Total Posts:23|Showing Posts:1-23
Jump to topic:

Implicit bias

feverish
Posts: 2,716
Add as Friend
Challenge to a Debate
Send a Message
1/26/2015 6:35:56 PM
Posted: 1 year ago
Judge not lest ye be judged...

Thought I'd share this interesting site I just came across with the DDO massive, apologies if it's old news.

https://implicit.harvard.edu...

Seen something like this before on some documentary, basically a set of scientific tests of unconscious bias and prejudice. Anyone who thinks that they have no views that are racist/homophobic/religously biased etc or that prejudice isn't highly pervasive and subtle may be surprised at their results.

For example, as someone who considers myself an anti racist I was initially surprised to see that I have a moderate unconscious bias towards my own race. I would be very interested to hear if other people are surprised at their results and also what people think about these tests and about unconscious bias in general. Apparently even people from the stigmatised groups tend to slightly favour the other group.

Try it, it's kind of fun. You don't have to answer any of the questions about yourself or your opinions if you don't want to or can't be bothered.
R0b1Billion
Posts: 3,733
Add as Friend
Challenge to a Debate
Send a Message
1/26/2015 6:51:14 PM
Posted: 1 year ago
At 1/26/2015 6:35:56 PM, feverish wrote:
Judge not lest ye be judged...

Hey I remember you, you were on here a long time ago. Come back from hiatus? I just came back after about six months off, I can never quite stay away from this place.

Somebody posted this test a long time ago (probably nobody that's around now was around then anyway lol). I was pretty disappointed with my results, I couldn't shake the feeling that it was designed to tell me I was biased no matter what I answered...
Beliefs in a nutshell:
- The Ends never justify the Means.
- Objectivity is secondary to subjectivity.
- The War on Drugs is the worst policy in the U.S.
- Most people worship technology as a religion.
- Computers will never become sentient.
feverish
Posts: 2,716
Add as Friend
Challenge to a Debate
Send a Message
1/26/2015 7:13:40 PM
Posted: 1 year ago
At 1/26/2015 6:51:14 PM, R0b1Billion wrote:
At 1/26/2015 6:35:56 PM, feverish wrote:
Judge not lest ye be judged...

Hey I remember you, you were on here a long time ago. Come back from hiatus? I just came back after about six months off, I can never quite stay away from this place.

Yeah, i remember you well Rob, hope life's treating you good man :) I logged in for what must of been the first time in years a week or two ago haha.

Somebody posted this test a long time ago (probably nobody that's around now was around then anyway lol). I was pretty disappointed with my results,

Interesting, so were you surprised or just disappointed?

I wouldn't take it as too bad a reflection on your character, more as an illustration of how ingrained in society and even in individual psyche certain prejudiced associations are.

It's really interesting to me that those of us with no conscious prejudice, and even those who one might expect to have an opposite preference may still reflect cultural stereotypes and value judgements at a deeper level. Even just to the extent that cognitive processes can be slightly distorted by it enough to show up in a test like this.

As the site points out these prejudices are also highly malleable and can generally be challenged and adjusted with deliberate effort.

I couldn't shake the feeling that it was designed to tell me I was biased no matter what I answered...

Are you saying you were left with the feeling that the test itself was biased?
Maikuru
Posts: 9,112
Add as Friend
Challenge to a Debate
Send a Message
1/26/2015 7:43:07 PM
Posted: 1 year ago
It's a well known test and the results bring up interesting talking points. An issue, though, which the creators themselves ended up having to address, is the fact that it isn't internally counterbalanced. That is, you will certainly have carryover effects from whatever order you receive (white/black and good/bad) and if you want to correct for that, you need to take the test again and hope you receive a different order. Of course, the ordering is random so there is no guarantee you'll receive the opposite order and you can't retake the test immediately or else you'll have carryover effects from across the tests. It's a pain in the asss lol
"You assume I wouldn't want to burn this whole place to the ground."
- lamerde

https://i.imgflip.com...
Maikuru
Posts: 9,112
Add as Friend
Challenge to a Debate
Send a Message
1/26/2015 8:50:52 PM
Posted: 1 year ago
By the way, hi Feverish. We've got you, me, and Rob in here. If we can get Raisor and Danielle in this thread, the combined DDO age of this thread would be old enough to drink.
"You assume I wouldn't want to burn this whole place to the ground."
- lamerde

https://i.imgflip.com...
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/26/2015 10:25:10 PM
Posted: 1 year ago
Actually, the test doesn't really measure racism. Kaufman suggests it's a in-group, out-group bias, and cites experiments that show the effect is equal for outgroups of the same race and different race to the participant.

https://www.psychologytoday.com...

