Get yer polls, bets, numbers & pretty graphs! Elections 2008

... and sorry for going all 2004. Mostly reading along on the other interesting stuff all you people are posting here!

0 Replies

Sorry nimh, but one more reference to maps for oe, the following shows how income follows the red blue state thing, the more wealthy voting Democratic.

http://taxprof.typepad.com/taxprof_blog/2005/05/median_income_d.html

Also another set of maps showing higher generosity mirroring red states than blue states. I guess blue states believe in government doing it, red states believe in charity?

http://taxprof.typepad.com/taxprof_blog/2004/12/generosity_inde.html

0 Replies

Back to 2007.

If we want to interpret correctly the Rasmussen Iowa polls, or any set of polls, we must look at the whole series, not at any two polls. Why? Because we must assume that any single poll can be outside the margin of error, while it's highly improbable that a series itself can be outside.
Besides, any serious series of polls does not use the panel method. So we don't see voters running to and fro the candidates, but different samples.

All this said, from that series one can safely tell that Huckabee had an early jump which he keeps, Romney and Paul are stable -but with very different levels of preference-, Giuliani and Thompson both fell and McCain is a question mark. The rest is mere speculation.

0 Replies

okie wrote:

Sorry nimh, but one more reference to maps for oe, the following shows how income follows the red blue state thing, the more wealthy voting Democratic.

http://taxprof.typepad.com/taxprof_blog/2005/05/median_income_d.html

No need to apologize, the subject matter is interesting. But the implication here is false.

The flaw of logic here lies in the assumption that if the wealthier states go blue and the poorer states go red, that means that wealthy people vote blue and wealthy people vote red. Sounds logical, but it isnt.

The equation of states with people is wrong because of what those "shades of purple" maps of OE's do show: that in reality, there are no "red" states or "blue" states that represent mere blocks of Republicans or Democrats; there are just variations of purple that represent states where there are lots of Democrats and lots of Republicans, even if there are more of the one than the other.

What's the relevance of that? Well, to make it more concrete: take California. Solid blue state, right? Yet almost 45% of the voters there - nine out of every twenty Californians - voted Bush in 2004. Now ask yourself, within California, who are those 45% of Bush voters, and who are the 55% of Kerry voters? Where do they live exactly? And which group is the more wealthy one, and which the poorer one? Consider, for example, that LA, the city, is wholly blue; while Orange county is a Republican stronghold.

Same issue plays across the country. The deep south is one of the poorest parts of the country. They are also majority-Republican voting states. But even in those states, some 35%-45% voted for Kerry. So even there, there are lots of Democrats as well as Republicans, and now ask yourself, within those states - who votes for which party? And which group is the wealthier one? In Alabama, do Democratic voters on the whole represent the poorer or the wealthier part of the population? Consider that the Democrats might take the odd enclave of higher-educated, higher-income urban voters, but that their bedrock of support in the state is the much larger African-American population.

Well, et cetera ad infinitum. There are lots of regional specificiies. In Idaho, Democrats have their best scores in the prosperous, urban parts of the state. But in most states, it's the urban city areas where the Democrats do best that are the poorest, while the Republicans are dominant in the more prosperous suburbs and exurbs.

Now luckily, we can skip this discussion altogether if the question is which party represents the wealthier or the less wealthy part of the population. We have no need to resort to geographical analysis of this kind to figure that out. Because political support by income level has been researched exhaustively, from simple opinion polls to demographically weighted exit polls to academic research. And the result, on the national level at least, is always the same. Not just are Republican voters wealthier on average, but there is a direct correlation. The wealthier the income group, the larger the share of support of the Republicans is.

The last presidential elections gave a perfect example of this, and I made a thread about it at the time. Let me repost the table with the data here:

2003 total family income:

Code:



                    % Total     Kerry     Bush     Nader 



Under $15,000           8        63        36        1 



$15,000-$29,999        15        58        41        0 



$30,000-$49,999        22        51        48        0 



$50,000-$74,999        23        44        55        1 



$75,000-$99,999        14        46        53        0 



$100,000-$149,999      11        43        56        1 



$150,000-$199,999       4        43        57        - 



$200,000 or more        3        37        62        1

As I wrote in that thread:

Quote:

[Of course,] a lot of non-rich people voted Bush too. And if you use Kerry's definition of rich (which indeed is not mine), an overwhelming majority of Bush's voters was non-rich.

