I don’t want to be at the center either. I’m very much on the left (liberal more than socialist, anti-authoritarian would be most accurate). But I do concern myself with how things play in the center, and I am convinced that one of the key ways our politics gets pushed rightward is by the media’s blatant and illegitimate claims about where the center is.

I would like to combat that. My goal is to pull the frame of our debate leftward until the real brownshirt, racist crap on the right is off the table, and some sensible and popular social democratic and left libertarian ideas are back (or newly) on it.

That isn’t what I meant.

I don’t know what your exact political positions are, but everything I’ve read from you here and in the past suggests that you are crammed over on the left side of your graph like almost everyone else who’s worth reading for information (as opposed to enemy tracking).

The point is that your graph and the decisions you made to create it, has a rhetorical effect, whether you intended it or not. And whether or not those questions were the right choice for the problem you were trying to address as a researcher.

That rhetorical effect is to implicitly accept the beltway frame about where the edges of debate are.

Can you not see that at first glance, this picture is not saying “readers of left blogs mostly follow a coherent set of positions” but instead “readers of left blogs are almost all at the far left of the spectrum, while readers of right blogs stretch evenly from moderate to right wing”? Clearly your full text does not support that interpretation, but because of the labels you’ve chosen, it’s what the picture says to anyone who doesn’t read the whole thing.

It looks just like something some right-wing hack would link to as “evidence” that left bloggers are a bunch of radical DFHs. That would intellectually dishonest, of course, but it wouldn’t matter. It would be effective, because the picture’s false apparent claim takes 2 seconds to comprehend, while comprehending the real claim requires reading the paper and a modicum of social science familiarity.

And having read more, and knowing what you are really saying, I still can’t look at your graph without having to consciously ignore the obvious visual interpretation.

I don’t think you are David Brooks, I just think you didn’t consider what the public result of this graph would be, and you seem to be arguing here as if that public result is irrelevant.

I’m pretty sure that’s why you are taking so much flack in this thread.

And with that, since you obviously have no intention of admitting you made a gross mistake, more than once, which everyone can see for themselves, I think we’re quite, quite through here. You do realize of course that the way you have conducted yourself here will influence people’s opinions of what you post in the future. Especially given the fact that you’ve admonished others to read what you write rather than attribute to you the positions they wish to challenge. There’s a word for that, starts with an ‘H’ . . .

Let me elaborate on what I mean by ‘moderate’ – it’s the median or majority opinion of the electorate over time on a variety of actions and issues. Thus it is the moderate position to oppose a war in Iraq – and always has been. What changed is not the people, but the information available. So being for an invasion because Saddam was actively assembling nuclear weaponry and had every intention of using it – that would be pretty reasonable if it were true. Being against staying in Iraq when it’s found out that there are no WMD, especially the nuclear kind, well that’s a moderate position too.

and

My own personal agenda is simply the feeling that what gets marked as ‘liberal’, the issues, the reasoning, the people . . . simply isn’t. Thinking that global warming is a real phenomena does not make one a liberal. Thinking that invading a country under false pretexts at the cost of trillions of dollars and millions of lives for no discernable gain is lunacy does not make one a liberal. Thinking that schools ought not to be privatized, or that it should be illegal for private individuals to own chimpanzees, gorillas, etc does not make one a liberal. I want in short, to obliterate the meme that ‘America is a center-right nation’.

and

No, this is not what I am saying. Go back to where I described a lot of people who formerly approved of the invasion of Iraq. They’re moderates – and always have been. Well then, why the reversal of positions? In a word, pragmatic considerations. Prior to the war, they were told that invasion would be cheap and easy (we’ll be greeted as liberators!), and that if this wasn’t done, the Evildoer would nuke Pocatello, or Laramie, or Cedar Rapids, or maybe Branson. Besides which, said Evildoer was responsible for 9/11. Of course, post- invasion, it comes out that there were no WMD’s, that far from being a cakewalk, occupation is money and manpower intensive, and according to our trusted news outlets, Saddam wasn’t responsible for 9/11 after all. Not a big surprise that the average voter changed their opinions about the invasion. Also not a surprise – the big difference between the netroots and the average citizen was that the DFH new all this beforehand; the unwashed masses only being clued in afterwards. Do you see how this will affect the results of your survey? And how this applies to vast range of issues? Joe Median will vote according to the information he has available, his political proclivities notwithstanding. Indeed, what distinguishes the far right in this respect is that additional information does not matter. Global Warming is made-up science. Saddam really was a ‘gathering threat’, etc. Despite impartial information to the contrary. So, your homogenization of opinion on certain issues might very reflect nothing more than being possessed of a certain minimal amount of information.

You want to claim both that these are moderate positions in the sense that they are held by a majority of the population, and that they are moderate in the sense that they represent ‘reasonable’ (your term) positions on things like the invasion of Iraq from people who (unlike right wing crazies) are prepared to change their minds when the facts change. Or else, if you didn’t mean this, somebody has been editing your posts on the sly (and I swear it wasn’t me).