It may not even be in-group, out-group. There's no reason to believe ANY explanation for the effect at this stage, IMO.
feverish
Posts: 2,716
Add as Friend
Challenge to a Debate
Send a Message
1/27/2015 6:34:36 AM
Posted: 1 year ago
Intriguing that we are all focussing on the racial bias test although there are many others on there, also that apart from the overt racist in the thread, everyone seems to be fairly sure that the test is flawed or doesn't mean what it seems to...

At 1/26/2015 7:43:07 PM, Maikuru wrote:
I it isn't internally counterbalanced. That is, you will certainly have carryover effects from whatever order you receive (white/black and good/bad)

Yeah, they acknowledge that this could be a slight factor on an individual's test results:

"the order in which you take the test does have some influence on your overall results. However, the difference is very small. So if you first pair gay people + bad and then pair gay people + good, your results might be a just a tiny bit more negative than they would be if you had done the reverse pairing first. One way that we try to minimize this order effect is by giving more practice trials before the second pairing than we did before the first pairing."

So you get more practice trials to adjust to the change, hopefully cancelling out some of the effect of you just becoming more familiar with how the test works as it goes on.

However if, as you acknowledge the order is randomised, then the overall statistics from all people who have taken the tests should not be affected, even if individual results are.

By the way, hi Feverish.
Hey man, good to virtually see you again :)

At 1/26/2015 10:25:10 PM, Garbanza wrote:
Actually, the test doesn't really measure racism.

I think if you mean racism in terms of a conscious belief that some races are inherently superior to others, then you are absolutely right.

However if you mean racism in terms of simply having some sort of preference for one group over another then I would love to know what else you think it could possibly be measuring?

Kaufman suggests it's a in-group, out-group bias, and cites experiments that show the effect is equal for outgroups of the same race and different race to the participant.

https://www.psychologytoday.com...

I think you and possibly Kaufman are actually misrepresenting the results of the experiments listed in that article.

The one experiment cited that addressed race rather than nationality does NOT say the effect is equal, it merely states that "the White-Black IAT-effect was largest for those in the Caucasian group, and smallest for those in the African-American group. Conversely, the White-Latino IAT-effect was largest for the Caucasian group and smallest for the Latino group. For those in the Caucasian group, there was no difference in the White-Black IAT-effect and the White-Latino IAT-effect."

So this shows that in group bias is definitely a factor but not that it explains the overall racial bias. In fact the Harvard IAT site addresses this issue as well, it fits with the findings of the experiment cited but disproves the conclusions of the article.

"Results from this website consistently show that members of stigmatized groups (Black people, gay people, older people) tend to have more positive implicit attitudes toward their groups than do people who are not in the group, but that there is still a moderate preference for the more socially valued group. So gay people tend to show an implicit preference for straight people relative to gay people, but it is not as strong as the implicit preference shown by straight people. We think that this is because stigmatized group members develop negative associations about their group from their cultural environments, but also have some positive associations because of their own group membership and that of close others."

So, yes in-group out-group bias certainly explains some of the effect on people who evidence a bias for their own group, but certainly can't explain the whole picture when you have results such as gay people favouring straight people more than their own group.

It may not even be in-group, out-group.

Well that is what the article you cited seems to be suggesting...

There's no reason to believe ANY explanation for the effect at this stage, IMO.