But [although] Bush drew a lot of votes from the 'bottom half' of the income scale as well, Kerry drew more. In fact, if it wasn't for those earning over 150,000 $, Kerry would have won the elections. [..]

Doesnt make any difference in terms of [Bush's] legitimacy of course - rich people deserve their vote just as much as poor people ;-).

But [n]ext time anyone on this board is going to go on about those "liberal elites" and how it's the Republicans who represent "the common folk", I'm gonna remind them of that. The poll shows: the richer, the more likely to have voted Bush. [..]

Or, as I wrote more concisely in another thread:

Among all income groups below $50,000 family income, Kerry won.
Among all income groups above $50,000 family income, Bush won.

0 Replies

On the same count, it's worth mentioning that the pattern of people being more likely to vote Bush than Democrat, the wealthier they were, even increased in 2004 compared to 2000:

Quote:

Money Matters

[..]

For all the talk of how religious voters made Bush's victory possible, their performance didn't change from 2000 to 2004. Four years ago, those attending church once a week or more were 42 percent of the electorate and gave Bush 59 percent of their vote [..]. In 2004, these voters were 41 percent of the electorate and gave Bush 61 percent of their votes [..].

By contrast, Bush improved his performance with voters at the upper end of the income ladder. Among those making less than $50,000, Bush actually lost ground, as his performance fell from 21 percent in 2000 to 20 percent in 2004. Among those making over $50,000, Bush's performance jumped 3 points, from 28 percent to 31 percent. And most of this improved performance was concentrated among the wealthiest of voters, those making over $100,000. In this group, increases in turnout and support for Bush raised the president's performance from 8 percent to 10 percent. In fact, Bush's gains among the wealthiest Americans account for a good chunk of his popular-vote margin of victory. [..]

I'm not finished yet, either - don't you wish you hadn't started me, Okie :wink:

I did some more research when this subject came up on A2K later again. Lash brought it up, and quoted a Washington Times article. The article touted how among even modest-income middle class whites, the Republicans, "the oft-described party of the rich", were now outdoing the Democrats.

But the data it cited proved my point. They still showed that the wealthier Americans are, the less likely they are to vote Democrat - it's just that among whites, specifically, the tipping point where a Republican majority is reached comes already at a fairly low level. Bush had a majority among every income group of whites over $23,300. Kerry had a majority among whites earning even less than that. Eg, "middle income" white voters went for Bush -- just like the rich -- but unlike the poor whites, who stayed Democratic.

And that, of course, was specifically among whites. Count all American voters, and the income threshold was, as mentioned, $50,000 - all income groups above went for Bush, all groups below went for Kerry.

This is where I can return to that local level. I explained above why I dont think analysing the relation between average income and party preference on a state level makes any sense. Few states are homogenous in income groups, so if the rich in a poor state vote Republican, or the poor in a rich state vote Democratic, your state-level correlation between rich states and Democratic preferences is meaningless. But data by House districts is a lot more specific, and make for interesting reading:

nimh wrote:

Here's another way to look at the question (borrowing slightly from an earlier post of mine on the House & Senate races thread). Look here:

2006 Election Guide - Senate, House and Governors' Races.

Maps, maps, maps. It's a political nerd's walhalla. Above all, because they are interactive maps.

Not just can you immediately get an overview of where the poor or rich districts are, or where (for House races) the urban and rural districts are, or the majority black, Hispanic and white ones. You can also mix and match!

Thump a few check boxes, for example, and you'll get only the poor, white, rural Democratic House districts - because yes, there are still some, as it turns out. In fact, of the 8 districts listed as poor, white and rural, five are still held by a Democrat (though none voted for Kerry). [..]

OK, now take out the race and urban/rural factor, and simply look at where the low-income voters live:

These are the 22 House districts where the median income is lower than $30,000.

They break down as follows..:

18 Safe Democratic seats

4 Safe Republican seats

0 Replies

okie wrote:

Also another set of maps showing higher generosity mirroring red states than blue states. I guess blue states believe in government doing it, red states believe in charity?

http://taxprof.typepad.com/taxprof_blog/2004/12/generosity_inde.html

As for this map - touche. But, eh.. is that kind of comparison of state-level statistics on social phenomena really a path you want to go down? Take a peek at this old thread of mine: Red states, blue states, and the stats on moral issues :wink:

(Of course, the same warning I made above against using state-level data when the states themselves are demographically diverse does hold for the kind of stuff I put in that thread too..)

Thanks for posting this stuff, though, dont want to discourage you - it's interesting.

