Race differences

[Carver]

Originally posted by GP

You don't understand how categorizations work. One doesn't have to prove race "exists". One merely has to have some definition for it, than one can examine how other variables correlate to it. Maybe the results are intersting, maybe not. If other interesting variables correlate to it, then the categorization is interesting itself. This is how the elements were identified. (Not first based on microcopic understanding...but first based on apperent effects.)

..."some definition for it". The problem is any definition will be contrived and artificial. There is no scientific definition of "race". A scientist looking at a DNA sample can only extrapolate and offer an educated guess as to the race based on various attributes. For example, the DNA may suggest a low level of melanin in the donor's skin and blue eyes. From this a scientist may guess that many people would classify the DNA donor as white.

There is no white/black/asian gene to be identified. What you perceive of as race is a collection of various attributes. Race is a human defined construct. Therefore, you can create any "race" you want but you won't be looking at nature - because nature didn't create the races, you did.

[TCO]

Originally posted by Carver


..."some definition for it". The problem is any definition will be contrived and artificial. There is no scientific definition of "race". A scientist looking at a DNA sample can only extrapolate and offer an educated guess as to the race based on various attributes. For example, the DNA may suggest a low level of melanin in the donor's skin and blue eyes. From this a scientist may guess that many people would classify the DNA donor as white.

There is no white/black/asian gene to be identified. What you perceive of as race is a collection of various attributes. Race is a human defined construct. Therefore, you can create any "race" you want but you won't be looking at nature - because nature didn't create the races, you did.

If other variables correlate to this categorization, than the categorization is interesting. And you have to admit that people already use this categorization in daily life, in affirmative action, etc.

Just becuase there is not a single gene doesn't change the obvious fact that racial/ethnic groupings are different in terms of heredity. They are extended family groupings. Just beacuse the boundaries are fuzzy doesn't mean that everything is the same.

[TCO]

Originally posted by Carver

What you perceive of as race is a collection of various attributes. Race is a human defined construct. Therefore, you can create any "race" you want but you won't be looking at nature - because nature didn't create the races, you did.

Agreed. That doesn't stop us from making the categorization and seeing if there are statistically meaningful group trends based on the categorization.

[Flubber]

OK if simply asking a question as to whether ethnicity could possibly cause other differences other than the physical makes me a racist . . . . I guess we have foreclosed all scientific enquiry into the area and know it all LOL.

A few points

1. I always believed we are one race and was using the term according to its more popular usage. I will use the term ethnicity from now on to more precisely confirm what I am talking about.

2. Would people consider it offensive if I were to suggest that maybe studies of different family could highlight the differences in between families? Are not ethnic groups merely a larger extension of that.


3. I see differences in my everyday life. I meet a family where everyone is very smart and another where they all don't seem to get it. A third where there is one brilliant kid while the remainder are average. Nature or nurture ?? probably both !! This can be within an ethnic group ( doesn't matter) but genetics could be part of it.

4. My basic understanding of the HGP is that it did not go so far as to say there are no intellectual differences between ethnicities but simply traced everyone back to a common genome. Fair enough . . . we all have common roots but it is obvious that people in different regions evolved somewhat differently. Exactly How?? and why would we assume that differences would ONLY manifest theselves in the physical look of a group?

I am not saying that other differences do exist but simply that it is irresponsible to assume otherwise. I have asked for links because unlike some on here I do want to educate myself further on what we know about genetics. last week I was informed that I have a genetic translocation in that some material that is supposed to be on chromosome 10 has switched places with some on chromosome 4. Nobody seems to know what this does or what impacts it may have. So excuse me if I try to ask a question now and again.


5. I find the response to my posts to be sad but probably predictable. I am called a racist (intended as a slur) because I simply question whether ethnic differences could exist. I asserted nothing and came down firmly as a "don't know" and asked for links showing that such ethnic backgrounds were irrelevant. Since I posed the question, none have been provided.


It may very well be that a study would find a difference in brain function between groups but that it could then be solved by substances more frequently ingested by one group versus another. We may find that it all comes down to training and education.

Too bad there are so many who want to deny even the possibility of a difference

[Ecowiz Returns]

I ask some questions to Fluber and GP answers them?