In your opinion are the huge differences in results favouring privileged and majority groups likely to be mere coincidence?
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/29/2015 4:05:48 AM
Posted: 1 year ago
At 1/27/2015 6:34:36 AM, feverish wrote:
In your opinion are the huge differences in results favouring privileged and majority groups likely to be mere coincidence?
No, I just mean it's not obvious what causes the effect and it's a big leap to assume it's implicit racism, although of course you're right that it depends on what you say racism is exactly.
I didn't actually check your link because I assumed it's the implicit association test where you have to match faces and words and they measure your speed and accuracy, so if it's not what you're talking about then I'll be embarrassed.

But does speed of matching words and images have to do with racism? It's set up so that it really seems as if it does, but imagine the same test with less emotive content. Suppose they had pictures of colors and then words related to texture and finish such as shiny, scratchy, smooth etc. , and suppose people made faster and more accurate associations with some colors and textures than others. What could we conclude? Personally, I would conclude that the faster associations had probably been rehearsed more than the slower, less accurate ones. In the case of the racism test, an obvious explanation is the ubiquityof advertising associating white people with positiveadjectives. You can't go away day without seeing them matched on a billboard somewhere. Only particular faces though. Some races more than others.

I don't think we can conclude that this test shows racism any more than if you said "tea" to someone in that word association game and they said "bags" that they're a tea dribker.

I agree that a lot of people say they're not raciat when they really are, but THAT'S the coinxidence. That's why this test seems so compelling, when actually it's probably just a gimmick.
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/29/2015 4:24:26 AM
Posted: 1 year ago
i just checked and it's what I thought. Sorry, i should have done that first. Yeah, it's interesting how much can be concluded from speed of word associations. Your question "what else could it be possibly measuring" seems like the wrong sort of question. It's measuring speed of associations between images and particular words. The images are already categorized by race and the words are already categorized good and bad (or whatever) so the experiment is set up to find "implicit racism" or not. In fact, we don't know what's going on and what causes the effect. "Implicit racism" is neither a cognitive model nor a psychological theory, it's just a judgmental label.

It could be, as I said, that participants of all groups are more familiar with seeing white faces on a computer screen, and that the associations have to do with the familuariry of the stimulus in that particular context. After all, a face on a screen in a testing situation is not a real person or a real interaction.
feverish
Posts: 2,716
Add as Friend
Challenge to a Debate
Send a Message
1/29/2015 4:24:36 PM
Posted: 1 year ago
At 1/29/2015 4:05:48 AM, Garbanza wrote:

I just mean it's not obvious what causes the effect and it's a big leap to assume it's implicit racism, although of course you're right that it depends on what you say racism is exactly.

Perhaps I should have been clearer, I don't believe this test is an accurate measure of how much of "a racist" an individual is at all. For example I don't consider myself a racist yet I still had a bias of white over black on this test, also as I've said, apparently a lot of black people themselves would score a similar result and presumably not that many would actually fall into the Uncle Ruckus category of racial self-hatred.

I do think it tells us something about attitudes to race though.

But does speed of matching words and images have to do with racism? It's set up so that it really seems as if it does, but imagine the same test with less emotive content.

Well I think that's part of the point, the positive and negative words have strong emotional resonance so it's hard to switch off the implicit associations we have with those emotions.

Suppose they had pictures of colors and then words related to texture and finish such as shiny, scratchy, smooth etc. , and suppose people made faster and more accurate associations with some colors and textures than others. What could we conclude? Personally, I would conclude that the faster associations had probably been rehearsed more than the slower, less accurate ones.

Maybe, but wouldn't that suggest a bias towards those things? We are drawn to and repeat or rehearse our preferences more than our prejudices surely?

In the case of the racism test, an obvious explanation is the ubiquityof advertising associating white people with positiveadjectives. You can't go away day without seeing them matched on a billboard somewhere. Only particular faces though. Some races more than others.

I agree, but it sounds like your describing a potential cause for racial bias rather than an alternative explanation for what is occurring.

I don't think we can conclude that this test shows racism any more than if you said "tea" to someone in that word association game and they said "bags" that they're a tea dribker.
I think if someone had significant trouble matching tea with positive things it might well suggest some negative attitudes towards tea!

I agree that a lot of people say they're not raciat when they really are,

I don't think I even said that, but maybe you are right.

but THAT'S the coinxidence. That's why this test seems so compelling, when actually it's probably just a gimmick.