0 Replies

I think the red blue state phenomena mainly illustrates the voting patterns of people, proportional to their connection to the land and how they make a living. Not in all cases, such as Jackson, Wyoming where the elites live for example. But if you look at farming communities and towns and cities where their economies are mainly driven by agricultural and other industries tied to the land, you generally see a red area.

You will prescribe it as my bias, but I firmly believe red state people have a better and more intimate understanding of nature and how we are so dependent upon it, and they better understand where everything comes from, such as food, energy, forest products, etc. They therefore understand the realities of life, the economy, and other matters of survival, and tend less to live in a world of idealism where you don't need to drill for oil but somehow by some magical government program we can solve the energy crisis, just one example. Such people are Democrats. People that live in cities far removed from where everything comes from, they are insulated from some of the realities of the market, and sources of everything we all live on, and therefore vote for the idealism of Democratic policies.

Of course, these are generalizations not always applicable, but tendencies are at least explained by some of the differences of lifestyle and how people make a living.

It seems to me that your income table, nimh, supports the fact that the people most reliant on welfare programs, mostly in urban areas most likely, support Democrats, while the people that are paying the bill vote Republican. My mother used to say the Republicans were for big business, Democrats for the little guy, and she was solid Democrat, but I have come to realize without big business, the little guy is pretty helpless, and by the way she has not voted Democrat for president for a long time. The party has pretty much left her.

0 Replies

nimh wrote:

Code:Rasmussen - Iowa - Republicans Date Romney Huckabee Thompson Giuliani Paul McCain 12/17 27 28 8 8 6 14 12/10 23 39 8 8 5 6 11/26-27 25 28 11 12 5 4 11/12 29 16 14 15 4 6

The McCain upsurge could be a fluke, no other poll is showing it, though one is tempted to credit the Des Moines Register endorsement for it.

An update on the McCain thing, since two more new polls of the Republican primary race in Iowa have come out. One is by CNN, which hasnt previously done a poll for the race (so no comparative numbers), the other is by ARG, which last did a poll on 26-29/11. And as of the ARG poll, this is no longer true.

In fact, ARG outdoes Rasmussen and now polls McCain at 20% - which is more than double the level he was polling at in late November, when he had 9%. So that's a poll seconding the Rasmussen finding of a sudden and unexpected McCain surge, after a decline of many months.

On the other hand, CNN has him just at 9%, which is also higher than he usually gets (the pollster.com trendline has him at 6.6%), but nothing comparable. So wait and see, but something unexpected might be happening. The DMR endorsement after all?

Re Huckabee dropping back and Romney recovering a bit, the two new polls don't show that out much. They have Huckabee at 33% (CNN) and 28% (ARG). That is indeed a bit lower than he had in a flurry of polls in early December, but when you look at ARG's previous numbers for him (27% in late November, 24% in mid-November and 19% in late October) there's no drop-off. You might speculate that his rise is levelling off though.

As for Romney, CNN has him at 25%, which is pretty much in line with other polls, while ARG polls him very low, at just 17%, which is down 11 points compared to its late November poll. No other polls showing a stark drop-off for Romney though, so let's say the data is inconclusive.

0 Replies

More indications of a budding resurgence of McCain?

Only just saw this. There's two new polls out for the Republican race in New Hampshire too, and there McCain is pretty clearly moving up. So perhaps there is a broader context to his possible (relative) rise in Iowa than just a Des Moines newspaper endorsement or a statistical fluke - perhaps there's a bona fide trend?

In New Hampshire, in any case, both new polls - by ARG and Rasmussen - have him significantly higher than he's been doing previously. ARG has him equal with the long-time favourite, Mitt Romney, at 26%! Rasmussen has him just 4 points shy of Romney at 27% vs 31%.

The two polls before that, by CNN/WMUR/UNH and Fox, also had McCain in the twenties (22% and 20%). That was already a departure from earlier numbers: in the month before, McCain had scored 20% or more in just 2 out of 14 polls, and in the previous four months, in just 1 out of 19 polls. And now four times in a row.

His budding change in fortunes sort of shows up in apple-to-apple comparisons too:

Code:

John McCain moving up in New Hampshire?



                 ARG    Rasmussen   CNN/WMUR   Fox



mid-December     26%       27%         22%     20%

early-December             18%         19%

late November    11%       15%                 21%

mid-November                           18%

early November             16%

late October     17%       16%

0 Replies

It makes sense.

