Shared publicly  - 
 
Some weirdness going on today with the Pulitzer Prize winning fact check site, Politifact. I know the editor Bill Adair, pretty well, and I support what he's doing. I think fact checking and calling out public untruths is something journalists should be doing much more of. The Politifact.com franchise is, in my view, a critically important addition to the news system. This makes criticism of it important, as well. Politifact needs to get this fact checking and "truth-o-meter" thing right. I hope it will. But some strange things happen along the way to that.

Last night Rachel Maddow went medieval on Politifact for the ruling I linked to below. I can see why she reacted as she did. Marco Rubio says a majority of Americans are conservative. Politifact looks into it, and says there's no evidence for that, but a plurality of those who identify as liberal, conservative or moderate choose conservative. Since "a majority of" is not "a plurality of" Rubio's statement is false, right? No, says Politifact. It's "mostly true." Huh? Then on Twitter this morning, Politifact responds to me like this: "Jay, our reasoning was that while it was short of a majority, it was still a plurality." To which I said: When you teach your kids what "plurality" means do you say to them, "basically, same as a 'majority,' so don't worry about it..."? I do not teach them that because that would be miseducating them.

For most of the morning I was just baffled by this, but I think I know where Politifact started to go wrong. The "tell" is in the first two sentences... "Liberals may want to argue with Sen. Marco Rubio’s remarks at the 2012 Conservative Political Action Conference in Washington. But they don’t have the evidence to argue with this statement: 'The majority of Americans are conservatives.'"

What we see going on here is Politifact straying from its declared purpose--strict factuality, civic exactness--into "needling." It is here anticipating liberal reactions and playing off them. But think about it: if Rubio is wrong to say conservatives are a majority, this should bother conservatives as much as liberals, or at least those who call themselves conservatives and who care about being strictly factual. Which is the proposition Politifact stands for.

Or even those who care about winning! Suppose you want to build a conservative majority in this country. Should you declare your work done because more Americans ID themselves as conservative rather than liberal or moderate? That's not a majority. Your work is not done! Politifact isn't addressing the factual issue so much as tweaking liberals, anticipating their arguments and trying to subvert them.

And that is how it went wrong. If we sketched a thought bubble above Politfact's head on this one, it would read. "Liberals can't stand it, but more people call themselves conservatives than identify as liberals. Watch: they're going to go crazy when we point this out..." The needling factor led them astray. That's how they got to, "Jay, our reasoning was that while it was short of a majority, it was still a plurality."
Liberals may want to argue with Sen. Marco Rubio’s remarks at the 2012 Conservative Political Action Conference in Washington. But they don’t have the evidence to argue with this statement: "The ...
21
5
James Salsman's profile photoScott Peterson's profile photoCraig Wiggins's profile photoMari Thomas's profile photo
62 comments
 
Politifact has had a slew of bad calls like this. I used to trust them as a source for straightforward facts, but those days are over. I'm not sure what led to their politicization, but it has hurt their brand. It seems their disingenuous "lie of the year" story was just the tip of their newly politicized iceberg.
 
Is there a pattern to Politifact's fails? Determining "truth" in politics will always be a squirrely game, but I agree that the concept of a well-established, credible fact-checker for national politics is a GREAT idea. but it seems to me that Politifact's finding have been consistently skewed right for at least the last year.
 
Is Politifact hoping to boost its credibility by seeking balance instead of truth? That's the path to destroying their brand.
 
+Bill Noble The pattern is that Politifact has gotten into the bad habit of reading people's minds (especially Democrats') and determining the truth or falsity of what Politifact determines that they meant to say rather than simply determining the truth or falsity of their words.

Rachel Maddow has been particularly insightful on the matter. http://bit.ly/wunblS http://bit.ly/xwZ6lD
 
Do we have to subtract Romney from the count on the self-identified conservatives? Or Bill Kristol? Or Andrew Sullivan? Or David Frum? All of who have been described by one faction or another as not really conservative. And what about liberals who no longer call themselves liberals but choose progressive instead? What about Obama who some people say is the most conservative president since Clinton? Seems as if Rubio and PF both ignore the possibility (fact?) that these labels don't mean what the they used to mean, and people self-identify in ways that are inconsistent with other people if not at odds with their own behavior and attitudes.
 
Hope you'll be posting this on PressThink for those of us who want to tweet this out!
 