At 1/29/2015 4:24:26 AM, Garbanza wrote:
Your question "what else could it be possibly measuring" seems like the wrong sort of question. It's measuring speed of associations between images and particular words. The images are already categorized by race and the words are already categorized good and bad (or whatever) so the experiment is set up to find "implicit racism" or not.

Right, and what could explain a measurable difference in speed and accuracy in being able to match white or black people with positive or negative things other than a slight cognitive dissonance caused by some sort of racial bias?

In fact, we don't know what's going on and what causes the effect. "Implicit racism" is neither a cognitive model nor a psychological theory, it's just a judgmental label.

I think it's interesting that people seem so defensive about the idea that such a test could possibly be meaningful at all, but I certainly agree that it would be judgemental and wrong to label an individual as "a racist" based on a test like this.

It could be, as I said, that participants of all groups are more familiar with seeing white faces on a computer screen,

Yeah, but that's a potential cause for (and perhaps result of) racial bias, not a lack of it surely?
F-16_Fighting_Falcon
Posts: 18,324
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 1:14:18 AM
Posted: 1 year ago
I took the European American vs Asian American test and feel kind of cheated. The order is which the test was presented to me was the SOLE factor in my final results.

At first, it put Asian American and American on one side and European and Foreign on another side. I was still getting used to the test. I didn't instantly recognize all the faces as either European or Asian. I didn't recognize whether some of the landmarks were American or Foreign although most were easy. I was slow and made mistakes.

I finally got the hang of it. Got used to the test. Then it paired European American with American and Asian American with foreign and as I got the hang of the test, I was fast and accurate.

Then the damn test told me that I have a preference for seeing European Americans as Americans and Asian Americans as foreign like a large percentage of people and suggested that I have implicit bias.

Bullsh!t.

I'm ethnically from an Asian country and I most certainly wouldn't associate being American with being European American. I know the FAQ about randomized order but I feel kind of cheated tbh.
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 1:51:34 AM
Posted: 1 year ago
At 1/29/2015 4:24:36 PM, feverish wrote:
At 1/29/2015 4:05:48 AM, Garbanza wrote:


I just mean it's not obvious what causes the effect and it's a big leap to assume it's implicit racism, although of course you're right that it depends on what you say racism is exactly.

Perhaps I should have been clearer, I don't believe this test is an accurate measure of how much of "a racist" an individual is at all. For example I don't consider myself a racist yet I still had a bias of white over black on this test, also as I've said, apparently a lot of black people themselves would score a similar result and presumably not that many would actually fall into the Uncle Ruckus category of racial self-hatred.

I do think it tells us something about attitudes to race though.

Maybe. The problem I have with it is the cognitive mechanisms are not broken down at all, but assumed. For instance, what mechanism do we use to identify black vs white faces? Is it just skin color? I doubt it, but even if it is, one race could have the advantage over the other based on brightness of the image, or on expectations or familiarity. There's also the issue of salience - which faces stand out from the rest due to particular features including unusualness.

And then the task doesn't simply ask people to identify black or white faces, but to separate them into categories, with a 50/50 distribution. We don't know how people do that. They could have separate mechanisms for "black" and "white" or they could have a system where white is the default, for instance, and black is seen as "different" so they're filing them according to a keep/go strategy. Or they could have some other strategy for sorting. The method they use has implications for the matching task. For example, if they're using a keep/go strategy, obviously, it will be affected by good or bad adjectives.

This is just one example of something that might be going on. Fact is, it's a complicated task and we're just guessing at the mechanism.

But does speed of matching words and images have to do with racism? It's set up so that it really seems as if it does, but imagine the same test with less emotive content.

Well I think that's part of the point, the positive and negative words have strong emotional resonance so it's hard to switch off the implicit associations we have with those emotions.

Suppose they had pictures of colors and then words related to texture and finish such as shiny, scratchy, smooth etc. , and suppose people made faster and more accurate associations with some colors and textures than others. What could we conclude? Personally, I would conclude that the faster associations had probably been rehearsed more than the slower, less accurate ones.