And with that, goodbye from me …

But – and this is key – this ‘left’ and ‘right’ don’t measure absolutes – they are relative terms. It plausibly makes sense to say that someone who is pro-choice anti-Iraq war and was against Bush’s tax cuts is more ‘leftwing’ than someone who is pro choice and anti-war, but favored the tax cuts, and that this person is more leftwing than someone who is anti-choice, pro-war, and liked the cuts. So you can make useful relative judgments. But you can’t and shouldn’t simply map these relative judgments onto absolute notions of whether someone is ‘far left’ or ‘far right. Furthermore, these relative notions of ‘left,’ ‘right’ and ‘center’ have no systematic relationship whatsoever with sensible moderation of the kind that ScentOfViolets is interested in.

And I have told you – what? – three or four or five times now that when I say ‘moderate’, I mean what the majority of the population thinks on a given issue. Not what you keep attributing to me. Like this:

I call myself a moderate for the good and sufficient reason that my opinions on a number of subjects match the opinions of the general public. So it seems that my definition agrees yours when it comes to relative placements along a left/right axis.

or this:

Let me elaborate on what I mean by ‘moderate’ – it’s the median or majority opinion of the electorate over time on a variety of actions and issues.

And yet, you say:

And I think that a lot of the confusion springs from a confusion between two different uses of the term moderate. You would like moderate to mean something like ‘sensible, prepared to accept valid evidence and change their views accordingly.’ And that is a common language use of the term which is perfectly fine. But it shouldn’t be confused with another common language use of the term – people who are moderate in the sense that they are close to the median.

When I encounter this level of refusal to read what is clearly stated multiple times, I’ve got to wonder just how wedded the individual is to an idea. And I just had an entire post about what I’ve bolded, with cites.

Now, until you can actually admit that you were wrong, that I’m using this word in exactly the same way you are, I really don’t see why I should continue this conversation. I’m not going to spend any more energy responding to someone who doesn’t bother to read what I write, who prefers to manufacture my position for me rather than answer substantively.

And I think that a lot of the confusion springs from a confusion between two different uses of the term moderate. You would like moderate to mean something like ‘sensible, prepared to accept valid evidence and change their views accordingly.’ And that is a common language use of the term which is perfectly fine. But it shouldn’t be confused with another common language use of the term – people who are moderate in the sense that they are close to the median.

The scale we use says absolutely nothing about moderation in the first sense of the term. But that’s not its purpose. It tells us about the second. And for these purposes, not only is it perfectly valid, but it captures a lot of interesting variation. Americans, contrary to what you seem to think, vary a lot in the degree to which they believe these things. And that variation is politically important. You say you have shown why this is statistically flawed – but as best as I can understand, your argument rests on an incomprehension as to what the scale is supposed to accomplish.

So, if your problem (as it seems mostly to be) is that you and others who you believe to be moderate in the sense of being sane, sensible and rational are being misdescribed, you shouldn’t worry. Our findings have absolutely no bearing on whether you are moderate in that particular sense of the term, nor are they intended to. The measures of left and right we are using here are trying to capture something quite different. However I have found this an interesting conversation (not only with you) in that the presumed identity between ‘moderate as sane and rational’ and ‘moderate as towards the center of the political spectrum’ seems to be extremely widespread among Americans. That is something that I find odd, but that I hadn’t especially appreciated before, and will be posting on at length in the future.

ScentOfViolets – there is something fundamental I am not getting about your underlying claim, so I will abstain from further argument. The questions were a part of a collaborative survey instrument that others came up with – as noted above, they were chosen to make them roughly comparable with Congressional vote data.

I’ll try to think of a better example. Meanwhile – to repeat my other statement – how do you get that ‘moderate’ voters aren’t really liberal or left-of-center? I call myself a moderate for the good and sufficient reason that my opinions on a number of subjects match the opinions of the general public. So it seems that my definition agrees yours when it comes to relative placements along a left/right axis.

Like most people, I don’t favor criminalizing abortion – a moderate position.

Like most people, I am against school vouchers – a moderate position.

Like most people, I am against the current occupation of Iraq – a moderate position.

Like most people, I believe that global warming is a real phenomenon – a moderate position. [1]

And so on and so forth. And yet, you claim that public opinion is nowhere near that ‘liberal’. What is the source of this discrepency? Is this a matter of asking the right questions in the right way? Because a lot of this seems pretty straightforward, for example the CNN/Opinion Research Corporation Poll that asked this question:

“With respect to the abortion issue, would you consider yourself to be pro-choice or pro-life?”

Am I making a suspect statement when I claim that not wanting to criminalize abortion is a moderate position when the figures to the above question came back: pro-choice 53%, pro-life 44%? This is why I am rather disgusted with attempts frame the above positions as ‘liberal’ rather than ‘moderate’, btw. It has a lot less to do with partisanship, and a lot more to do with accuracy. It smacks of a rather nasty technique being used to thwart the will of the people. It is, in short, undemocratic.

