Sunday, May 29, 2016

Experiment

Okay, let us consider the possibility that it really is the conservatives who are ignorant, aren't listening, and reflexively reject other points of view.

How are we going to measure whether that is true?  Something that would stand up when presented to a man from Mars.

11 comments:

RichardJohnson said...

How are we going to measure whether that is true? Something that would stand up when presented to a man from Mars.

Q: Do you consider Slate an unbiased new source?
Q: Do you consider the NYT an unbiased new source?
Q: Do you consider Pajamas Media to be an ignorant, racist news source?

A NO will definitely expose those who are "ignorant, aren't listening, and reflexively reject other points of view."


Have a questionnaire designed by a social psychologist or sociologist- you know, professions which lean 95% Demo/prog/lib. You know it will be objective.

Am I cynical? No, just realistic.

Sam L. said...

We have seen that colleges tend toward excluding the ideas of those who think differently, or merely have a point of disagreement. Let someone go to Hillsdale College, which I understand to be quite conservative, and see how they treat those who think differently, or disagree with them.

jaed said...

The problem here is who judges. The judge will have a viewpoint and that viewpoint won't be neutral, or at least we can't expect it to be.

I like Haidt's method for this: have each side try to make the case for the other side, and let the other side judge whether that case is convincing. (Example: have pro-lifers answer an abortion question from the pro-choice side, and vice versa. Then give pro-choicers the pro-choice answers (from both pro-choicers and pro-lifers), and see whether they can distinguish the pro-lifers' answers - and vice versa, have the pro-lifers grade the pro-life answers written by both groups.) Whichever group can more convincingly represent the opposite position in the judgement of those who hold that position wins.

This doesn't exactly tell us who is more ignorant, isn't listening, or reflexively rejects other points of view, but it's probably close enough for our Martian to accept it as an approximation. Someone with these characteristics is less likely to be able to accurately represent those other points of view.

I like this method because it doesn't require anyone to be objective. It is thus suitable for a situation where anyone's objectivity is likely to be questioned by one or both sides, and where objectivity can't be tested because what is objective truth is the very thing in question.

james said...

Ignorant or knowledgeable about what? How do we weight things we know that aren't so, vs things we don't actually know?

Details, themes, big picture? How important we judge the various themes to be will vary a lot, of course, and that makes the big picture: we probably can't get good agreement on that. But someone who doesn't realize that themes of security and freedom and trust all play roles in, for example, gun control questions, we can safely call ignorant.

jaed's solution is good.

Roy Lofquist said...

Tough row to hoe there, AVI. According to my betters I'm a knuckle dragging anti-science troglodyte. This from people who think that the Ricci Tensor is a new kind of yoga, that Schroedinger's wave equation is about surfing and that Georg Cantor was a borscht belt comedian. Try to engage them with things like numbers and such and they insist that's not real science. Real science has pictures and charts and meerkats. Gotta have some meerkats.

Never try to teach a pig to sing. It wastes your time and annoys the pig.

bs king said...

I actually would wager the hardest part of this would be defining your demographic in a meaningful way. The way I see it, there are three parts to this:

1. Who's a conservative
2. Of those named in #1, which subset are we most concerned with?
3. What is the liberal equivalent of #2?
4. In a comparison between those from #2 and those from #3, who is more ignorant and faster to reject new ideas?

My rationale:
1. With people increasingly defining themselves as independent/unaffiliated, we're getting to the point where selection bias is a major problem. Someone pontificating extensively about a conservative or liberal issue in one area could subsequently claim to have no affiliation. So are we letting people define themselves or letting others define them?

2. When I hear my liberal friends say "those conservatives" they are almost always referring to some mix of their older relatives/crazy friends or acquaintances/Fox News/religious leaders they heard on TV/politicians. When my conservative friends say "those liberals" they are almost always referencing some mix of their younger relatives/crazy friends or acquaintances/liberal media folks/protestors or academics they saw on TV/politicians. Which of these categories are we concerned about? Are we talking general population or something specific? Even if the general conservative population was more ignorant/less receptive, are their elected leaders? What about colleges? Do we care if the bottom quartile is more ignorant or the upper quartile? Etc etc.