Originally posted by GP
In what way would examing conductivity in different metals help us understand electrical conductivity in general. As a scientist, having different samples with different performance makes it easier to understand function in general. This is how medical research in general is often done.

See, the point here is in the identifiable diference.
As far as science has proven, there is no identifiable genetic diference between any two humans so that we may present that diference as racial.
I have no problem in choosing diferent individuals.
What I present to you is that, as far as science already proved, when a genetic phenomenon is to be studies, it is likelly to reach equal results by sampling diferent people from one single "race" as it is form sampling from diferent people from diferent "races". The "race" factor is yet to be proven accountable in any unrefutable scientic work.

Originally posted by GP
No. I don't think that the HGP has identified the genes that effect intelligence and how and then correlated these genes by racial group. If they have, please point me to the results.

So you say that, although no detectable diference was detected in the genome of diferent humans that would trace us to their "race", if we choose a particular part of it, we could find one?
Given the implausability of it, I have to wonder the reason of such wishfull thinking, don't I?

Originally posted by GP
No for the reasoning above. Also, I wasn't aware that DNA could not be used to distinguish Africans, Europeans, etc. Is that true? Mitochondiral DNA certainly can be used this way. I'd lilke to see some source that says nuclear DNA can't be used to resolve the difference between a European and an African. I bet if I give two unmarked samples to a crime lab, they can use DNA testing to determine race. (I'm guessing here...but since you are certain...please rpovide the source.)

Shouldn't you be the one providing the source, since you are the one saying that Mithocondrial DNA can be used?

Originally posted by GP
You ignored the main point of his example: you can have wide variance within categories and still have statistically meaningful differences between the categories. This is basic math. Think it through.

Now that's where you're wrong.
As any statistician knows almost without thinking is that, the moment you have high variance, mean loses statistical significance. The higher the variance, the more likelly it is you get any value other that one near the mean. In those cases it is said that the mean is insuficient to explain the population.

Originally posted by GP
You don't understand how categorizations work. One doesn't have to prove race "exists". One merely has to have some definition for it, than one can examine how other variables correlate to it. Maybe the results are intersting, maybe not. If other interesting variables correlate to it, then the categorization is interesting itself. This is how the elements were identified. (Not first based on microcopic understanding...but first based on apperent effects.)

The problem of picking an irrelevant characteristic to stratify your sample with is that you may end up believing that it was that strata that explained the phenomena, when, in fact, a multitude of other factores may be the ones expalining it. The only thing you need is that the determining factors are themselves highly correlated with the stratification you chose.
An example:
You divide the world population by "races": you end up saying that race is the reason from some diference in, say, height, while overlooking some factors that are highly correlated with the "race" stratifications (nutrition, income, health conditions and sorts).
That's why it should be moro productive to start by stratifying according to those factors first, instead of some hystorical mith.

[TCO]

Originally posted by Ecowiz Returns


So you say that, although no detectable diference was detected in the genome of diferent humans that would trace us to their "race", if we choose a particular part of it, we could find one?
Given the implausability of it, I have to wonder the reason of such wishfull thinking, don't I?

Maybe there is an English or other communications barrier here.

If the HGP wants to prove that race is not correlated with genetic intelligence, it needs to do:

1. Know which codons affect intelligence and how (this hasn't been done to my knowledge. Correct me if you know better.).

(Then)

2. Take DNA samples from people of different races (you DON'T NEED a "DNA race detector" here, you label the samples when taking them from the subjects. Like DUH!!! BTW, this is also the only way to identify a "DNA race detector" to start with. (Think about it). )

3. Look at how the codons identified in 1, vary with 2.

Now you are the one saying HGP proved that there is no correlation of genetic basis of intellect with race. I say that until you produce some experiment approximating what I showe above, your claim is full of kaka.

[Ecowiz Returns]

I don't question the issue of the name being "race" or whatever. Calling it ethnicity may be as wrong as calling it race, if we start implying that the diferences are imbued in our genetic code, therefore implying there is a natural basis to diferentiate people.

An ethnic group is distinguished according to culture, not natural characteristic and in no time did the comon use of the word "race" as ever been related to cultural diferences. The matter was, always, natural ones.