McCain is kind of the known quantity. Not perfect, but has his good points, and known. Giuliani came up as this tough, charismatic, Dem-beating option. But he just couldn't get through the scrutiny. Huckabee is kind of the anti-Giuliani -- an actual religious conservative, Southern, aw-shucks, etc. But he's also just not standing up to closer scrutiny, once it started.

Romney is trucking along in the background as someone with impressive resources who many people just don't quite trust.

Who's that leave? McCain.

0 Replies

Anyone remember what McCain and Obama promised to do if they each became their party's nominee? I thought of it because it seemed so unlikely at the time that it would be possible -- that both McCain and Obama would be their party's nominee -- and now looks like it actually could happen. Don't remember what the promise was though. (Something about funding...?)

0 Replies

okie wrote:

It seems to me that your income table, nimh, supports the fact that the people most reliant on welfare programs, mostly in urban areas most likely, support Democrats, while the people that are paying the bill vote Republican. My mother used to say the Republicans were for big business, Democrats for the little guy, and she was solid Democrat, but I have come to realize without big business, the little guy is pretty helpless

That's an internally consistent perspective. I disagree with it, of course :wink: . But it has the virtue of having a link with the actual numbers on voting patterns. In that sense, it's already a lot better than the implication suggested by the "Taxprof" blogger with his maps: that the Democrats are somehow the party of the wealthy, the elites, while the Republicans are the party of the common folk. Such an assertion does not hold up when you add more data.

Regarding the notion of 'welfare voters', though, I do disagree. I mean, going back to those data by income level: among all groups under $50,000 income, a majority voted for Kerry. Those groups represent 45% of the US population Exclamation

. That's hardly just welfare families.

Now for sure, the divide between Kerry and Bush voters in the $30-50,000 bracket was very close: 51%/48%. But Kerry won the voters under $30,000 in income more handily, with some 60% of the vote; and that's still 23% of the population. A quarter of the population - thats hardly just the welfare class. That's a lot of working poor and struggling middle-class families.

0 Replies

This seems to be what I was thinking of:

Quote:

Both Obama and McCain have long supported public financing of elections. McCain has a strong record on supporting campaign finance reforms, a position which has not been popular amongst some of his GOP colleagues.

McCain's manager said yesterday that "Should John McCain win the Republican nomination, we will agree to accept public financing in the general election, if the Democratic nominee agrees to do the same."

http://2008obama.blogspot.com/2007/03/mccain-joins-obama-on-campaign-funding.html

0 Replies

McCain will NEVER get the GOP nod.

0 Replies

sozobe wrote:

Romney is trucking along in the background as someone with impressive resources who many people just don't quite trust.

Who's that leave? McCain.

True. I was thinking the race would basically become Romney vs Huckabee, with McCain as a kind of dark horse possibility. But he might be more prominent than that, creating a true three-horse race.

I do think Okie's right that he's got a long way to go though. On the Republicans thread, Okie said "his overall potential still may be limited," and I think that could be right. What stands in the way of him being a safe haven for last-minute doubters, like Kerry was for the Democrats in Iowa in '04 when the Dean campaign imploded, is that so many core conservatives are resentful of him for a number of reasons.

At least I hope so - if one is to accept the 'match-up' polls that measure how the different Democrats would perform against the different Republicans, McCain is by far the strongest Republican. Stronger than Giuliani and a lot stronger than Romney or Fred Thompson (jury's still out on Huckabee's post-surge performance).

0 Replies

nimh wrote:

The last presidential elections gave a perfect example of this, and I made a thread about it at the time. Let me repost the table with the data here:

2003 total family income:

Code:% Total Kerry Bush Nader Under $15,000 8 63 36 1 $15,000-$29,999 15 58 41 0 $30,000-$49,999 22 51 48 0 $50,000-$74,999 23 44 55 1 $75,000-$99,999 14 46 53 0 $100,000-$149,999 11 43 56 1 $150,000-$199,999 4 43 57 - $200,000 or more 3 37 62 1

In your discussion of the income of Kerry's voters, you didn't specifically mention the under $15,000 group which does indeed show Kerry's highest percentage of support, which I think does indicate I was correct to conclude this group that comprised 8% of the voters to very likely include the most people that pay no income tax, and receive more welfare programs. It could also include more young people that have just been indoctrinated by the educational system to vote Democrat. Later when they get a career and become exposed to life, they may tend to become more conservative. Just a theory.