[1] Let’s look at that last, since it illustrates what I was trying to say earlier:

Executive Summary

Overall, a large majority of the American public is personally convinced that global warming is happening (71%). Surprisingly, however, only 48 percent believe that there is consensus among the scientific community, while 40 percent of Americans still believe there is a lot of disagreement among scientists over whether global warming is occurring.

Martin – yes, you are quite right that there is no necessary intellectual coherence between these various positions. There is a whole cottage industry in political science discussing where voters’ attitudes on these issues come from.

I think you are arguing something like the following. On most of these issues, there is a non-crazy position and a crazy position. People answering these questions, and ending up on the left on our scale, are really all just taking the non-crazy position, while the batshit right wingers over there are taking the crazy one. Therefore, all we can say is that the people who appear to be on the far left pole are non-crazy, not that they are genuinely left wing or ideologically coherent.

No, this is not what I am saying. Go back to where I described a lot of people who formerly approved of the invasion of Iraq. They’re moderates – and always have been. Well then, why the reversal of positions? In a word, pragmatic considerations. Prior to the war, they were told that invasion would be cheap and easy (we’ll be greeted as liberators!), and that if this wasn’t done, the Evildoer would nuke Pocatello, or Laramie, or Cedar Rapids, or maybe Branson. Besides which, said Evildoer was responsible for 9/11. Of course, post- invasion, it comes out that there were no WMD’s, that far from being a cakewalk, occupation is money and manpower intensive, and according to our trusted news outlets, Saddam wasn’t responsible for 9/11 after all. Not a big surprise that the average voter changed their opinions about the invasion. Also not a surprise – the big difference between the netroots and the average citizen was that the DFH new all this beforehand; the unwashed masses only being clued in afterwards.

Do you see how this will affect the results of your survey? And how this applies to vast range of issues? Joe Median will vote according to the information he has available, his political proclivities notwithstanding. Indeed, what distinguishes the far right in this respect is that additional information does not matter. Global Warming is made-up science. Saddam really was a ‘gathering threat’, etc. Despite impartial information to the contrary. So, your homogenization of opinion on certain issues might very reflect nothing more than being possessed of a certain minimal amount of information.

I don’t claim that this is only one consideration that has been neglected, btw, or even the most important. I am saying that I don’t see anywhere in your paper where you’ve accounted for these sorts of issues.

But if you want to look at this in relative terms, which is what our data tells us about, then the ‘crazy’ (inverted commas deliberate – I am still wearing my pol-sci hat here) is widely distributed. The median American voter is a mix of ‘crazy and ‘non-crazy’ positions.

Wearing my stat hat, I think that there is some confusion here – there is a difference between the ‘median’ opinion, and the opinion of the random median voter. In fact, the median opinion on a variety of issues is pretty well known, and is moderate to liberal. This does not conflict your claim that the median voter is a mix of different ostensibly left/right opinions; it merely means that, say, one person could be firmly anti-gun control and also believe that global warming is a real phenomenon. To put it another way, statistically, the ‘average person’ is a myth. Nor does the distribution of opinion necessarily reflect a grab-bag of ideological opinion; it could quite easily instead be the result of a mix of information, some highly partisan, some not.

So the point here is that the median left-blog reader is quite different from the median American citizen – he or she is not close to the central set of positions on these (and I would strongly predict, other) issues. And that is a pretty important difference. You may argue that the ‘leftwing’ position on this is the moderate position, properly considered. And that is a perfectly legitimate claim. But it rests on an implicit normative argument, about what is crazy right wing nonsense and what is a reasonable position to take, that I have no business taking as a political scientist. When I am writing as a blogger, it is a different matter of course, and I do allow myself strong opinions as demonstrated by ample empirical evidence.

Again. Please. Read what I actually write; I said quite plainly up above that the ‘moderate’ or middle position was defined by majority opinion over time on various issues. I made no attempt at a normative argument at all.

To put it a different way – there is a lot of variation both among American voters and American politicians on limited scales such as this one, and on more detailed instruments too. Political scientists want to capture this variation, using statistical instruments, and argue that it provides us with some insight at least into where people’s positions rank in relative left-to-right terms (or in two dimensional issue spaces, but that is another world of hurt). This has clear limitations, as discussed above. But it does provide us with useful knowledge. This knowledge may not map well onto your sense of what is sensible, plausible moderation, and what is not, but that is not what these instruments are suppose to do. They don’t try to capture what is ‘sensible’ or what is ‘reasonable’ (given that nearly everyone believes themselves to be sensible and reasonable no matter how far out they are, this would be a heroic claim). Instead, they try to capture the underlying logic of observed substantial variation in the positions American citizens, activists and voters take on various political issues.

Hey, I’m all for doing good science myself. And no, what you attribute to me up above? That is not what I am thinking or what I had said before. Really, you’ve got to actually read what I write. Let me say this again – your tools appear to be defective. They don’t measure what you think they are measuring. They have both range and bias issues.

How did you happen to select these particular questions anyway?