The first test I took was structured so that the the answer for Group A was on the left and Group B on the right. The second round had Positive terms on the left and Negative terms on the right. The third round had Group B and Positive terms on the left with Group A and negative terms on the right. The final round had Group A and Positive terms on the left and Group B and negative terms on the right. At first I thought that the issue was that rounds 1 and 2 were establishing the pattern of Group A on the left, then Positive Terms on the left, them Round 3 changed that pattern, putting Group A on the right instead. Then round 4 moved back to the originally taught pattern of Group A and Positive terms on the right. Given the way the pattern was established, then broken, then resumed I wasn’t surprised that my result showed a preference for the pairings as they were in Round 4.

The next test I took, which was the following day, however, did not follow the same format. In that test the first two rounds had Group B and Positive Terms on the left, them Round 3 had Group 3 and Positive terms together on the left. The final round had Group A and Positive terms on the left with Group B and Negative Terms on the right.

My conclusion is that the issue with the first test wasn’t the established pattern as I had thought. The issue was simply that the first time I was faced with classifying multiple categories at the same time I was slower than classifying single categories. By the time I got to round 4 I had started to get the hang of the multiple categories, so round 4 went faster. The same thing happened when I took my second test today. It had been more than 24 hours since I took that first test, so took me a few minutes to get the hang of classifying multiple categories at once. By round 4 I had it down and so was faster there, once again showing the “expected” result. From that point, however, the difference in times between round 3 and round 4 was gone, since I didn’t have the learning curve for the multiple classification. This is why I believe that I showed no preference in the latter tests, other than the Obama/Bush abnormality. My guess is that if I took another of the tests right now I might show a bias, unless the expected “preferred” pairings where presented in Round 3 instead of Round 4, since its been several hours since I took the test. That might not be the case however. I may not need to re-adjust to the multiple classification thing now since I seemed to get it down fairly well earlier.

As for the compounding of disadvantages you mentioned, I agree completely that it is an issue. I think that you summarized quite well why teachers, etc do not notice their bias, but added together with several other individuals showing a small bias it can add up to a significant disadvantage.

I’m not sure what you did differently once you got used to how the test worked, but if that means you were deliberately trying to fool the test (either by going a bit slow on the black=bad section so that your speed on the black=good section looks better by comparison, or by focusing on specific black people whom you hold in high regard), then that would mean you really do hold the bias that the test is supposed to reveal. You can’t deliberately conceal a bias if you don’t have it in the first place. (I’m not saying you definitely did these things, just giving examples. Only you can know for sure if they apply to you.)

I wonder if maybe that might explain your Bush/Obama result too – maybe you were trying to compensate for what you thought was a preference for Obama, but that preference wasn’t as strong as you thought it was, and you ended up overcompensating? (Again, I’m not saying this is necessarily what happened – only you can know that.)

I started rereading Valian’s book this weekend, and one of the points she makes early on is that advantages and disadvantages accumulate, just like compound interest: Small biases that are almost imperceptible individually can be repeated so often that they add up to a huge advantage or disadvantage at the peak of one’s career. A boy might be 1% more likely than a girl to be called on in class, 1% more likely to be praised for getting a good grade, 1% more likely to be given constructive feedback when he gets a bad grade, 1% more likely to be encouraged to take the most challenging classes, 1% more likely to be encouraged to apply for a prestigious summer program, 1% more likely to be written a letter of recommendation that highlights his achievements and abilities (as opposed, for example, to his appearance and his personality), and so on. If you look at the various teachers, counselors, professors, bosses, and colleagues that are responsible for each one of those 1%’s, none of them look like they’re biased, and probably none of them think that they’re biased. But as a result, the boy is far more than 1% more likely to become an engineer, or a CEO, or a partner in a law firm, or a Nobel Prize winning physicist, or President of the United States.

That’s why this is a complicated problem.

I took 6 of the tests on the IAT site. On the first two my score showed the “exepcted” result, which was the same strong preference most others apparently had. Once I got use to how the tests worked however, I scored a result of no bias (or whatever they call it, I can’t remember). The only test on which I scored a preference was the one that compared Barrack Obama to George W. Bush. According to the test I strongly prefer Bush. That result couldn’t be farther from the truth, so I’m not sure what it indicates. Maybe it indicates a flaw in the test or the conclusion it forms. Maybe it indicates that the order in which they present the data or which side the put which answers on impacts the result. Or maybe its just a a fluke and is a result that just needs thrown out.

Let me know how you interpret the results. I don’t know that the results qualify as acing it, but aside from that weird Bush/Obama result it appears I can now regularly complete the test with a score of no automatic preference.

I have no explanation for this score, and cannot, therefore, give you any guidance. Possibly it relates back to a childhood spent in a foreign country of which I have fond and positive memories, where 99% of the people were darker in skin color than I was. Or, since I’m white, maybe I’m self-hating? You’d tell me if I’ve shown any deficiency in self-esteem, wouldn’t you?

I’ll have to poke around in that site a little more.

“If we accept your premise that men’s average ability is greater than women’s, it is rational to see a male candidate with a particular test score, and to assume that he’s actually a little bit better than his test score indicates. Likewise, it is rational to see a female candidate with the same score, and assume she’s actually a little bit worse. (I’m oversimplifying here, but this is the basic idea.)

This is rational, . . . .”

First, it is not my premise that men’s math abilities are higher, if we are using the word ‘ability’ to connote inherent traits. What I said is that as a group they score higher on average than women, which I see as a proxy only for current competence, not innate ability.

Second, I’m not a statistician, but I believe that your conclusion quoted above is not rational. If a man and a woman have the same test scores, I don’t think that the fact that men on average score higher than women on average adds any meaningful data to the test scores themselves. If a score of 10 denotes a skill range between 8-12, it does so for both test-takers.

