There’s not supposed to be a national debt. When Ronald Reagan ran for president in 1980 he said that tax cuts generated so much economic growth that it would be possible to cut taxes, raise defense spending, and balance the budget without cutting middle class entitlements. All we would need to do would be to cut, “waste, fraud, and abuse.”
I went to the trouble of formatting a table with the relevant columns and putting it in my first message, then I linked to the entire table in that post, and then I linked to the entire Pew report in subsequent posts, and linked to a graph of the data in yet another post.
How do you know it’s wrong?
Pew is a respected sociological polling firm, widely used by the press and people on both sides of the political aisle. Exit polls can be skewed for all the reasons I posted, and are not typically used for purposes other than studying voters and their patterns. Your data doesn’t even mention Libertarians, which is what we were actually discussing. My comments about Republicans and Democrats were actually a digression from the topic.
You still have to make the gist of your argument here in an intelligible fashion.
Because there’s a wealth of polling data on the other side. If it’s one survey v. many, I’m placing my bets on the many, as any self-respecting statistician should until given a damned good reason to the contrary.
Nonetheless, if Pew is wrong about liberals and conservatives, it calls into question their other results in the same survey.
I went back to the Pew survey, and looked at how they’d partitioned their universe. All I can say is, their way of looking at and making these divisions is quite debatable. I have no qualms with the survey data, in terms of their ideological partitioning, but I’d question the partitioning itself. I don’t think they’re analyzing things in the right frame.
My argument was pretty simple: In specific, that it’s not true that the majority of libertarians are ‘at least’ upper middle class. I then generalized it to point out that it’s also not true that Republicans are wealthier than Democrats. The Pew data shows that there are significantly more Republicans in the lower income strata and more Democrats that are upper middle class or higher than Republicans.
Is that true? Can you cite some other polling data other than your exit polls? You might be right - I’m not here to apologize for Pew. I cited them because they’re a perfectly respectable firm to cite, but it’s possible that they are wrong or that their poll was an outlier or a non-representative snapshot in time. I’m open to seeing other data.
This may be the best argument against their data - if they don’t have the divisions correct, then they could be smearing results across categories. Again, the report itself doesn’t go into a lot of detail about their methodology, so I don’t know if that’s true or not. It depends on how they derived those categories. Did they build them up by inference, by examining the responses to other questions, or did they present those categories directly and ask respondents to pick one? I don’t know.
No, it doesn’t. I’ve been through it, and it doesn’t say that anyfuckingwhere. So quit saying it does, or quote the exact language here. You won’t, because you can’t find it, because it doesn’t fucking exist.
Here’s a Quinnipiac poll of registered voters from last month. Question 1, broken down by income: people earning <50K favored Obama’s re-election in 2012 by 39-34%, people earning between 50-100K favored the generic Republican by 40-36%, and people earning over 100K favored the generic Republican by 43-36.
For the GOP, things went from -5 to +4 to +7 as you went up the income ladder.
I could get more, but for some reason my browser isn’t showing the links at pollingreport.com as links, so I’m having to Google polls, which isn’t as easy as you’d expect.
Hell, it’s your cite, that you’ve been using as the primary support for your argument.
You might want to read it.
:headdesk:
Be careful of the word ‘cut’. Here in MN people use the word ‘cut’ when they are actually having a larger budget.
I love your evil mind! Let’s do this!
Here’s the relevant part of the table again:
Income Level Libertarians Conservatives Liberals
$75,000+ 31 21 29
$50,000–$74,999 10 14 16
$30,000–$49,999 18 26 20
$20,000–29,999 14 15 12
<$20,000 14 16 13
Note: 21% of Conservatives are in the $75,000+ category, and 29% liberals.
In the top two quintiles, liberals outnumber conservatives 45% to 35%
31% of conservatives are in the bottom two quintiles, vs only 25% of liberals.
Unless you’re quibbling about “Republican” and “Democrat” vs “Conservative” and “Liberal”, doesn’t that pretty clearly show what I was saying?
Well, as much as I’m equating “Conservative” to “Republican” and “Liberal” to “Democrat”, you’re doing much worse - you’re using election polls, at a time when the Republicans haven’t even put up a candidate. You’re also using registered voters as a proxy for the population at large. There’s plenty of skewing that can be going on in an election poll - plenty of Republicans voted for Obama, and currently plenty of liberals say they won’t. Also, this particular election is skewed by big labor issues, which is going to pull in more blue-collar people for Obama. The Democrats are also very good at ‘get out the vote’ drives and voter registration, especially among the poor. That’s what ACORN was doing - registering primarily poor voters from the inner cities, where they are likely to be heavily skewed Democrat.
Let me show you one major way in which election polling can skew towards poor Democrats: Poor Democrats are more likely to benefit from government programs than are poor Republicans. Therefore, they may be more motivated to vote - especially in a year when the main issues are where and how to cut government spending. Also, if farm subsidies are at risk, that may pull in more rich Republican farmers. The Tea Party is a major factor in Republican politics this year, which may skew the voting data. I’m not sure in which direction, though.