Maybe, but wouldn't that suggest a bias towards those things? We are drawn to and repeat or rehearse our preferences more than our prejudices surely?

No, actually, people can respond faster to unusual or disliked stimuli in some circumstances. It's complicated.

http://link.springer.com...

In the case of the racism test, an obvious explanation is the ubiquityof advertising associating white people with positiveadjectives. You can't go away day without seeing them matched on a billboard somewhere. Only particular faces though. Some races more than others.

I agree, but it sounds like your describing a potential cause for racial bias rather than an alternative explanation for what is occurring.

It depends on the mechanism. It could be an extremely shallow matching/familiarity process. For instance, if people are used to seeing faces of a particular luminosity or contrast or arrangement of features, it might affect their reaction times to faces and strategies in the sorting task.

I don't think we can conclude that this test shows racism any more than if you said "tea" to someone in that word association game and they said "bags" that they're a tea dribker.
I think if someone had significant trouble matching tea with positive things it might well suggest some negative attitudes towards tea!

I agree that a lot of people say they're not raciat when they really are,

I don't think I even said that, but maybe you are right.

Haha true, I shouldn't have said "I agree".

but THAT'S the coinxidence. That's why this test seems so compelling, when actually it's probably just a gimmick.

At 1/29/2015 4:24:26 AM, Garbanza wrote:
Your question "what else could it be possibly measuring" seems like the wrong sort of question. It's measuring speed of associations between images and particular words. The images are already categorized by race and the words are already categorized good and bad (or whatever) so the experiment is set up to find "implicit racism" or not.

Right, and what could explain a measurable difference in speed and accuracy in being able to match white or black people with positive or negative things other than a slight cognitive dissonance caused by some sort of racial bias?

A lot of things. A HUGE LIST of things could cause it. I've already put forward some examples.

In fact, we don't know what's going on and what causes the effect. "Implicit racism" is neither a cognitive model nor a psychological theory, it's just a judgmental label.

I think it's interesting that people seem so defensive about the idea that such a test could possibly be meaningful at all, but I certainly agree that it would be judgemental and wrong to label an individual as "a racist" based on a test like this.

Maybe people are defensive, but I'm really not. I'm happy to accept the idea that we have implicit racism, I just think there's no reason to think that this test measures it.


It could be, as I said, that participants of all groups are more familiar with seeing white faces on a computer screen,

Yeah, but that's a potential cause for (and perhaps result of) racial bias, not a lack of it surely?

Perhaps, but that's not what you or the authors are claiming. You're claiming that it's the negative associations with black faces (because of race) and positive associations with white faces (because of race) that are causing the difference in reaction times. If it's simply issue of salience or familiarity of stimuli, then that's a very different cognitive process than the one you're putting forward. I think.
feverish
Posts: 2,716
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 11:15:44 AM
Posted: 1 year ago
At 1/30/2015 1:51:34 AM, Garbanza wrote:
At 1/29/2015 4:24:36 PM, feverish wrote:
At 1/29/2015 4:05:48 AM, Garbanza wrote:

The problem I have with it is the cognitive mechanisms are not broken down at all, but assumed. For instance, what mechanism do we use to identify black vs white faces? Is it just skin color?

I doubt it, but even if it is, one race could have the advantage over the other based on brightness of the image, or on expectations or familiarity. There's also the issue of salience - which faces stand out from the rest due to particular features including unusualness.

And then the task doesn't simply ask people to identify black or white faces, but to separate them into categories, with a 50/50 distribution. We don't know how people do that. They could have separate mechanisms for "black" and "white" or they could have a system where white is the default, for instance, and black is seen as "different" so they're filing them according to a keep/go strategy. Or they could have some other strategy for sorting. The method they use has implications for the matching task. For example, if they're using a keep/go strategy, obviously, it will be affected by good or bad adjectives.

This is just one example of something that might be going on. Fact is, it's a complicated task and we're just guessing at the mechanism.