You can tweet a Google+ link. I did. "I think I figured out how Politifact got itself into the twisted position it has staked out today. My Google+ post: http://t.co/RvR8sRCD
 
This plays well into the "Come on, everybody knows this is a center-right country." myth started by... well, I don't know who started it.
 
Err no, most Americans labeled themselves liberal or moderate.
 
Because I can't resist being an advocatus diaboli, let me make a point broader than the politics of the moment. Except for its founding documents and for the New Deal, the US has always been a remarkably conservative country, despite our mythmaking. We've been technologically very innovative, but socially we have chosen mostly to be passive victims of change, or active resisters. In all the great progressive movements of the last 2 centuries - anti-slavery and suffrage for instance - we've been several decades behind the other wealthy countries.

As for modern Conservatism, I'm with Inigo Montoya: I don' think it means what you think it means.
 
+andy schneider Except it's also true that a plurality of Americans identify as liberal or independent. :-)
 
You're right, Andy. I was sloppy in my post. I should have quoted Rubio's exact phrasing: "The majority of Americans are conservatives."
 
Another quibble. If you look issue by issue at what divides "liberal" and "conservative," a solid majority of Americans identify largely with liberal positions, nearly across the board. Different approach than simply asking "Which are you?" and different results.
 
Your second paragraph is FUBAR, particularly the 3rd and 5th sentences. Rubio didn't say "most," he said majority Further, most doesn't mean a majority. Seriously, reread and fix the second paragraph. Its errors make this post useless.
 
It has been obvious for months that +PolitiFact is extremely poor because of repeated subtle bias. fair.org has been the superior fact checking site for decades. The last error they made was in 1993.
 
+andy schneider 1. I hope I wasn't being disingenuous, and I hope I can get you to see it this way: according to PolitiFact, 40% identified as conservative in the 2011 Gallup poll and so 60% did not identify as conservative (56% positively identified as something other than conservative). Even in informal usage, "most" did not identify as conservative and "most" identified as something other than conservative. It seems black and white and not anything like "mostly true".

2. It's not just "my" logic, but I would absolutely agree that most Americans identify as conservative or moderate. It's also true that most identify as non-moderate. It's clearly not true that most identify as conservative.

3. My impression is that "liberal" can be considered a negative identification, so I'd hazard a guess that many who identify as moderate actually hold liberal (by American standards) opinions. If that's right, even if you try to read Rubio's mind to mean something like "more individual Americans are conservative than liberal" it's still not close to "mostly true".
 
I sometimes wonder if what PolitiFact does is let people offer statements for correction, and maybe listens a little too carefully to those posing it. I saw one where Michelle Obama said that her husband "helped" soldiers get new benefits (VA? GI bill? I don't remember). They rated it as less than "true" because he "only" supported it, and voted for it as a Senator, and she was (they claimed, without context) making it sound like he'd signed the bill himself.

I can easily imagine a Republican supporter making such a statement. It's exactly the kind of not-quite-a-lie that use to bash opponents. "There she is, trying to claim credit for her husband, as if he had signed the bill into law!" I suppose if you're empathic enough, you might find yourself thinking, "well, gee, that is a point. He is President, someone could be confused about the role he played, and maybe she is taking advantage of that."

But, to be honest, that's precisely the kind of - should I be G-rated, and say "BS"? - that you'd expect a place calling itself PolitiFact to cut through. The statement is flat out true. PolitiFacts explanation of why it wasn't completely true should receive a "partly true" rating, at best.
 
This is clearly a plural reason why Politico is such a plural player among the plurality of our nation's plural media. Plural = Major.
 
I want to know more about the money dynamics in big media boardrooms where pandering to the oligarchy is balanced against maximizing ratings by covering from the perspective of the rapidly shifting demographic center. If this keeps up, the strong leftward trajectory inherent in exploding hispanic populations will result in a catastrophe for the uncompromising right.
 
Bill Adair, PolitiFact honcho, has this to say: "Our goal at PolitiFact is to use the Truth-O-Meter to show the relative accuracy of a political claim. In this case, we rated it Mostly True because we felt that while the number was short of a majority, it was still a plurality. Forty percent of Americans consider themselves conservative, 35 percent moderate and 21 percent liberal. It wasn’t quite a majority, but was close.