I’m not saying the polling data is necessarily wrong, I’m saying that there are plenty of ways that it could be skewed to be non-representative of the population at large.
Remember what we’re ultimately trying to ask here: Does income make you more likely to be in one party or another? Does being wealthy tend to make you Republican? Do wealthy people vote more against government because they need it less? Do poor people tend to be liberals because liberals protect the poor? Or do poor people tend to be more conservative or libertarian because big government often works against them and for the middle class? Or what?
These types of sociological questions are very different than election poll results. To give yet another example of possible bias, consider that really disenfranchised communities may not bother to vote, yet they may be disproportionately liberal or conservative.
However…
I feel for you. I’ve been trying to do the same thing, and it’s very difficult to find data other than that Pew study and past election poll results. What makes it worse is that partisans on both sides have reasons for spinning the data, sometimes in the same direction. Some Democrats want to show that wealthier people are Democrat, bacause Democrats are smarter, more educated, etc. Some Republicans want to show that Republicans are richer, because they work harder, are more entrepreneurial, yada yada. So you can find spin in both directions, by both sides.
Heritage broke down electoral districts in the 2008 election and found that the majority of the wealthiest districts voted Democrat. They used that as evidence that Democrats are rich metro types, and Republicans are poorer, ‘salt-of-the-earth’ types. I find that questionable (just because a majority of blue districts are wealthier doesn’t mean that the people in them who vote Democrat are the wealthiest). And in fact, I found several reference to a paper critical of Heritage’s conclusion - but in each case the reference led to a dead link. So I can’t find the actual paper.
You’d think there would be more and better data on this stuff, but apparently there’s not. Maybe it’s because neither side uniformly has a vested interest in the outcome, so no one bothers to go looking. I don’t know.
By the way, why don’t you try to be pleasant while having this discussion? It should be clear to you that I’m trying to debate this in good faith, with the best data possible. I don’t really have a vested partisan interest in the outcome: I’m not a Republican, and I’m not a class warrior so I don’t particularly care who’s rich and who’s poor. I jumped into this fray because the claim made by China Guy was more than that: He was basically saying that if you’re a libertarian you’re likely to a rich person. That’s simply not true and I took exception to it and backed it up with the Pew Poll.
Now we’re digressing into general demographic trends of all the parties. That’s fine, but it’s a complex issue and there’s a lot of conflicting data. It doesn’t help anyone to enter that kind of discussion by hurling insults at the other side and assuming everything they say is a lie. Try to be a little more open minded.
No, it doesn’t, because the differences you are pointing out are exceeded by the margin of error of the sample; there is 13% of Libertarians unnaccounted for, 10% of conservatives and 8% of Liberals. It is a very sloppy poll at best, and I am surprised you keep harping on it. It does nothing for your argument.
I don’t know what you see, but my screen doesn’t show the ‘Republicans’ and ‘Democrats’ columns that you refer to.
Fail.
In this case, it’s far from a quibble. Their 33% of Americans that they categorize as conservative or liberal leaves something to be desired compared to a nation where 20% of the electorate, at absolute most, is in play, and that’s only under landslide conditions. Realistically, less than 10% of the electorate isn’t pretty solidly in one camp or the other.
So we’re talking 1/3 v. 9/10. You bet I’m ‘quibbling’ over that difference.
That should, if anything, increase the validity, since we’re talking party ID, not personality.
No, I’m using party preference as a proxy for the more conservative and more liberal segments of the general population. People who aren’t even registered are largely apolitical, so this makes complete sense.
Sheesh.
First of all, the fact that the numbers don’t add up to 100% is not a ‘margin of error’. We don’t know why they don’t add up to 100%, but Pew is careful with their work, so I would assume they accounted for it. For example, if the missing numbers were due to unanswered questions on the form, and they did a statistical analysis of the gaps and found no skewing, then the only effect of the spoilage would be to reduce the sample size, which would increase the margin of error but only by a very small amount. If all the surveys with missing income data came from high income areas, I would assume Pew would mention it or correct for it in some way.
Second, the reason I brought it up again was because I had to answer RTFirefly’s statement that the Pew report did not say that Republicans were under-represented in the top quintiles. I quote: “No, it doesn’t. I’ve been through it, and it doesn’t say that anyfuckingwhere. So quit saying it does, or quote the exact language here. You won’t, because you can’t find it, because it doesn’t fucking exist.”
I found that rather confusing, because to me it did pretty much say that.
However, I did just find a rather large difference between the Pew Poll and RTFirefly’s election polls: I was using 'Conservative" and “Liberal” as roughly equivalent to “Republican” and “Democrat”. But they’re far from it in this survey. I just had a look at the “Values, Issues Positions and Politics” table, and it shows a wide overlap between party affiliation and these categories. So if that’s what he was getting at, then he’s correct.
For example, 50% of libertarians in the poll are Republican or lean Republican, but 41% of them are Democrats or lean Democrat. I found that surprising, so I dug deeper to see how they defined “Libertarian”, and it’s generally anyone who claims to be fiscally conservative and socially liberal. That’s a pretty expansive definition, and would include ‘blue dog’ Democrats and Republicans like Olympia Snowe or independents like Joe Lieberman. Likewise, 22% of the conservatives are Democrats or lean Democrat, and 12% of the liberals are Republican or lean Republican.