I can't help feeling you are making it way more complicated than it is. I think if you are shown a face, you can tell pretty much immediately if it is white/black male/female, maybe not with some faces but certainly those in this test. It doesn't take much cognitive process to work it out. The words also are heavily emotionally loaded, you know straight away if it is positive or negative.

The only real cognitive process is matching the category by the rule of the round, not distinguishing between whether the faces are black/white or the words are good/bad. I simply don't agree that it is a complicated task.

No, actually, people can respond faster to unusual or disliked stimuli in some circumstances. It's complicated.

http://link.springer.com...

OK, but as I say, I think it's clear that the task isn't measuring the speed you identify the stimuli, rather the accuracy and speed by which you class it as positive or negative.

It depends on the mechanism. It could be an extremely shallow matching/familiarity process. For instance, if people are used to seeing faces of a particular luminosity or contrast or arrangement of features, it might affect their reaction times to faces and strategies in the sorting task.

See above. In my opinion the difficulty people have isn't with whether a face is black or white.

If that was in fact the issue then that shouldn't affect the results anyway, since you do it both ways round, wouldn't any delay in discernment be cancelled out?

If you were slower to identify black faces then you would be slower to match them as good, but also slower to match them as bad so no bias would be recorded.

I don't think we can conclude that this test shows racism any more than if you said "tea" to someone in that word association game and they said "bags" that they're a tea dribker.
I think if someone had significant trouble matching tea with positive things it might well suggest some negative attitudes towards tea!

what could explain a measurable difference in speed and accuracy in being able to match white or black people with positive or negative things other than a slight cognitive dissonance caused by some sort of racial bias?

A lot of things. A HUGE LIST of things could cause it. I've already put forward some examples.

With all due respect, I really don't think you have, despite a concerted effort. You have shown that in-group bias is a factor, which it definitely is, but I have shown that clearly can't fully explain the results because of the overall out group bias displayed by black Americans.

You've made several good points that to me illustrate wider cultural bias in society and could well explain why racial bias might exist, even in non racists.

And you've made some points about how brightness and other visual factors could play a part in discerning whether a face is white or black.

All of your points are factual but I don't believe any provide an alternative explanation for what these tests are measuring other than a bias for one race over another.

Maybe people are defensive, but I'm really not. I'm happy to accept the idea that we have implicit racism, I just think there's no reason to think that this test measures it.

Ok, but it seems like not just you but everyone else in this thread, apart from myself and the Stormfront refugee, seem very certain that this test is bullsh!t, yet none of you really seem too sure why.

It could be, as I said, that participants of all groups are more familiar with seeing white faces on a computer screen,

Yeah, but that's a potential cause for (and perhaps result of) racial bias, not a lack of it surely?

Perhaps, but that's not what you or the authors are claiming. You're claiming that it's the negative associations with black faces (because of race) and positive associations with white faces (because of race) that are causing the difference in reaction times. If it's simply issue of salience or familiarity of stimuli, then that's a very different cognitive process than the one you're putting forward. I think.

A difference in speed of recognising the stimuli would not produce the results these tests produce. It would be cancelled out by the alternate combination of matchings in the preceding or subsequent rounds, whereas an implicit bias would be reinforced.

Even if the only reason one has a racial bias is because one sees more of a certain race, that doesn't mean that one does not have a racial bias.
feverish
Posts: 2,716
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 11:30:22 AM
Posted: 1 year ago
At 1/30/2015 1:14:18 AM, F-16_Fighting_Falcon wrote:
I took the European American vs Asian American test and feel kind of cheated. The order is which the test was presented to me was the SOLE factor in my final results.

At first, it put Asian American and American on one side and European and Foreign on another side. I was still getting used to the test. I didn't instantly recognize all the faces as either European or Asian. I didn't recognize whether some of the landmarks were American or Foreign although most were easy. I was slow and made mistakes.

I finally got the hang of it. Got used to the test. Then it paired European American with American and Asian American with foreign and as I got the hang of the test, I was fast and accurate.

Then the damn test told me that I have a preference for seeing European Americans as Americans and Asian Americans as foreign like a large percentage of people and suggested that I have implicit bias.

Bullsh!t.