Third, I disagree that a rational action can be sexist. Maybe we are using the term differently, but to me, sexism is a belief system that supplants rationality.

A rational action may certainly have a disparate detrimental effect on a woman or on women generally, as in your example of an employer who refuses to hire a woman for an inconsequential job in order to retain the services of another uniquely valuable but sexist employee. The employer in that instance may be called callous or craven or desperate, but his/her frame of mind is not sexist, despite the fact that his action discriminates against women and is illegal.

If you needed the services of a top surgeon to save your life and you had available to you the services of a top surgeon who was a sexist pig or a non-sexist doctor just completing her surgical residency, you would not be sexist yourself, or endorsing sexism by choosing the more experienced surgeon.

As for your solution to the math employment problem at #115, I concur that would be a superior way to solve the problem if you had the necessary data, but as far as I can tell, you’ve still committed the sin you complain of, by assigning more than half the potentially winning slots to men. You’re still using sex as a proxy for skills, albeit with greater statistical accuracy.

I’ll go ahead and note a couple of things, though. First, I’m not claiming that I’m immune to the impacts of stereotype messages as they apply to others. I’m nowhere near that advanced on the path yet. I was referring specifically to the impacts of external stereotype messages on myself (and the ability of an individual to do the same). Second, I can’t claim to be able to teach others to not be impacted by stereotypes. That is well beyond my ability. All that I can do is say, it can be done and personal responsibility does work, and hope that it sparks an interest in someone to look into the concepts for themselves. I’m not a teacher, I’m simply a messenger. If down the road someone hears about these concepts again and thinks to themselves, this is the same sort of thing that weird guy on TSD was talking about, maybe I should look into it more, then it has been worth it. It is a radical concept and its not something that you hear about and suddenly start doing.

If you’ve discovered a way to totally get rid of all your internalized stereotypes, that is really something. But unless you can teach everyone else how to do exactly what you did, it’s of limited use.

I’d be interested to know what you think of the Implicit Association Test. As someone who’s escaped the Matrix (or whatever the right analogy is), you should be able to always ace it, right?

Johanna, thanks for the clarification. I apologize for misinterpreting your meaning.

Unfortunately your view that “…it’s absolutely not true that it’s possible not not internalize…” stereotyping messages is just as foreign to me as my view that it is very possible is to you. Actually, I take that back. Because I use to internalize those same messages and know people who still do it, its not a foreign concept to me. It’s the idea it is so ingrained into our psychology that people just accept it without realizing it can be changed. I hate to use this analogy, but I don’t know how else to explain it. It’s like the movie the Matrix. Once you are told about the Matrix and see it, its impossible not to see it and its hard to understand how everyone isn’t seeing it, especially when you tell them it is there.

Anyway, to be honest, I think in the end, we need to have room in our philosophy to accept that people aren’t perfect and we can’t expect them to be. Saying “Well, just don’t internalize that” expects them to be, even for adults.

Except those were basic, uniformed questions, not questions designed to spark debate or thought. They were literally the questions of someone who has never done real, solid thinking or reading about the topic, someone looking for a 101 education. There mere fact that you asked them demonstrates you don’t yet understand it. Also, “I will say, however, that I stand by my assertion that it is possible to choose not to internalize stereotyping messages. Again, that comes from learning the concepts of taking 100% responsibility.”

I highly suggest taking Johanna’s advice, reading that book and doing some actual research, because it seems like you’re just repeating your own biases to yourself. Because it’s absolutely not true that it’s possible to not internalize them. It’s possible to actively work against them and attempt to overcome them, but that’s not the same thing and it’s not only not guaranteed, it can’t be done the way you suggested on this thread.

I will say, however, that I stand by my assertion that it is possible to choose not to internalize stereotyping messages. Again, that comes from learning the concepts of taking 100% responsibility.

Actually, I was assuming that you *are* fairly learned, or at least well-read, and familiar enough with academic writing to know that it’s almost invariably dry as dust. So I wanted to mention that Valian’s book is an exception.

Why, if you understand that people can’t choose not to internalize stereotyping messages, did you suggest that they do so?

Why did you even posit the fact that maybe eliminating gender studies would eliminate the creation of stereotypes if you understood that stereotypes don’t arise FROM the studies?

Why, if you’ve thought about it extensively, did you ask questions like “I wonder, where does the reinforcing stereotype come from? Is it only based on observations in the classroom? Or do the results of the studies that show males perform better also impact this? If no one was ever told that males performed better at math, would that stereotype exist?” when the answers are actually the sort of thing you learn very early if you do research?

Or I guess the question is, you say you’ve thought about it extensively and I am sure that you believe that you have. But have you done any actual research, reading, and learning? Because if you’re just listening to yourself, you’re not listening to reality.

I honestly thought that the conversation was going well and we were making progress. I was apparently wrong.

And no, to be honest, you have not thought about this “plenty.” Until a few hours ago, you didn’t understand that sexism can be unintentional, that it is not a matter of opinion, that studies don’t cause stereotypes, and that people can’t just choose not to internalize stereotyping messages. I’ll give you the benefit of the doubt that you understand all these things now, but these are just the basics. It gets much more complicated than this.

I’m not saying this to be unkind, or to discourage you from learning. But if you want to be an ally here and to help offer solutions, you really do have a lot more learning to do.

I will make the clarification that when I say, what “we” need to do, that is my opinion about the hypothetical situation that we’ve been discussing. I assumed that was obvious, but apparently it was not. I have thought about this plenty, but I appreciate your suggestion nevertheless. It seems that I’ve hit a nerve, though, so I’m happy to end the conversation here if you’d like.