"We don’t expect our readers to agree with every ruling we make. We have published nearly 5,000 Truth-O-Meter ratings and it’s natural that anyone can find some they disagree with. But even if you don’t agree with every call we make, our research and analysis helps you sort out what’s true in the political discourse."

http://www.washingtonpost.com/blogs/erik-wemple/post/maddow-politifact-and-the-non-sequitur/2012/02/15/gIQABxKiFR_blog.html
 
I'm sorry, but if you can use readily available facts to prove a statement such as "A majority of Americans do not consider themselves conservative", then regardless of Bill Adair's weasel words about agreeing or disagreeing with their call, the Rubio statement that Politifact rates as Mostly True is Objectively False. Non-conservatives outnumber conservatives 60-40! (not even "close" at all). There is no agreeing or disagreeing or when something can be so clearly proven to be an objective fact.

Therefore, PolitiFact cannot be trusted to be factual. This episode is the objective proof. You can disagree, but you would still be wrong!
 
Uh, yeah, that's sort of the point I was trying to make, albeit humorously, in my "plural=major" remarks.
 
I'd like to see more of you picking up on the fact that Rubio did not speak in terms of self-identification regarding the stated conservative majority. So why the focus on self-identification supposedly showing him wrong? In addition, PolitiFact deserves pointed criticism for shifting the question to party identification from the issue of conservative/liberal after failing to find information on the leanings of moderates. If more than a third of moderates lean toward conservatism then Rubio's right, and neither PolitiFact's nor Maddow's opinion about it matters.
 
"Rubio did not speak in terms of self-identification regarding the stated conservative majority. So why the focus on self-identification supposedly showing him wrong?" Because the way fact checking of this type works is, first you have to render the statement in checkable form, which always involves some interpretation. In this case the rendering is, "Well, let's look at how Americans identify themselves in the most reliable survey research we have." Then you can apply the check. So the answer to your question is: Rubio's statement would not be readily checkable without the rendering step. That's why. Of course since it is an act of interpretation, the rendering can be criticized, as well.
 
The answer, good Mr. Rosen, admits the point of my rhetorical question: We end up with a ruling concerning a subtle straw man of Sen. Rubio's statement. The act of interpretation alters the nature of the claim. It's reasonable to look at self-identification and add to that statistics for the conservative lean of self-identified moderates. It isn't reasonable to substitute for the latter the Republican lean of self-identified Independents. Yet the conversation has swirled around PolitiFact's failure to add two and two together for the wrong equation.
 
+Bryan White, so, if I understand you correctly, you're arguing is that, because we only have three categories, but one of them might "lean conservative", we should pretend that we have more than three categories, and declare that it's "mostly true" to make a completely baseless assumption about the quantity that leans.

Back in high school (and college), when one did that - when one pretended to knowledge one did not have based upon ideas that might (or, might not) be reasonable - in order to make statements on an essay exam, the proper term was "bullshitting", not "making statements that should be called mostly true by fact checkers".
 
+Bryan White what straw man? The statement was objectively false when Rubio said it, and it's objectively false the way +PolitiFact reported it. Conservatives don't identify as a majority when "moderates" are included, or when it's split yes/no, or when people are asked to describe themselves, and they get even less when it's split five ways: www.gallup.com/poll/152021/Conservatives-Remain-Largest-Ideological-Group.aspx The fact that they are rapidly losing ground because of demographic shifts makes it even worse. But there's a larger question here: Why should liberals care about this? If liberals hear that conservatives are the majority, it will make them more likely to vote and it will make conservatives more complacent.
 
John Palmer wrote: "if I understand you correctly, you're arguing is that, because we only have three categories, but one of them might "lean conservative", we should pretend that we have more than three categories, and declare that it's "mostly true" to make a completely baseless assumption about the quantity that leans."

You seem to have significantly elaborated on my statement. In short, you do not understand me correctly.
 
James Salsman asked: "what straw man?"

Did Marco Rubio say that most Americans self-identify as conservatives? Because that's the claim PolitiFact checked (along with whether adding the percentage of Independents who lean Republican exceeds 50 percent).

Salsman also wrote:
"If liberals hear that conservatives are the majority, it will make them more likely to vote and it will make conservatives more complacent."

Conservatives who are convinced that PolitiFact is part of a liberal conspiracy owe you debt of gratitude, sir! :-)
 
+Bryan White , fair enough.

Just curious: Do you have a magical touchstone that identifies "conservatives" and separates out self-identification? I assume it would also identify "true Scotsmen" with equal efficacy?
 
Are you asking me, Andy? How else would I know that PolitiFact sought to add the percentage of Independents who lean Republican to the percentage of self-identified conservatives?