I'm ethnically from an Asian country and I most certainly wouldn't associate being American with being European American. I know the FAQ about randomized order but I feel kind of cheated tbh.

Hey man, thanks for sharing your experience with it.

If you are sure the order was a big factor, then I would encourage you to try it again, hopefully you will get the reverse order and then have a better idea.

In any case, please don't feel bad about the results. One thing I think me, Garbanza and the test creators can all agree on is that these tests say more about a society than they do about an individual. It certainly doesn't mean you dislike yourself or consider your own culture alien or any of that stuff.

As someone with a strong dislike for racism and a mixed race child, I could choose to be peeved by my own results on the black/white test. Everyone seems to take this test way too personally! I think it's the overall results that are interesting.

But on a personal level, I think it would still be pretty natural if when bombarded by media images and suggestions of Asian culture as being foreign and different, you might have a slight mental delay classifying them as American, even when you don't consciously feel that way yourself.
F-16_Fighting_Falcon
Posts: 18,324
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 12:31:41 PM
Posted: 1 year ago
No, I'm pretty sure the order was the sole factor. I just didn't feel like trying it multiple time hoping to get a different order. I think you misunderstand. I don't feel "bad" about the test results for what they say about me because I'm confident that was entirely order-based.

I feel like the test creators's hypothesis was that people associate Asian American with foreign and I feel "cheated" that they used my results to confirm their hypothesis when I know for a fact that it shouldn't have been. It would have felt different if order caused them to reject their hypothesis. '

It is not so much the specific content of this test but more like feeling that I've been unwittingly used as a tool by the experimenters to further their theory i.e. "yep, we knew it all along. People associate European with American. Your results confirm this."
ford_prefect
Posts: 4,139
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 3:14:53 PM
Posted: 1 year ago
Yeah I took two separate tests that I know for a fact I have biases about. One test gave me the correct order first, and the other test gave me the incorrect order first. Just as I predicted, the test that matched my biases first gave me a very high score, since I struggled mightily to overcome both my bias and the fact that the categories were switched on the second round. On the other hand, the test that didn't match my bias first, my score was neutral because at first I was slow because it's hard to overcome the bias, but just as I was getting used to it, it switched back and then the difficulty of switching is compensated by the fact that you're going along with your bias now.

So essentially the order does make a huge difference in your individual results. However, the overall statistics, given that they are randomly assigned, are still significant because the order bias cancels out in the long run.
gingerbread-man
Posts: 301
Add as Friend
Challenge to a Debate
Send a Message
1/30/2015 3:39:26 PM
Posted: 1 year ago
At 1/26/2015 6:35:56 PM, feverish wrote:
Judge not lest ye be judged...

Thought I'd share this interesting site I just came across with the DDO massive, apologies if it's old news.

https://implicit.harvard.edu...

Seen something like this before on some documentary, basically a set of scientific tests of unconscious bias and prejudice. Anyone who thinks that they have no views that are racist/homophobic/religously biased etc or that prejudice isn't highly pervasive and subtle may be surprised at their results.

For example, as someone who considers myself an anti racist I was initially surprised to see that I have a moderate unconscious bias towards my own race. I would be very interested to hear if other people are surprised at their results and also what people think about these tests and about unconscious bias in general. Apparently even people from the stigmatised groups tend to slightly favour the other group.

Try it, it's kind of fun. You don't have to answer any of the questions about yourself or your opinions if you don't want to or can't be bothered.

took the test, and indeed I am a racialist - against my own race no less.

I think everyone should accept the fact that we are all racist to some degree due to our unconcious biases. Things would get better much quicker if we did accept that fact.
Not my gumdrop buttons!

Debates currently in voting period:

http://www.debate.org...
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/31/2015 4:14:27 AM
Posted: 1 year ago
At 1/30/2015 11:15:44 AM, feverish wrote:
The only real cognitive process is matching the category by the rule of the round, not distinguishing between whether the faces are black/white or the words are good/bad. I simply don't agree that it is a complicated task.

