Championship Subdivision Football | FCS Football | Stadiums | Blogs | Forums

So after I typed my most recent post in the Capn's latest "Conservatives are Racists" thread I turned to my wife and said, "I wonder if anybody's ever looked at whether or not there are racial differences in testosterone levels." That's based on the fact that higher testosterone levels tend to make males more aggressive.

So I Googled it. Lo and Behold:

http://www.ncbi.nlm.nih.gov/pubmed/1621259" onclick="window.open(this.href);return false;

Racial and ethnic variations in serum testosterone levels were investigated among a large sample of male Vietnam era veterans. Based on geometric means, significant average differences were found between 3,654 non-Hispanic white and 525 black individuals.

Also:

Results were interpreted as having considerable potential for explaining some of the race differences in the incidences of cardiovascular diseases, hypertension, and prostate cancer.

Now, I'd have to buy the paper for $39.95 to look at which mean testosterone level was higher right now. My bet is the mean testosterone level of Black males was found to be higher than that of White males.

What say you? When I boot up my work computer tomorrow I'll probably be able to get to the full article and find the answer because my employer probably has a subscription to the journal in question.

Old study. Other researchers have looked at the same and accounted for other variables (including age, percent body fat, alcohol, smoking, and activity) and no significant differences were found between testosterone levels in whites and blacks.

Higher testosterone in black compared with white men has been postulated to explain their higher prostate cancer incidence. Previous studies comparing hormone levels by race might have been limited by size, restricted age variation, or lack of representation of the general population

After applying sampling weights and adjusting for age, percent body fat, alcohol, smoking, and activity, testosterone concentrations were not different between non-Hispanic blacks (n = 363; geometric mean, 5.29 ng/ml) and non-Hispanic whites (n = 674; 5.11 ng/ml; P > 0.05) but were higher in Mexican-Americans (n = 376; 5.48 ng/ml; P < 0.05). Non-Hispanic blacks (40.80 pg/ml) had a higher estradiol concentration than non-Hispanic whites (35.46 pg/ml; P < 0.01) and Mexican-Americans (34.11 pg/ml; P < 0.01). Non-Hispanic blacks (36.49 nmol/liter) had a higher SHBG concentration than non-Hispanic whites (34.91 nmol/liter; P < 0.05) and Mexican-Americans (35.04 nmol/liter; P < 0.05).
CONCLUSIONS:
Contrary to the postulated racial difference, testosterone concentrations did not differ notably between black and white men. However, blacks had higher estradiol levels. Mexican-Americans had higher testosterone than whites but similar estradiol and SHBG concentrations. Given these findings, it may be equally if not more important to investigate estradiol as testosterone in relation to diseases with racial disparity.

http://www.ncbi.nlm.nih.gov/pubmed/17456570" onclick="window.open(this.href);return false;

OMGosh you can NOT make this up. Check out this quote I found at the article at http://www.academia.edu/4064241/Black_M ... rspectives" onclick="window.open(this.href);return false;.

Higher testosterone levels are linked to violent behavior. High tetosterone results in lower IQ in males but better fighting and mating skills.

Don't get TOO upset though. If you find that quote you'll find it preceded by some mitigating language.

Old study. Other researchers have looked at the same and accounted for other variables (including age, percent body fat, alcohol, smoking, and activity) and no significant differences were found between testosterone levels in whites and blacks.

I think the study I linked had a larger sample size.

However, the study you linked appears to have been based on a "scientific" survey sample of American males in general while the one I linked was based, at best, on a "scientific" survey of Vietnam veterans; and probably not that. Also the one I linked made no effort to adjust for factors influencing testosterone levels. But all that means is that they may have found explanations for why the mean differs. Doesn't mean there's not a difference.

Finally, the authors of the study you linked committed the sin of saying their results showed "...concentrations were not different." That is not true. Actually, in their study, the sample mean testosterone concentration for non Hispanic Black males was higher than that of non Hispanic White males even after adjusting for the factors they adjusted for. It's just that it wasn't ENOUGH higher to result in the conclusion that there was sufficient evidence, at 95 percent confidence, that the mean level of the larger population of Black males was greater than the mean level of the larger population of White males.