"And more to the point: independents who "lean conservative" shouldn't be counted as conservative."

Why not? Not that we have any such data in the first place (we have Independents who lean Republican).

Is it impossible for a person to be conservative if they do not self-identify as a conservative?
 
Andy Schneider wrote: "this idea that we're attacking the wrong thing because Rubio said "are" instead of "self-identify as" is a semantic mess at best."

You think "are" is the key word? Do you know the rest of what Rubio said? Without looking it up? I do. If you don't, go look it up and tell me again how self-identification as a conservative is the right way to check Rubio's statement.
 
John Palmer wrote: "Do you have a magical touchstone that identifies "conservatives" and separates out self-identification? I assume it would also identify "true Scotsmen" with equal efficacy?"

I just love it when people try to imply that somebody else is guilty of a fallacy when there's really no evidence of it at all. I look forward to seeing you deal with the things that I write rather than things imagined or things you would like to see me write, John.
 
It's hard to use the Socratic method with people who refuse to answer questions. :-)
 
+Bryan White the answer to the question you asked me is: Rubio said, "The majority of Americans are conservatives." My question to you is: What makes you believe that statement could be construed as true in any way?
 
+Bryan White , when you said "I just love it when people try to imply that somebody else is guilty of a fallacy when there's really no evidence of it at all."

I just love it when people claim I "imply" something when I did nothing of the sort. I asked, since you do not accept self-identification, if you have a magical touchstone - I also asked if said touchstone allows you to deny those whom you find disagreeable ("No true Scotsman would do such a thing, so, when I said no Scotsman, I was still correct!").

If you think self-identification is insufficient, then there are two possibilities:
1) There is no identification method that is possible - and Rubio was as a high schooler "bullshitting" on an exam, and hardly "mostly true".
2) There is an identification that is possible- and I (admittedly, in a teasing manner) asked what this identification method was. "A magical touchstone, perhaps?"

So, which is it? There is, or isn't? If there is, what is it?
 
James Salsman wrote
"the answer to the question you asked me is: Rubio said, 'The majority of Americans are conservatives.'"

I could have sworn the question I asked you was of the yes/no answer type ("Did Marco Rubio say that most Americans self-identify as conservatives?").

Salsman continued:
"My question to you is: What makes you believe that statement could be construed as true in any way?"

Any way other than via measuring the number of self-identified conservatives, you mean?

I think that it is appropriate to measure the truth of Rubio's statement according to self-identification plus the leaning of moderates (a stat we don't have, so far as I know) because of the context of Rubio's remarks. He specifically attached his statement about the conservative majority to identification with "liberal" in contrast to "conservative. I say that means it's reasonable to use the leaning of moderates.
 
+andy schneider , respectfully, communication must be a two way street - I have seen, especially in online communities, a situation in which a person has become "the wrong one" and each further explanation is used as the basis for further nitpicking and attacks. The burden can't just be on Bryan to explain - it must also be on others to listen, carefully, and try to understand. As one might guess from my comments, I don't think Bryan is correct. But I am trying to understand his objections.

(Obviously, this can only be done when there's a good faith effort to communicate on both sides. Trolls, noise makers, flame-warriors, and so forth, can suck up all of your time and energy :-) )
 
+Bryan White scroll up. You asked me what Rubio said, not whether he said "self-identify" which you asked someone else. What is your source for the self-identification question? The Gallup URL I posted above clearly shows that you are mistaken.
 
@ James Salsman, who wrote:

"You asked me what Rubio said, not whether he said "self-identify" which you asked someone else."

My post at 2:22 p.m. disagrees with your assessment, sir. And I don't know what you mean by "the self-identification question." Identify that and we'll see if I'm mistaken or not.
 
+Bryan White it looks like I'm mistaken about the exact wording of your question. But it doesn't make any difference. Most people not only don't self-identify as conservatives, they don't vote that way, either. I've backed up that assertion with sources. If you think it's wrong, what sources agree with you?
 
Andy Schneider answered (!):

"It's not impossible, but it's an absolutely meaningless distinction."

Okay, we agree it's not impossible. But I don't know what you mean by "it's an absolutely meaningless distinction." What distinction are you talking about? A distinction between conservatives who do not identify as conservatives and conservatives who do? Why would you say it's meaningless?
 
I'll take a stab at James Salsman's claim ...

James wrote: "What is your source for the self-identification question? The Gallup URL I posted above clearly shows that you are mistaken."