3. An add on to #2, but we should make sure it's a fair comparison. We shouldn't compare the general conservative population to academic liberals, or our personal friends to elected leaders. Obvious, but people do this all the damn time.

4. Okay, if we do have two comparable groups that are accurate representations of "conservatives" and "liberals", now what? Well, I like jaed's thought. I also like the study I wrote about here:
https://graphpaperdiaries.com/2016/02/07/people-our-own-worst-enemy-part-8/

In that study they presented people with a math problem (part of a larger test) that was based on "results of a new study examining the effectiveness of gun control legislation". For some it made gun control look like it decreased crime, for others it made it look like it increased crime. They found ability to accurately do the math was profoundly affected by what you already believed. I think that's a pretty good proxy for ignorant/rejects new ideas.

So if I had my druthers, I'd take members of Congress, give them a full math test, and have a problem in there whose correct answer would conflict with something they believed. Control for mathematical ability on the rest of the test and control for voting record to approximate strength of belief, and I think you'd have some interesting data. Oh, and tell them all you were just testing to see which party had better math skills, to make sure they were as incentivized as possible to give correct answers. If your convictions are so strong that you can't even hypothetically answer something that would conflict with them, I'd say you have a problem.

PS: In the study I mentioned above, Republicans were both more likely to get the problem right if it agreed with them, and more likely to get it wrong if it didn't agree with them than Democrats. Not sure what to make of that.

Grim said...

The easiest way to test it would be to see if the conservatives could, or could not, fairly reproduce the arguments of their opponents. If you can explain their assumptions, principles, and arguments, you've been listening. If you really can't do that -- or do it in a way that is wildly unlike what they are actually saying -- you have not been.

Roy Lofquist said...

Grim,

That would be a lot easier if progressives would frame their arguments in a manner that can be logically disputed. An awful lot of them amount to stream of consciousness non-sequiturs with lots of adjectives.

Note that this a tactic that Trump is using against them. He makes outrageous nonsensical statements and all they can do is call him names because they can't parse what he actually said. If they figure it out he simply denies that he said it. Classic doubletalk.



Grim said...

Roy,

I wasn't intending to test the validity of their arguments, but only the proposition: "consider the possibility that it really is the conservatives who are ignorant, aren't listening, and reflexively reject other points of view."

If they give a bad argument, based on flawed assumptions, without a coherent logic, that's fine. If I can reproduce it accurately, I was listening and am not ignorant of their point of view. My rejection then won't be reflexive, but based on a demonstration that the assumptions are flawed or that the logic doesn't follow.

On the other hand, good logic doesn't guarantee truth preservation in the real world because physical objects aren't logical objects. Logical objects are alike throughout, whereas any two physical objects will be at least a little different. Thus, reality is analogical, and logic isn't always a sure guide.

Assistant Village Idiot said...

Bethany- brilliant

Grim - that experiment was tried over at Volokh Conspiracy about 3 years ago. It was right after that UCLA study about academic sources for news reports came out. Nostrils flared.

Roy Lofquist said...

Grim,

Can't say I disagree. I am making more of a utilitarian argument. If you can honestly say that you try to understand different points of view and have, indeed, changed your mind on significant subjects over the years then you are on more solid ground if you stop listening when you determine that you've either considered the argument before and have not heard anything new or that the person addressing you is asserting something that is absurdly incorrect.

The utilitarian twist is that you owe people careful attention both because of common courtesy and that you might learn something new but you also have a high duty to yourself and to those close to you to make the best use of your limited time on this mortal coil.

To assert that their is some higher moral imperative, a duty to society that compels you to surrender agency is to ask you to betray yourself.

Short form: I agave up guilt for Lent.