democrats now claim that they are "certain" to vote in November. This is probably how ABC got it's likely voter screen. Among registered voters, they have 37-27 Dem-Rep. But among likely voters, it's 33-32.

This is why I HATE likely voter screens being used by different pollsters! It makes it almost impossible to tell anything about what's going on! It becomes like comparing apples and oranges! Honestly, who knows who's likel to vote in July? We know that registering to vote tends to mean that people end up voting, but exactly HOW likely?

We've seen lots of new registrations for Democrats, which SUGGESTS that there are lots of new voters who won't necessarily be caught by tight voter screens in July, but how can it be PROVED? It CAN'T!

Flat Fact: We Just Don't Know!

NATE has talked about this problem, the difference between "precision" and "accuracy." Using a likely voter screen probably increases precision, i.e. the probability that you can spot trends from poll to poll by the same company and making the grouping tighter.

But, it is wildly inaccurate to apply a likely voter screen in July. Who the hell knows who's going to vote in November? Thus, apply the wrong screen and you make ALL the results of your polls essentially meaningless.

We know old people are going to vote because they've voted before. Are young people going to turn out in equal numbers? Maybe not! If they don't then the race will be very tight. If they do, then it's a blowout for Obama.

Worse, since other pollsters apply different screens, we it's much tougher to compare polls by different companies.

It's much better to publish only the Registered Voters. That's gives us an Obama +8 lead which is pretty constant (somewhere between 8% and 10% of registered voters).

Is that really accurate? Of course not! The race is definitely closer than 8%, but is it 3%? 4%, 5%? Who knows? It all "depends" on who you think is likely to vote. That's why Nate's model is so interesting, even if it is UNPROVEN!

Sometime in late September, maybe October this question ("how certain are you to vote?") becomes meaningful, but young people are DOING STUFF in July!

Americans, even ones who say they are "following the election closely" probably are NOT really doing anything of the sort.

Everybody who reads or posts on this site is a FREAK! We're WEIRDOS who actually enjoy talking about politics and analyzing polling in the middle of July! That Ain't Normal people!

It's asking a LOT of voters to be ready to vote in July.

If it were easy to predict who's likely to vote and how the election will trend by November, Nate's model would be a LOT less complicated.

You'd just take the registered voter totals, add in the likely voter screen and VOILA! You've got your result.

Some sites actually attempt this based entirely on some arcane technical formula based on computer trading markets and expert opinions: http://www.forecastingprinciples.com/PollyVote/

http://Polyvote claims to have had a "near perfect prediction of the 2004 election."

"Polly says McCain trails at 47.7% vs. Obama 52.3%.
Polly sees the Democrats ahead by 4.6%."

I certainly won't argue for Pollyvote, but they have as good a chance of being right as the Washington Post Likely Voter Screen, and in 2004 they were a LOT better at predicting the actual outcome of the election than Gallup!

One more thing: check out the "movable" statistic...

2008-07-18T14:01:00.000-04:00

One more thing: check out the "movable" statistics on page 6 of the ABC/WP poll.

Most movable
White Catholics: 39%
Clinton supporters: 36
Independents: 35
Moderates: 35
Moderate/liberal Republicans: 42
Whites: 34
Married women: 33
Unmarried men: 31
Age 18-29: 31

Least movable
Blacks: 7%
Liberal Dems: 12
Obama primary supporters: 17
Conservative Republicans: 19
Liberals: 20
Age 65+: 22
Conservatives: 22
Single women: 23

I would love to know how these match up against the variance of the coefficients in the regression model.

Wow, some quality posts in this sleepy little thre...

2008-07-18T13:55:00.000-04:00

Wow, some quality posts in this sleepy little thread.

I think this has been asked before, but I don't remember seeing an answer: Does anyone have any data on how much of an impact GOTV efforts have? I'd imagine that very few likely voter models take them into account, but it would be fun to incorporate into any turnout model.