Seriously they should not have done that. You are supposed to be very careful in how you state the results. They did not show that the concentrations were not different. In fact if you had to bet you'd have to bet that the concentrations are indeed different after adjustment for the factors. To see how confident you'd be in that bet I'd have to see the actual p value instead of just seeing "p > 0.05."

It's a very important concept. When you do a statistical hypothesis test you are starting off assuming "no difference" and placing a very high burden of proof on the side of proving that there is indeed a difference. Usually 95% confidence. But when you fail to achieve 95% confidence you have not shown that there is "no difference." You have merely failed to generate sufficient evidence to conclude that there IS a difference at that 95% confidence level.

It's just amazing to me how very high power researchers routinely do that crap of saying that they've shown "no difference" because they failed to generate sufficient evidence for 95% confidence. It's endemic and it's really bad.

Some serious trolling going on tonight.

Must be why they have units the size of my forearm.

I know you guys won't care but I feel compelled to type this. To me there is an aspect of that paper Skyfetti linked that is disturbing. I see the problem all the time but that paper is a good example of it. I Googled each author of the paper to get the institutions they are indicated as affiliated with. Here is the list:

Sabine Rohrmann - University of Zurich
William G. Nelson - Johns Hopkins
Nader Rifai - Harvard
Terry R. Brown - Johns Hopkins
Adrian Dobs - Johns Hopkins
Norma Kanarek - Johns Hopkins
James D. Yager - Johns Hopkins
Elizabeth A. Platz - Johns Hopkins

A lot of prestige and credibility associated with those institutional names.

But as I wrote above they incorrectly stated their results. I already quoted the abstract and I subsequently looked at the entire paper. The conclusions section includes this Statement:

In conclusion, in this large, nationally representative sample, there was no difference in circulating testosterone concentrations between non-Hispanic black and white men overall.

First of all, one can't even evaluate that Statement because they did not report what the unadjusted mean testosterone levels were. What they're referring to is estimated mean testosterone levels derived through "...applying sampling weights and adjusting for age, percent body fat, alcohol, smoking, and activity."

But the big one is making the statement that there was no difference. There actually WAS a difference in their sample non-Hispanic black and white male estimated adjusted mean testosterone levels. Their estimates 5.25 ng/l for non-Hispanic black males and 5.11 ng/l for non-Hispanic white males.

But the difference in sample adjusted means is not LARGE enough to be statistically significant at their chosen rejection level; which is the 95% level. The proper way to state it is something like, "There is not sufficient evidence, at the 95% confidence level, to conclude that the levels in the populations represented are higher once they are adjusted as described."

Or you could do something like put a 95% interval around the estimated difference between the two adjusted estimated means and say that you're 95% confident that the difference is betwee - x and +x. And you could discuss what that means.

But what they did was completely wrong. Two things about it: 1) It's very disturbing that we have people that smart and with that kind of prestige doing that and 2) that language would not have made it through peer review if peer review was really a quality control process like people think it is. If you did that on a college freshman introductory Statistics course test you'd get marked wrong for doing it. The idea that you can never show "no difference" with a statistical hypothesis test is a "big one" in statistics. You wouldn't get away with it.

And the end result is that people will cite that paper as showing there is "no difference" when the paper didn't show that.

One other thing: The paper is deficient in that it does not show the actual p values derived. It just lets you know if they were <0.05 or <0.01. That does not allow you to really judge things. For example: Say someone compares a difference and end up with p = 0.10. And let's say they were using the 95% rejection level. You know it didn't meet the standard for them to accept the difference as shown to exist. But you also know that if they'd picked the 90% rejection level the evidence would have been sufficient. It's significant at the 90% confidence level.

Yes I know that I'm just a regular guy criticizing people with very impressive credentials who are affiliated with very impressive institutions. Also they all probably have way higher IQs than I do. But I'm right. And that's bad.

Peer review is ridiculously lame these days.

It is a circle jerk.

Just because I like doing stuff like this I made an illustration of the problem with doing what those researchers did. I randomly generated two samples of 10 observations each. One comes from a population with a mean of 1 and the other comes from a population with a mean of 0.5. So we know that the means are different. The two samples are the numbers in columns labeled "Sample of Population 1" and "Sample of Population 2" in the figure below.

Next down is the result of a statistical test to compare sample means (t test). p = 0.14. That means the result would not provide sufficient evidence to conclude that the means of the two populations are different at what people typically use as the rejection level for the "null" hypothesis that the two populations are equal. Typically the 95% confidence rejection level is used which means p <0.05 would be necessary.

