Krugman being Krugman

[badlydrawnjeff.livejournal.com]

"There's the great line, of course, about how his "lived experience" (see? I told you, he's doing empirical work!) confirms that motivated cognition "is not, in fact, symmetric between liberals and conservatives.

...

That Krugman is too thick to see that one can't infer anything about the quality of partisans' reasoning from the truth or falsity of their beliefs is ... another element of Krugman's proof that ideological reasoning is symmetric across right and left!"

Comments

[the-rukh.livejournal.com]

Well, Krugman was talking about three things. He acknowledged the study that says that people use data to enforce their pre-held beliefs, but he also went on to point out that they TYPES of beliefs matter and furthermore what people do when they find their beliefs are wrong matter. Liberals are much more likely to acknowledge after the fact that their belief was wrong and work to fix it, a la Obamacare website spectactular failure, while conservatives are much more likely to deny deny deny, a la Iraq war.

The guy who thinks he can respond to Krugman simply ignores all that and makes a claim about how people interpret data then accuses Krugman of being wrong for something he didn't argue. Well, ok then. It's a pretty awful response piece.

[oslo.livejournal.com]

Liberals are much more likely to acknowledge after the fact that their belief was wrong and work to fix it, a la Obamacare website spectactular failure, while conservatives are much more likely to deny deny deny, a la Iraq war.

Kahan's response to this was just to say that it would be nice if we could somehow test an assertion like this for its validity. And lo - indeed, we can, and indeed we have! And it turns out, according to Kahan at least, that the evidence shows that Krugman is in fact wrong to believe that liberals somehow have the better form of "motivated cognition."

I wouldn't take it too personally or seek to spin it one way or the other. Jeff just wants to mock Krugman. You just want to defend your tribe. I think the better approach is to think about the way one assesses new evidence that may count against or support our established political positions. We know that Jeff isn't very good at doing that, nor is he very good at recognizing that he isn't very good at it, and the OP and his comments here are more telling than he realizes. As for you - I suppose I don't yet know, but I'm not optimistic, given your comments here.

[the-rukh.livejournal.com]

If I thought Krugman was wrong, I'd say I thought Krugman is wrong. I have no qualms about telling someone of "my tribe" that I think they're wrong, as you should know.

[goumindong.livejournal.com]

Ahh but Kahan is wrong here. He has not tested what the thinks he has tested.

If confronted with evidence which suggest that their world view is wrong will they take that with the same weight as anyone else? Certainly no, no one does, not even Kahan. Everyone weights by their priors. If your priors are good there is nothing wrong with this. You should be skeptical of evidence which goes against what prior evidence has lead you to believe proportional to the strength of the prior evidence

The question is, mainly, whose priors are better? Whose priors are based on evidence and whose priors are degenerate. The answer to that is clear. So a test which suggests that "liberals" use ideological reasoning as much as conservatives doesn't help us here. The question is "how are those beliefs formed?" and within the "liberal information structure" those beliefs are formed by testing. Within the conservative information structure those beliefs are not.

Essentially Krugman is making a point about priors and how priors are formed institutionally. And Kahan comes in and says "well the people who aren't tasked with creating those priors aren't updating them properly".

Moreover i would suggest that the non-experimental evidence is more valid in this case due to observation effects. Granted actually evaluating a large group of information updating (via the publications of said information sources) would be a task far beyond most researchers and even if you had a team that could do it, you would be accused of bias the instant you published (because you would have to actually evaluate how the evidence was accepted rather than yes/no responses)

[the-rukh.livejournal.com]

The big problem is the US has only two parties, so we try to somehow shoehorn them in to this left and right half of the whole, and it's just nonsense. They are two parties of plenty of possibilities. Its much easier in other countries that have lots of parties to see each party as representing a certain group, not just 1/(total parties) of the whole. There's no party in the US that represents communism. There's nobody who represents hardline environmentalists. We have a moderate party and an ideological purist party.