Turnout, if I remember correctly, was the one area Nate struggled with during the primaries. It would be fun, and useful, to think of a way to model turnout for the general.

Frank, good point about the Sunday polling. I wonder if ABC tried to weight its poll to correct for demographic differences among the days; I really wish these pollsters would offer more transparency.

I think the fact that young voters are more knowledgeable about key issues is fascinating. With all the focus on the internet as an organizing and fund-raising tool, I think its role as content provider has been ironically overlooked. I wonder to what extent young people are more well-informed because they are more likely to fact-check online.

michael,As a person who posts Anonymously here now...

2008-07-18T13:46:00.000-04:00

michael,

As a person who posts Anonymously here now and then, I'd like to mention that if it really bugs you, you can always leave. Maybe all the posts should be anonymous, and you can choose to sign your name in the post instead. Would that be better? Now why don't you quit being a control freak and start talking about polling. Thanks.

-- Anonymous

Dale Petrie's post is on target, for sure. Obama ...

2008-07-18T13:45:00.000-04:00

Dale Petrie's post is on target, for sure. Obama is a new phenomenon and the McCain has the truly difficult task of making him seem at once like something dangerous, risky and exotic and, at the same time, a conventional, hustling flip-flopping politician. No mean trick, and I don't see the McCain crew as having the skills to pull that off. Petrie's main point is worth reiterating. The voter models do nothing to account for the MASSIVE increase in turnout from Blacks and youth, as well as the underreported story that Mccain is suffering a 15 point gender gap and a 40 point gap with Latinos. If Af-Ams makeup 18% of the vote (for instance) rather than last time's 13%, and go for Obama 95-5, both of which are highly probable, that makes it game over. Af-Ams, youth and Latino are almost always underpolled, due to the landline issue, whereas as stay-at-home white seniors are pretty reachable on their landlines.

Obama has two challenges: His (1/2 race) and the Muslim smear. My hope and belief is that the folks who will not vote for him because of one of those two reasons would not vote for a Dem anyway, and will be more than counterbalanced by the new and enthusiastic support from young, Af-Am, and Latino voters...

whispers -It's not unverified nonsense. This is a ...

2008-07-18T13:40:00.000-04:00

whispers -

It's not unverified nonsense. This is a quick and breezy summation; I didn't give you the details of the psychographic studies. Of course the model is a generalization - the neat little pigeonholes are a simplification.

That's what models are - they are a map of the territory, but they are not the territory. But they do capture some sense of different cultural values for folks raised in different times.

People make choices for a variety of reasons; humans' behavior is harder to explain than that of ants or carbon atoms. But don't attack a model as a simplification - because that is what models are there to do.

All the talk about voter turnout in historical perspective presents an assumption that the present election will follow similar patterns to previous elections - that's a model.

Your own explanation (about retired people having time on their hands) is also a generalization. But I won't attack it, because it alSO contains some truth in it. It is an incomplete model; it does not map all the variables. And neither did my generational culture model.

Interesting story, footsy. Neat how you pigeonhol...

2008-07-18T13:09:00.000-04:00

Interesting story, footsy. Neat how you pigeonhole entire generations into neat little stereotypes.

Older people vote at higher rates because they have more time on their hands, esp. retirees.

You can discard the rest of your stories as unnecessary and unverified hypotheses.

Anon @ 1:30, Modeler re. ABC poll: What I found in...

2008-07-18T12:31:00.000-04:00

Anon @ 1:30, Modeler re. ABC poll: What I found interesting on that poll was them reporting that "in this four-night poll, Obama held a much larger lead Sunday
compared with Thursday through Saturday results." (p.4, 2nd para). ABC just sees it as a an indication of the changeability within the electorate still being high.