That means we’re looking at an entirely different way of breaking down the population as compared to the election surveys, and they really can’t be compared.
Easy, sparky. I just simulposted the same observation.
And you don’t find that disconcerting? I guess I just have higher standards than you do.
I guess that’s the closest to a mea culpa you can ever get to. Again, would it kill you, Sam, to actually spend a couple of minutes checking out your own cites for a sanity check before you post? You make us dear readers do that and then get in a snit when said cite smells fishy and makes you look like cherry picking data.
A mea culpa? Give me a break. The whole thing about Republican vs conservative and liberal vs Democrat was a side argument having nothing to do with what I originally posted that cite to refute. And, it doesn’t disagree with what I said - it’s not sufficient evidence. So the relative wealth of Democrats vs Republicans is still up in the air. We wouldn’t even have had that discussion had I limited my cite to libertarians and not thrown in the sideshow about Republicans and Democrats.
But the whole point to the original cite was to refute YOUR uncited claim that most libertarians are at least upper middle class. The PEW poll DOES have a column for ‘libertarians’ so it’s still on point with respect to your claim. We can debate whether it’s too expansive a definition or not, but it’s still evidence (although not concrete evidence) that your statement was false.
So now it’s your turn: Put up some data backing up your own assertion, or issue a retraction. I’m required to cite every claim I make, so you are too. Fetch it up.
Yeah, you just consistently espouse Republican positions that aid the wealthy at the expense of the poor and weak.
You can’t expect the fact that you’re a Canadian to suddenly get you noncombatant status in the American political wars, after you’ve been arguing hard for one side in those wars for the past dozen years. For all intents and purposes on this board, you’re a committed Republican partisan who by happenstance finds himself north of the 49th parallel.
Look, you’re the one who relied heavily on a particular cite that you’d apparently not even skimmed for the highlights, let alone understood. As a Doper, I refuse to have any open-mindedness about that. It makes me tired and grouchy, and makes me point out that no, your cite doesn’t say that anyfuckingwhere.
Nice of you to acknowledge this, sort of, after the fact, but it’s a bit of a recurring theme with you lately. I know it’s hard to resist when somebody is wrong on the Internet (wonder if we could talk Randall Munroe into letting that panel be the official GD logo), but maybe you should try to argue well in half as many threads or something, so that you’d have time to understand your own cites, and hell, your own arguments.
I mean, that handwaving argument about how exit polls potentially could be biased applies to any conclusions based on a statistical sample, period, including every Pew survey ever. When you don’t know you’ve made an argument claiming the invalidity of not only your big cite, but the entire field of statistics, you’re spread too thin.
Side argument or not, it’s not up in the air. Ultimately, what it means to be a Republican or a Democrat is to vote for Republicans or Democrats. And we know what’s going on there.
There are really only two ways the exit polls can be screwed up re income: either affluent Democrats or poor Republicans are reluctant to respond to exit polls.
I’m gonna toss the first one out, out of hand. There’s a reason why the whole ‘effete liberal’ meme caught on, and that’s that we affluent liberals are visible. As a group, we don’t hide our lamp under a bushel.
So it comes down to whether poorer Republicans are underrepresented in exit polls.
Even supposing such an underrepresentation, certain recent elections should have compensated for it by bringing out more lower-class Republicans. I’m referring to 2000 and 2004. In both elections, both the GOP and the media did a bang-up job of painting the Democratic candidate as an out-of-touch elitist, telling white lies (Gore) or flip-flopping (Kerry). Meanwhile, in 2000, Bush was the guy you’d have a beer with, and in 2004, he was also the guy leading our wars and keeping us safe from the terrorists. Plus if there was ever an election where white working-class conservatives were fully mobilized and turned out on the basis of social issues, 2004 was that year.
So what happened?
In both cases, as you go down the income ladder, you get way more Democrats, and you get way more Republicans as you go up it.
Nothing seems to change that, even the things that you’d think would have to if anything would.
IMO, ceding the definition of Libertarian et al to the model used by the Pew Report cited here is a mistake:
What ties this model to actual behavior? They created these survey-defined groups, give them labels that seem to correspond to political parties, but when you look at the numbers it doesn’t add up. They can’t identify 42% of the survey respondents, but we know there isn’t 42% of the voting population in the “middle”.
We already know that polling for party (or at least liberal-conservative) identification and issue identification gives dramatically different results.
I spent probably an hour looking up your cites, reading your source material and searching the internets for some kind of data. Libertarians evidently don’t show up on most studies of income breakdown or at least I can’t find it in my time. My bad.
That said, as a vociferous self styled libertarian maybe you would actually be more familiar with some libertarian for dummies facts and figures and be able to provide the same. I mean, I spent much more time reading your cites than you evidently did. And have probably spent more time trying to find a demographic income breakdown than you have.
Your turn to fetch it up and please read your cite first