I'm assuming you're talking about my statement that we don't have stats for the leaning of moderates (I'll eagerly accept correction if I'm guessing incorrectly). The URL you linked gives stats for whether party members identify as conservative or liberal (or whatever). The parties are already represented in the data for self-identification as a conservative, so it's completely improper to try to add those back in. It certainly does not represent the falsification of my statement that we don't have data on whether self-identified moderates lean conservative.

If that's what you're doing it's a complete mishandling of the statistics. I trust I've misinterpreted your remarks.
 
+Bryan White that Gallup study broke it into five categories: strongly and somewhat liberal and conservative, and moderate. In it, conservatives were fewer than in the conservative/moderate/liberal surveys or the conservative yes/no surveys, but more than the conservative/moderate/progressive surveys. Is there some fact that we disagree about here?
 
John Palmer wrote:

"I just love it when people claim I "imply" something when I did nothing of the sort."

My interpretation of you was charitable, so I'll be interested to see how you try to argue I interpreted you incorrectly.

"I asked, since you do not accept self-identification, if you have a magical touchstone"

If I don't accept self-identification then how do you explain my writing the following: <b>It's reasonable to look at self-identification and add to that statistics for the conservative lean of self-identified moderates.</b>(?) I'd accept self-identification by itself if that was the measure Rubio spoke of. So we have you veering off on this unreasonable "touchstone" thing based on your dubious assumption that I do not accept self-identification. If I were to assume such non-rational assumptions on your part I'd be doing you a disservice. It is charitable to take the "magic touchstone" as your derisive analogy to the method you suppose I use to identify the true Scotsman (conservative).

Mr. Palmer continued to elaborate:
"- I also asked if said touchstone allows you to deny those whom you find disagreeable ("No true Scotsman would do such a thing, so, when I said no Scotsman, I was still correct!")."

If the "touchstone" is not supposed to serve as the analogy to which I refer above, my imagination fails in the effort to make your commentary even borderline coherent.

"If you think self-identification is insufficient, then there are two possibilities:
1) There is no identification method that is possible - and Rubio was as a high schooler "bullshitting" on an exam, and hardly "mostly true".
2) There is an identification that is possible- and I (admittedly, in a teasing manner) asked what this identification method was. "A magical touchstone, perhaps?"

So, which is it? There is, or isn't? If there is, what is it?"

As to the second option, you suggested that the magic touchstone would permit me to identify the conservatives according to the "no true Scotsman" fallacy. If you do not believe that my method would likely have produced an instance of such a fallacy then it's very hard to imagine why you would add the comment.

Therefore, I find your defense lacking. It is reasonable despite your defense to believe you implied I was probably guilty of a fallacy. I stick with the method I mentioned from the first: "It's reasonable to look at self-identification and add to that statistics for the conservative lean of self-identified moderates." No touchstone and no rejection of self-identification except as the sole measure for judging Rubio's statement.
 
James Salsman wrote:

"Is there some fact that we disagree about here?"

Apparently so. You wrote that I was mistaken about something. I'm still not sure what you were talking about. If you're not sure either then I'll take it that you've decided to drop the suggestion.
 
Andy Schneider wrote:
"I say it's meaningless because we're talking about this in the context of what a politician said about the way he believes the American voting base feels about the issues. If someone is "at heart" a conservative, but doesn't identify as a conservative, it's extremely unlikely that their "deep-down" conservatism is going to impact the political sphere at all."

Our representative government is intended to represent even those who do not vote or engage in any sort of political activism. I think you overstate the case by calling their political beliefs "meaningless."
 
+Bryan White ,:"If I don't accept self-identification then how do you explain my writing the following: <b>It's reasonable to look at self-identification and add to that statistics for the conservative lean of self-identified moderates.</b>"

Ah, but I didn't say you disregard self-identification; I said you don't accept it - if a person merely self-identifies as "conservative" that's insufficient.

"It is charitable to take the "magic touchstone" as your derisive analogy to the method you suppose I use to identify the true Scotsman (conservative)."

Indeed. Self-identification is insufficient; now we have to pull in others. Which others? People who claim to be able to categorize often fall into the trap I alluded to.

As for this:
"Therefore, I find your defense lacking. It is reasonable despite your defense to believe you implied I was probably guilty of a fallacy."

You are aware of who "implies" and who "infers", right? You felt that I had implied; I informed you that I had not. You're now saying it's reasonable to believe you know my motivations better than I.