I don't mean consciously, I mean in terms of brain processing. That's what reaction times measure - speed of processing. So recognition tasks are faster than choice tasks for example because they contain fewer processing steps. Imagine if you had to program a computer to recognize race of faces from localized brightness patterns like on our retinas. It would be complicated.
Here's a theory I've made up about how processing of race could lead to differences on the matching task. This is just to show you how it might happen that way. I don't know what the actual mechanism is, obviously. Suppose the steps went like this:

1. Is it a face? (yes, continue; no, stop)
2. Is it a familiar, expected face? (yes - GOOD; no, continue)
3. Is it a face of a known category? (if yes, access memory of types)
4. Is it friendly or threatening? (friendly - GOOD; threatening - BAD)

I know you'll take issue with these processing steps and rightly so, but the only point I'm trying to make is that the MECHANISM of categoration could affect the speed of matching to adjectives. In the example above, the expected faces link to GOOd very early on. To link them to bad adjectives would require an extra processing step.

You might say this is racial bias anyway, and I agree my example is bad, but the point is that without knowing the mechanism of the task, we can't come to any conclusions about what the difference in reaction times mean.
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/31/2015 4:24:33 AM
Posted: 1 year ago
At 1/30/2015 3:39:26 PM, gingerbread-man wrote:
took the test, and indeed I am a racialist - against my own race no less.

I think everyone should accept the fact that we are all racist to some degree due to our unconcious biases. Things would get better much quicker if we did accept that fact.

yikes. take it to the religion section. :P

I mean even if you're right, who cares? Should we all go out and buy hair shirts? (sarcastic rhetorical). Actually, this is the worst, when bad science ends up with "right on" conclusions. Nobody wants to get too critical. I've seen it before. Fact is though, without understanding the nature and processes of these "unconscious biases" it's not worth anything. It's just a cute effect.

If we understood the mental processes underlying it, though, instead of making vague PC guesses, then it might actually have some practical utility.
gingerbread-man
Posts: 301
Add as Friend
Challenge to a Debate
Send a Message
1/31/2015 4:34:02 AM
Posted: 1 year ago
At 1/31/2015 4:24:33 AM, Garbanza wrote:
At 1/30/2015 3:39:26 PM, gingerbread-man wrote:
took the test, and indeed I am a racialist - against my own race no less.

I think everyone should accept the fact that we are all racist to some degree due to our unconcious biases. Things would get better much quicker if we did accept that fact.

yikes. take it to the religion section. :P

I mean even if you're right, who cares? Should we all go out and buy hair shirts? (sarcastic rhetorical). Actually, this is the worst, when bad science ends up with "right on" conclusions. Nobody wants to get too critical. I've seen it before. Fact is though, without understanding the nature and processes of these "unconscious biases" it's not worth anything. It's just a cute effect.

If we understood the mental processes underlying it, though, instead of making vague PC guesses, then it might actually have some practical utility.

The test is a tad dodgey - interesting, but dodgey. Definitely don't think we should beat ourselves up about it - humans tend to gravitate to their own tribe and often are warey of anything different to our "norm". Its just tiring when you keep hearing...I'm not racist...But.... When everyone carries around some bias of some sort about race, sex, age etc.

Just think it would be better if people were honest about it and acknowledged it even thought it is no PC to do so.
Not my gumdrop buttons!

Debates currently in voting period:

http://www.debate.org...
Garbanza
Posts: 1,997
Add as Friend
Challenge to a Debate
Send a Message
1/31/2015 4:46:12 AM
Posted: 1 year ago
At 1/31/2015 4:34:02 AM, gingerbread-man wrote:
At 1/31/2015 4:24:33 AM, Garbanza wrote:
The test is a tad dodgey - interesting, but dodgey. Definitely don't think we should beat ourselves up about it - humans tend to gravitate to their own tribe and often are warey of anything different to our "norm". Its just tiring when you keep hearing...I'm not racist...But.... When everyone carries around some bias of some sort about race, sex, age etc.

Just think it would be better if people were honest about it and acknowledged it even thought it is no PC to do so.

I suppose so. For myself, if people are going to be racist deep down, I'd rather the lied about it and then got all weird and awkward trying to compensate and conceal it in social situations. That stuff used to annoy me but now I think it's as good as it gets for racist people.