So if someone was looking at the results without knowing beforehand as we do that the samples do come from populations with different means and they'd set the rejection level at the 95% level they'd say they didn't get sufficient evidence to conclude that a difference exists because p = 0.14. But they obviously would be incorrect if they said that they'd shown that there was no difference in the populations. And THAT is what those highly credentialed researchers from all those very prestigious institutions did with respect to the populations they were looking at.



Now, you CAN make a statement about how big the difference might be at most and talk about whether or not it could be big enough to be of "practical" significance. You can see that in the last two rows of the figure. You could do a 95% confidence interval for how big the mean difference might be. In the case of the example if you used the sample data you'd get a 95% confidence interval of -0.02 and 1.60. The fact that the 95% interval "covers" zero is another way of showing you don't have enough evidence at 95% confidence to say there is a difference. Zero is in the 95% interval so the difference could be 0. But you'd also have to consider the possibility that the difference is as high as 1.60 looking at it that way.

Another thing you could do is do a one sided 95% limit and say "I'm 95% confident the difference is no greater than this." That's the last row. With the data in the table you could be 95% confident that the difference between the two population means is no greater than 1.46. Somewhere in the range negative infinity to that.

To relate that to the study Sky linked: With one model the authors adjusted the testosterone sample means to Whites 5.10, Blacks 5.24. With another they adjusted them to Whites 5.11, Blacks 5.29. Those differences were not large enough to allow for 95% confidence that a difference in adjusted means of the populations being represented exists. Zero difference is among the 95% interval possibilities. But, at the same time, the difference in the adjusted means of the populations being represented could be LARGER than the 0.14 and 0.18 differences in the adjusted sample means. If someone was to put a one sided limit for how big the difference could be using each model you would get possible differences somewhat larger than 0.14 and 0.18.

I don't know if I can really express how bad it is that they stated it as showing there is no difference or how bad it is that the peer review process didn't result in them having to correct that statement. It creates a completely false impression among those who don't know to look for this sort of thing. And believe me when I tell you that this particular type of misstatement is VERY common in peer reviewed literature.

We get it....you hate them colored boys....


Find a new dead horse...there's not much left other than pulverized nothing

JSO at around 2,000 words in this thread.

My favorite: "OMGosh"

Skjellyfetti wrote:JSO at around 2,000 words in this thread.

My favorite: "OMGosh"

skelly must be bored with Afghanistan since Obama is in charge.

No more counting dead bodies...so jellybean is stuck counting words.

clenz wrote:We get it....you hate them colored boys....


Find a new dead horse...there's not much left other than pulverized nothing

Trolls gotta eat.

clenz wrote:We get it....you hate them colored boys....


Find a new dead horse...there's not much left other than pulverized nothing

Sigh. I'd completely changed horses.

By linking that study Sky inadvertently swerved into an example of a huge problem we have in the world of "science." But I guess getting most people to understand it isn't a realistic objective. It's not something you can do in 50 words or less and I'm learning that 50 words or less is pretty much the attention span of most people in this country.

JohnStOnge wrote:

clenz wrote:We get it....you hate them colored boys....


Find a new dead horse...there's not much left other than pulverized nothing

Sigh. I'd completely changed horses.

By linking that study Sky inadvertently swerved into an example of a huge problem we have in the world of "science." But I guess getting most people to understand it isn't a realistic objective. It's not something you can do in 50 words or less and I'm learning that 50 words or less is pretty much the attention span of most people in this country.

No...basically you're a racist and us intelligent people tend to ignore racist pedophiles

clenz wrote:

JohnStOnge wrote:
Sigh. I'd completely changed horses.

By linking that study Sky inadvertently swerved into an example of a huge problem we have in the world of "science." But I guess getting most people to understand it isn't a realistic objective. It's not something you can do in 50 words or less and I'm learning that 50 words or less is pretty much the attention span of most people in this country.

No...basically you're a racist and us intelligent people tend to ignore racist pedophiles

Feed the troll, feed it!

Nah. JSO isn't a troll. Trolls don't write these sorts of diatribes.... they try to provoke them.

Skjellyfetti wrote:Nah. JSO isn't a troll. Trolls don't write these sorts of diatribes.... they try to provoke them.