I have explained what I had hoped you would infer: that categorization carries a risk of a particular fallacy, and that it is therefore proper to be skeptical of a person who claims such a categorization.

I'll also note that you did not actually address my question: Is there a method for splitting people into conservatives? If so, I'd like to hear of this function; you already know that I think it would be akin to a magical touchstone, and prone to the No True Scotsman fallacy. Or, was Rubio speaking as if he were an insufficiently prepared high-schooler on an exam, and merely saying something he felt sounded good, with no factual basis whatsoever? Claiming he's not wrong because we can't prove he's wrong, because, hey, we don't have enough information to answer the question at all, only proves more strongly that his statement can't be called "mostly true".
 
John Palmer wrote:

"I didn't say you disregard self-identification; I said you don't accept it - if a person merely self-identifies as "conservative" that's insufficient."

Insufficient for what? If there's one person who employs more ambiguity while writing than I do then perhaps it's you.

"Self-identification is insufficient; now we have to pull in others. Which others?"

I think I've identified the method specifically in this thread on at least three separate occasions, now. If you can locate just one of the three (they're all in agreement) then we're set.

"You felt that I had implied"

I'm honored to have the reasoning I spelled out for you spun back to me in terms of my supposed subjective impression ("felt"). No doubt you did not mean to imply such a thing.

"I informed you that I had not."

If only it was that easy. Sometimes we imply things regardless of our intent. As I mentioned in response to your explanation, you put me in the position of either assuming that you mention the "no true Scotsman" fallacy trivially as a needless distraction or else you mentioned it deliberately to make a point. The latter still seems like the charitable interpretation even if it follows that your present denial is dishonest. Indeed, your explanation agreed with my interpretation on a number of points, notably excepting your intent.

Here's the bottom line: "It's reasonable to look at self-identification and add to that statistics for the conservative lean of self-identified moderates."

Deal with it.

(edit to add this bit that escaped)

"I'll also note that you did not actually address my question: Is there a method for splitting people into conservatives?"

I'm totally dodging your question by repeating the method over and over again.

"If so, I'd like to hear of this function; you already know that I think it would be akin to a magical touchstone, and prone to the No True Scotsman fallacy."

You're mistaken, sir. Mr. Palmer assures me that there was no intent to imply that my method would likely result in a fallacy.

"Or, was Rubio speaking as if he were an insufficiently prepared high-schooler on an exam, and merely saying something he felt sounded good, with no factual basis whatsoever? Claiming he's not wrong because we can't prove he's wrong, because, hey, we don't have enough information to answer the question at all, only proves more strongly that his statement can't be called "mostly true"."

You're precious, John. It's been Rubio's detractors who have confidently stated that he's wrong despite the lack of data for the leanings of moderates (want some quotations from this thread?). My strongest statement on the issue (made elsewhere prior to now) is that Rubio is probably correct, based on an expectation that a conservative preference among moderates would break the approximate 30 percent threshold needed to exceed 50 percent favoring conservatism over liberalism.
 
+Bryan White "Sometimes we imply things regardless of our intent. As I mentioned in response to your explanation, you put me in the position of either assuming that you mention the "no true Scotsman" fallacy trivially as a needless distraction or else you mentioned it deliberately to make a point."

And I did mention it to make a point it was not the point you chose to claim I made. You could have asked - you chose not to. You made a choice; you inferred.

"Here's the bottom line: "It's reasonable to look at self-identification and add to that statistics for the conservative lean of self-identified moderates."

I will take from this that you don't have anything that allows you to categorize; you think it's "reasonable to look" at self identification, and then, you want to add something that's very vaguely worded, begging the question of whether those statistics exist and are meaningful.

In addition to the vagueness, you have not established that this is in any way correct - one can argue that it would be as correct to take those who call themselves conservative and subtract from that "statistics for the moderate lean of self-identified conservatives". Otherwise, self identification as a moderate would not make one "not-conservative" but self-identification as a conservative would make one "not-moderate" - that's not very sensible, to privilege one form of self-identification over another.
 
Andy Schneider wrote:

**OK, let's cut this argument short by once again addressing the single most appropriate question of the entire thread:

"Or, was Rubio speaking as if he were an insufficiently prepared high-schooler on an exam, and merely saying something he felt sounded good, with no factual basis whatsoever?"

YES. That is the entire point.**

It's telling that statements such as the above ("Rubio was speaking ..." "YES") have no more evidential basis than supposedly exist for Rubio's statement.