0 Replies

sozobe wrote:

This seems to be what I was thinking of:

Quote:
Both Obama and McCain have long supported public financing of elections. McCain has a strong record on supporting campaign finance reforms, a position which has not been popular amongst some of his GOP colleagues.

McCain's manager said yesterday that "Should John McCain win the Republican nomination, we will agree to accept public financing in the general election, if the Democratic nominee agrees to do the same."

http://2008obama.blogspot.com/2007/03/mccain-joins-obama-on-campaign-funding.html

That would be nice...

I think an Obama/McCain general elections race would be a big relief compared to previous races, and might do the political culture of the country a lot of good.

It's funny how either party's choice looks different depending on whom you imagine the opponent to be..

For example, I think that Obama, with his visceral distaste of both the more plebeian and nasty sides of campaigning, would be well suited as Democratic candidate against McCain. Against McCain, Hillary might look shrill or mean-spirited or the like, so Obama has an advantage there.

Against Giuliani, on the other hand, the race looks completely different to me. Obama would be eaten alive by Rudy; against a candidate like that, you need a calculating, determined, tough-as-nails Democrat.

A truism, I suppose: in a fight, you can only be as nice as your opponent is - go much beyond that, and you'll be killed...

The other Republican candidates I see as having some in-between effect.. wouldnt it be nice if you could let the other side choose its candidate first? :wink:

0 Replies

Totally!

0 Replies

In the Obama '08 thread, especially, there has over time been a lot of discussion about Hillary Clinton's negatives. In polling terms, I mean: her high unfavourability ratings, the high number of people who say they absolutely won't vote for her, et cetera.

If I can roughly paraphrase, one school of thought emphasises that her unfavourability rating is anywhere between 40% and 50%, which leaves little space to get elected, as even winning over everyone who is neutral about her wont be enough.

The other school emphasizes that the number of people who view her strongly unfavourably, or who say they will under no circumstances vote for her, is more equivalent to the core Republican support - say, 30%-40%, and that the rest can be persuaded still.

That latter argument is fair enough: someone who tells a pollster that he has a "somewhat unfavourable" opinion of Hillary is hardly by definition lost. The campaign season will see an onslaught of political arguments and media coverage, and plenty of opportunities to change some minds.

In the case of Hillary, that argument is weakened, however, by the fact that people know her already very well - or at least think they know her very well already. She's been prominently in the public eye for a decade and a half, after all.

In turn, however, Blatham especially has emphasised that a large part of the negative impressions of Hillary are the result of a sustained propaganda campaign, perpetuated by a barrage of media narratives and conservative talking points rolled out over time - which would still catch up with any Hillary rival if he were nominated instead as well.

Voters dont really know her, this argument goes - a large part of who they think they know is a caricature, and now is the time on Sprockets when people get to find out that she's not actually a fireblazing witch, but in fact quite a sensible politician and an actual human being. Especially the sentiments of the group that feels only moderately negative about her are only skin deep and can thus be corrected.

That argument raises the question: has this unfavourable impression of Hillary shown much flexibility over time? Has the proportion of people with a negative view of her varied much over the years as the propaganda for and against her had its ebbs and flows?

I started wondering about this while reading the latest NBC/WSJ poll, so I used that poll's history to chart it out. But feel free to point to the data from any other poll if it shows a different picture. This, however, is what the NBC/WSJ polls have shown:

http://img530.imageshack.us/img530/5268/hillaryfavratingsjo3.png

As you see, a weakness in this chart is that polling on Hillary was strongly concentrated in 1999-2001 and 2006-2007, with just three polls in the middle. But the impression is striking nevertheless.

Public perceptions of her in the past year or two is pretty much exactly where it was back around Y2K. Moreover, it has shown little development so far as her presidential campaign has received more and more coverage. The only thing you can say is that both the number of people with strongly positive feelings and with strongly negative feelings has increased, with the number of those feeling neutral dropping. The overall balance is stagnant.

The only weird interference is back in early 2001, when two subsequent polls diverged from the average, the first one being more positive for Hillary, and the second one more negative than all the other polls. Speaking instinctively, I'd blame random statistical variation for that.

0 Replies

If the "Somewhats" and "Neutrals" split down the middle, it looks like Clinton loses narrowly. And if alot of them stay home, she could lose worse.

The Very Negative band seems to be rising a bit faster than the Very Positive of late.

0 Replies

Get yer polls, bets, numbers & pretty graphs! Elections 2008

Related Topics

Quick Links

My Account

able2know