That brings us to a second point, the parties also are from two totally different places. Any sort of rebellion in the democratic party has long since drained out. Its an establishment party of moderate ideals. Its just what it is. Its the oldest US political party.

On the other hand, what used to be the liberal party turned pro-big business turned conservative party has turned in to the ideologist party. Its all about people being "true conservatives" and whatever the heck that means. Apparently conservativism being defined by an eclectic collection of wedge issues. Honestly it's more a collection of competing ideological theocracies right now. The base is a throng of people who either buy in to the talk radio of how evil liberals are and embrace conservatism by default, even though they're really not any sort of ideological purists and may not really care about any of the conservative ideas, and people who really do buy in to conservative ideas. But I think the people who buy in to the electic mish-mash are more a noisy minority than anything.

[goumindong.livejournal.com]

No. That isn't right. While people use ideological reasoning all the time, ideologies are not homogenous across parties(even among the Republicans) and ideological groups do not receive their information from the same sources though there certainly is more overlap among aligned ideologies than there is among non-aligned .

You're confusing the two party system with something that has actual effect on politics. That the US has two parties is simply a result of coalition forming occurring outside of the legislature rather than within. A multi-party parliamentary system would not change how ideologically motivated people received their news, nor would it change the makeup of congress, nor would it change how votes go down. Fundamentally every political system is two party, and it cannot be changed until we figure out a way to get around the fact that legislation must either pass or fail.

[the-rukh.livejournal.com]

Well that's kind of my point. All humans use their preconceived beliefs to interpret data. That's just part of what we do, it makes sense to try to fit what you see in to what you know. However parties that are built differently are naturally going to react differently to conflicting data, and furthermore as you say, they're not heterogeneous bodies, they're fairly diverse.

You think the two party system doesn't have an effect on politics? That's an odd thing to say. But I'm not saying that people will not be swayed by their preconceived ideologies, I'm saying that its easier to realize they're not just two sides of a coin.

[goumindong.livejournal.com]

No, the two party system does not have an effect on politics. Our regional representation system does; but at the end of the day policy votes are up or down and there is no way to change that. In parliamentary systems the same coalitions are formed as in congressional/legislative systems the difference is that in a parliamentary system they form the coalition at the same time they form the government and that is where the horse trading goes on. In congressional/legislative systems the horse trading goes on between individuals (E.G. the gang of 8) and between caucuses (like the progressive caucus).

My point is a bit deeper into that. The ideological information dissemination methods which filter our thought are not simply split in 2. The libertarians have their own and the theocrats have their own and the progressives have their own and the environmentalists have their own and the pragmatists have their own.

I mean look at how I am having to explain to Peristaltor about what the economic orthodox is. Yet he would probably consider himself a liberal as I do but he still thinks that Krugman's explanation of how money systems work is wrong because he follows different ideological information sources.

[the-rukh.livejournal.com]

You don't think the two party system has an effect on politics. Ok. That's weird, but your right.

[goumindong.livejournal.com]

No. Regionalism does. But the fact that its 2 party doesn't. Coalitions will be formed regardless of labels that individuals place on themselves. Whether we had the republicans and then 30 different parties in a coalition doesn't matter. It would be the same people (see: Regionalism) and they would vote the same way. Wouldn't make a difference that Rand Paul got to call himself a member of the Libertarian party in his campaign literature.

[goumindong.livejournal.com]

Other countries only have two parties too. You're mistaking what they call parties with what we call parties. What they call parties are more akin to interest groups, or within the congress the caucuses which make up the groups of ideologically aligned people.

You are seeing the end effect, that coalitions form within the parliamentary body rather than without it as having effects on the populace which do not exist.

There are people who represent the hard line environmentalists. They're within the Democratic party. Just as Libertarians actually exist, they're within the Republican party

[the-rukh.livejournal.com]

I think you're mistaking what I'm saying. I'm stating that by having two parties its easy to just picture them as two halves of the whole that operate in the same yet opposite ways. My point is it's not that at all. They're two interest groups that are not representative of the whole spectrum.