One would think that Republicans have never done focus group research on people's reactions to the word "liberal." If you're willing to make that kind of assumption then how do you accuse Rubio without looking hypocritical?
 
John Palmer wrote:
*I will take from this that you don't have anything that allows you to categorize; you think it's "reasonable to look" at self identification, and then, you want to add something that's very vaguely worded, begging the question of whether those statistics exist and are meaningful.*

What do you mean I "want to add something that's very vaguely worded"? Do you know my motives better than I do without asking me or something?

What's vague about whether moderates lean liberal or conservative? Do you find the GOP/Democrat Party leanings of Independents similarly vague?

My posts do indeed "beg the question" of whether we have stats for the the liberal/conservative lean of self-identified moderates. Perhaps you'll get around to commenting on whether such data would be useful in fact checking the statement from Marco Rubio that PolitiFact tried to check by using the GOP/Democratic Party leanings of Independents. But more likely you'll find yet another way to name-drop a term for a common fallacy in response to my writing when you haven't discovered any fallacy.

*In addition to the vagueness,*

You mean the vagueness you charged was a begged question up above? Apparently you've answered that question with the ease of assertion. When do you share the evidence on which you base the assertion?

*you have not established that this is in any way correct - one can argue that it would be as correct to take those who call themselves conservative and subtract from that "statistics for the moderate lean of self-identified conservatives".*

I'd say that any person who previously had no complaint about adding the percentage of Independents who lean Republican to the percentage of self-identified conservatives to test a claim that conservatives constitute a majority has the burden of proof for dismissing the parallel role for the conservative/liberal leaning of moderates. I think a majority would accept the method I've proposed without argument. I'll pass on taking up the burden of proof for convincing [i]you[/i], John.
 
+Bryan White, "What do you mean I "want to add something that's very vaguely worded"? Do you know my motives better than I do without asking me or something?"

Your chosen method is that it is "reasonable to look" at self identification and then "add to that statistics for the conservative lean of self-identified moderates". But you don't define this supposed "conservative lean", and beg the question - your statement only makes sense if a previous question "are there good, sufficient statistics for defining 'conservative'?" has been answered positively. "Lean" is also a very vague word in this context - how much lean? What does it mean to "lean"?

Further, if it's possible to gather these statistics of those who have "conservative lean", why are we bothering with self identification anyway? We already have our answer - we'll refer to those magical statistics to find who is and isn't conservative!

So, what you're talking about adding is vaguely worded - it doesn't point to something definite, or clear. It is therefore vague.

As for your "wanting", you have a point - someone may have put a gun to your head and said "Don't you dare accept self identified conservatives as all conservatives! Add something - conservative lean of self-identified moderates, or something" and you might stating that it's reasonable to "add" against your will. Or, you might have another explanation - I'm willing to listen. But without strong statistics that correctly categorize who is and isn't conservative and/or leaning, I'm not willing to accept that you didn't want to, but, by golly, the evidence forced your hand.

Note that gathering such statistics of who is conservative, or who "leans" conservative, while not accepting self-identification, points back to my questions about magical touchstones. How do you categorize?

As for this:
"Perhaps you'll get around to commenting on whether such data would be useful in fact checking the statement from Marco Rubio that PolitiFact tried to check by using the GOP/Democratic Party leanings of Independents."

I'd as soon argue the use of invisible pink unicorns as for statistics for which I have no evidence of existence or sufficiency. But you can't rank a statement "mostly true" if you have no evidence to support it.


As for this:
"I'd say that any person who previously had no complaint about adding the percentage of Independents who lean Republican to the percentage of self-identified conservatives to test a claim that conservatives constitute a majority has the burden of proof for dismissing the parallel role for the conservative/liberal leaning of moderates. I think a majority would accept the method I've proposed without argument."

What, so someone has to specifically complain, quickly enough to suit you, or the burden of proof shifts from the person making the assertion, to the person who failed to complain? No, I think I'll stick by the old rules; you want to make an assertion, you're the one responsible for backing it up.
 
John Palmer and Andy Schneider are the twins of self-stultification.

Palmer:
(Y)our statement only makes sense if a previous question "are there good, sufficient statistics for defining 'conservative'?" has been answered positively.

So we just assume that it can't be answered positively and declare Rubio a "bullshitter"? You get a pass for drawing conclusions without basing them on evidence, then. Nice.

