mercredi 18 septembre 2013

Confirmation Bias, stereotypes, and how do you interpret this statement?

Hi.



I was reading this blog and comments:



http://abagond.wordpress.com/2010/08...irmation-bias/



I saw this:


Quote:








Originally Posted by Mira

Confirmation bias is a harmful thing, because it help inaccurate assumptions exist. a stereotype starts and continues to exist, and after a while it’s almost impossible to get rid of it.



But I also believe it’s impossible to escape confirmation bias. It seems that humans work that way. Even those who should know better (scientists) can’t escape it. Not to mention regular folks. And yes, whether you’re the one being affected by a stereotype doesn’t always mean you’ll be free from it.




But if it cannot be escaped, then does that mean that we're doomed to continue being harmed by it, and so complaining about it is a worthless exercise? Or, is it possible that even if it can't be "escaped", it can at least be reduced, which would itself be useful and beneficial?



Also, does the existence of confirmation bias and other biases mean that it is important, if one wants to be a rational thinker, to seek others' input on one's views, as they may be able to notice things you may have missed due to your biases?



One poster ("RR") claims that stereotypes (this thread and blog is about "racism" issues and similar stuff) may be right. I saw this response to him:




Quote:








Originally Posted by Jasmin

And RR, you prove once again that you don’t have any educational background in the things you talk about. You really should spend more time in a classroom learning rather then speaking nonsensically.




Where can one get the necessary "educational background" for this topic when you don't have a lot of money to spend on a college or something? Does this mean you need to take huge, long courses in everything you want to critically analyze? There's so many subjects out there -- what to do?



Another response to "RR"'s posts:


Quote:








Originally Posted by Thaddeus

The following statement is an example of a stereotype that is not subject to confirmation bias:



Men are taller than women.




Au contraire. That statement can very easily be subject to confirmation bias. I think we could easily engineer a situtaion in which the average woman’s height was taller than the average man’s height and then ask people, after the situation, to write down what they thought the avergae heights were. I’d lay dollars to donuts most would place the women as shorter than the men.



What you have there, RR, is a generalization. If you feel that it holds true in all circumstances, regardless of what your eyes SHOULD be telling you, then you have a stereotype which is being held in place by confirmation bias.



Thanks for closing out your time here on this board with a clear demonstration that, while you may be a talented polemicist and rhetorician (and what should one expect from a journalist and computer salesman?), you have a very shaky understanding of science and its logical foundations.



See you in the funny papers, Stevie!




However, I'm wondering about this example, of a supposed "stereotype" that is "true". As to me, when I see such a statement ("Men are taller than women"), I interpret it as saying "on average, males of the human species are taller than females, averaged over the whole species". It looks like this may actually be true, based on the data given here: https://en.wikipedia.org/wiki/Human_...ound_the_world . Yet he says the statement is not necessarily true. Which would mean the interpretation would have to vary. He says there are "situations" in which the heights are the opposite way, but the claim, as stated, doesn't seem to refer to a particular situation. So what is going on? Isn't it so that when a claim like that is made, there is an intended interpretation? Isn't the intended interpretation of that claim ("Men are taller than women") usually the same as the one I've made? Or is that just my own confirmation bias -- what, if anything, does this claim "usually" mean? I suppose, however, that someone could intend for a different interpretation, in which case we could say that maybe the claim is wrong then if it is wrong under that intended interpretation. Is this the answer? I also notice there are some caveats mentioned about the cited data, which may bias it -- but is it so wrong as to make the apparent height difference between men and women nonexistent? (E.g. a systematic bias by all researchers to pick short girls or tall guys, or both) I suppose it's possible. Is this the answer? Are both of these valid answers?





via JREF Forum http://forums.randi.org/showthread.php?t=265516&goto=newpost

Aucun commentaire:

Enregistrer un commentaire