However, another plausible explanation may be that there is a difference between people that are out on Friday and Saturday nights in mid-July, and those who spend these nights at home answering to polls. In other words - I suspect the ABC poll to have over-sampled seniors, people with week-end jobs (e.g. retail, healthcare, taxi driving), and similar groups, and under-sampled groups like college students, or families with kids that have taken a full week off after July 4th. So, it may well be that the sample structure was completely different from March (where college was still going on, and the weather probably less attractive for spending a week-end outdoors), and ABC was actually comparing apples to pears. I think, a few more polls will be needed to find out whether the youth (as reported by PEW) or the seniors (as reported by ABC) will be more likely to vote in November.

Pete Kent: "The trip to Europe will seem odd – a Presidential candidate stumping in Berlin? Wait and see how that spins!" I think you are missing some important points here. First of all, there are some 20,000 American citizens living in and around Berlin (which gave Obama something like a 70-30 vote in the local Democrats Abroad primary), and Germany still hosts the largest number of US soldiers in the world after the USA and Iraq.

More importantly, however: According to the German Foreign Affairs Office, since WWII some 18 million US soldiers plus there families have been living for an extended time in Germany. Many of them have made friends here, and a good number have taken home a German wife. In the 1950s, nearly 600,000 Germans immigrated to the USA, plus some 200,000 in the 1960s.

Let's, for simplification, assume that 450,000 of the 1950's German immigrants to the USA were female, in average 22 years old by the time of immigration (i.e. born between 1938 and 1948), and still alive (the other 150,000 either being male, or having died in the meantime). That would mean they account for something like 5% of all American women between 70 and 80 years.

These women are likely to still have friends and relatives in Germany, with whom they regularly correspond and/or visit. A few personal examples: Friends of my parents-in-law, who reside in Texas, just wanted to pay us a visit a few weeks ago, but for health reasons decided to only visit their relatives in Munich and skip the travel from there to Hamburg. My god-father, living near Boston, has visited us something like three times over the last five years. My god-aunt is regularly informing me about the well being of her two sisters (one in CA, one in NJ), both of whom I visited with my parents when I was a child.

In other words: An Obama speech in Berlin, which succeeds in further fueling the already high enthusiasm for his candidacy among Germans, has enormous potential for reaching back into McCain's core demographics, via German seniors talking / phoning / e-mailing to their siblings / cousins / school friends that are now living in the USA. No other location in Europe, with the possible exception of Rome, could provide such a multiplier effect.

As to the event itself: As it appears now, the speech will be held at the Berlin Victory Column, which is something of the equivalent to the Washington Monument, and located opposite of the Brandenburg Gate in some 1,3 km distance from it (provides for nice backdrops of the gate, while allowing Mrs. Merkel to save face). Check out for details:

http://www.ypfp.org/obamas_berlin_speech_can_he_have_his_cake_and_eat_it_too

The last foreigner who spoke publicly in open in Berlin was the Dalai Lama this Mai - he attracted a crowd of some 25,000. Public viewing of the European Soccer Championship this summer attracted some 200,000 people, and the Love Parade, which until 2006 used to
take place in mid-July(http://en.wikipedia.org/wiki/Love_Parade) attracted up to 1.5 million people.

This gives you something like the lower and upper ceiling of the attendance you may expect. My personal estimate is at least 150,000 listeners. If Obama's team can (with the short time still remaining) organise some live music and performance ("Die Ärzte" would be the obvious choice; http://en.wikipedia.org/wiki/Die_%C3%84rzte) and turn it into a Love Parade-like happening, you may even get more than half-a-million people there, and possibly live coverage on MTV. In terms of imagery and PR, it will be a knock-out, even if the American punditry may criticise it for location and content.

Last but not least: I think there is a fundamental difference between an "Atlantic" candidate like Kerry (partly grown-up in Berlin, based in MA) and a "Pacific" candidate like Obama (Indonesia, HI) devoting time to Europe. In the case of Kerry, it served for re-inforcing his (narrow) "Atlantic" / New England background. Obama's profile, OTOH, will be broadened further, adding an "atlantic touch" to his existing "Mid-West to Pacific appeal".