Schneider:
(T)he only meaningful metric for these labels is whether or not people self-identify. "conservative lean" is something that only the person doing the leaning can evaluate, since, as is mentioned several times upthread, the labels have no objective definition.

If the labels have no objective definition then how can self-identification serve as a meaningful metric? Subjective definitions might make "liberal" and "conservative" mean the same thing to different people.

Take a bow, gentlemen.

One more for Mr. Schneider:

(T)his has gone so far astray from anything relevant to the actual subject at hand that I'm convinced that dude just wants something to pick apart.

With the possible exception of my evaluation of Mr. Palmer's attempt to argue he implied no fallacy on my part, I've stuck pretty narrowly to the subject: The right way to evaluate majority conservatism is via self-identification plus the conservative lean of moderates. The bulk of the far-ranging discussion has consisted of others trying to pick that apart, generally using spurious methods such as the self-stultifying techniques noted above.
 
You're doing a good job being the only person who agrees with you.

I expect John Palmer will give you a pass on supporting your assertion with evidence.
 
+Bryan White , you are a very sore loser, aren't you?

Look: I explained what bullshitting was - it's a common definition. A student who brags they bullshitted their way to a C (or even a B) is saying something we all understand - speaking (well, typically writing) without knowledge of the facts.

I pointed out the flaws in your proposal: if we had statistics for who moderates who lean conservative, we'd have statistics for what constitutes "conservative" and wouldn't need self-identification, and, claiming that there's a squishy middle ground (moderates who lean conservative) just as easily lends itself to the argument that there's a squishy middle ground on the other side (conservatives who lean moderate).

Does this prove that Senator Rubio was bullshitting? No. But we have established a good case for it. I'll let people draw their own conclusions. If they think he had some measure by which he could make that statement, they can rest assured that a politician would never engage in bullshitting while speaking to a friendly crowd, and hope that the Easter Bunny will bring them some extra nice candy for their faithfulness in unlikely things. (Okay, granted, I'm laying it on a little thick here.)

You're very good at finding fault with those with whom you disagree. However, if one wants to learn the truth, one must apply such a skill just as rigorously to those with whom one agrees, as well - and especially with one's self.
 
John Palmer wrote:

*you are a very sore loser, aren't you? *

No doubt. Mother Jones' Kevin Drum makes my point:
http://motherjones.com/kevin-drum/2012/02/maybe-majority-americans-really-are-conservative

Does this prove that Senator Rubio was bullshitting? No

Dang! I lose again! lol

But we have established a good case for it.

Well, yeah, if we ignore the flaws in your case. Suppose we had the whole population who took the survey that showed 40 percent identifying as "conservative" and had them take a self-identification poll with 50 options to choose from. And only 5 percent choose "conservative." Does that prove Rubio wrong all the more?

Your case rests on the legitimacy of using a poll with more choices than Rubio countenances.

Now consider something very important about the poll Drum references. It's a poll of likely voters, yes, so it's not the perfect vehicle for justifying Rubio's statement. But if you look inside the numbers you see something important. The population sample breaks down as plurality Democrat. Yet over 60 percent favor "conservative" over "liberal."

I'll let people draw their own conclusions. If they think he had some measure by which he could make that statement, they can rest assured that a politician would never engage in bullshitting while speaking to a friendly crowd, and hope that the Easter Bunny will bring them some extra nice candy for their faithfulness in unlikely things.

Excellent job of making it appear that you bring an anti-politician bias with you to your analysis. If only Drum could think like you he wouldn't have looked for additional survey data that might support Rubio.

You're very good at finding fault with those with whom you disagree. However, if one wants to learn the truth, one must apply such a skill just as rigorously to those with whom one agrees, as well - and especially with one's self.

What makes you think I don't do that? Do you have any examples that do not (fallaciously) beg the question? Oh, wait a minute. I get it. You're not implying anything. Pfeh.
 
+Bryan White, congratulations - now, you have brought forth evidence that would support that America's likely voters are majority conservative. That doesn't quite comport with what Rubio said, but, it's much better than declaring that one should take self-identification, and add something to that, based upon an assumption that there are statistics that would support the add. It does not excuse Politifact's nonsense, but it does give Rubio a leg to stand on.

As for this:
"What makes you think I don't do that? "

The very fact that you've never mentioned that, gee, it is pretty silly to say "let's take self-identification, and add something to it from statistics that may or may not exist, and if they did, would make self-identification unnecessary."
 
Your answer was supposed to avoid fallaciously begging the question.
Add a comment...