The human knowledge of our diversity has evolved from dangerous concepts such as "blacks are not human beings" or "indians do not have soul" to the notion brought up by the Genome Project that said much more than we have some comon origin. It really said that none of us have particularly diferenciate genetic characteristics. They could test it because they, in fact collected samples from enough number of people from diferent "races".

Prior to the Human Genome's discovery I also thought that there were diferent human races, although I didn't aknowledge any special diferentiating characteristic between them. Now, I don't see the point of such a concept.

That's why I see no point in talking about "races", any more.
And talking about genetic diferences between diferent ethnic groups is, to me, just as wrong.

Now back to you:
"All I am saying is that genetic differences that might make a difference in one area (systemically and on average) could make differences in another area. I know that individual variances will be large since individual characterictics would swing wildly around any group mean but perhaps race X is more inclined to certain types of reasoning (logical, spatial, intuitive) while race Y is inclined to others."

You have to aknowledge that arguments such as these are not new. In fact, they are the basic argument of weel known racists.
The fact that you use it does not make you a racist yourself, but, if you didn't had a slight racist bias you would have some dificulty in arguing so much on the base of racial diference.

"I present a balanced outlook that comments that racial variance we see is likely due to socioeconomic factors and outright racial discrimination but that there does not seem to be enough science to disprove the idea that the genetics of race may matter."
Again, you argue that there is a possibility that something like a genetic racial diference exists, even if that was not shown in the Human Genome Project. Why?

"There are obvious physical differences between the races"
Unproven fact!
Given the high variance the mean does not inform!
Other posters pointed out that even the examples you gave are proven to be incorrect!
Do you keep that, other than the colour of one's skin there is any noticeable phisical evidence of "racial diference"?
Do you still think that much is an important phisical characteristic?
Why?

The moment you say that one should test diferent races, you say that race may matter. That is racist bias. It may not be xenophobic racism but is racism nonetheless.

I've noticed you didn't realized some of the implications of the Human Genome Project: the fact that many people, from diferent "races" were sampled and no evidence was found that would allow to genetically pinpoint someone's "race", the concept is no longer valid.
With no scientific backing to the notion of "race" any belief in it is no longer an informed rational one; its a prejudicial bias.
And there is no nice way to say it.

And, again, it is not a question of PC. Thank God this time the correct stance is the PC stance, otherwize I would be politically incorrect (in certain contexts, unfortunately, my position is still politically incorrect and that's not the reason for me to change it).

[TCO]

Originally posted by Ecowiz Returns
Shouldn't you be the one providing the source, since you are the one saying that Mithocondrial DNA can be used?

Here are two sources showing that mDNA can be used to identify ethnic groupings.

(easy to read source, but less definitive)

http://www.psc.edu/science/Merri/merri.html

(more definitive scientific source. Just read the first para. This answers your desire for a source showing the relation of mitochondrial DNA to race/ethnicity.)

http://www.biomedcentral.com/1471-2156/2/13

Major haplogroups [2] are continental or ethnically specific. Three of them (L1, L2, and L3) group sub-Saharan African lineages, nine (H, I, J, K, T, U, V, W and X) encompass almost all mtDNAs from European, North African and Western Asian Caucasians. Finally, haplogroups A, B, C, D, E, F, G and M embrace the majority of the lineages described for Asia, Oceania and native Americans.

[Flubber]

Originally posted by Ecowiz Returns
I ask some questions to Fluber and GP answers them?

Well I am not on here that often so I guess he picked up the argument

Originally posted by Ecowiz Returns
As any statistician knows almost without thinking is that, the moment you have high variance, mean loses statistical significance. The higher the variance, the more likelly it is you get any value other that one near the mean. In those cases it is said that the mean is insuficient to explain the population.

.

All true but as the sample size increases and increases it becomes harder and harder to discount the mean as an accurate indicator of a trait in the population. The man/ woman height thing is an example of this.

Other things


1. Do you acknowledge that there are some physical characteristics that are shared by groups when divided into groupings based on ethnicity or do you see this as a circular argument in that the physical differences exist because you have divided people based on their physical differences.

2. Is it your claim that there are NO genetic commonalities shared by ethnic groups as compared to other groups (on average)?

3. Lets forget the buzzwords. It may be interesting and informative to study larger people versus smaller, righthanders versus lefthanders, red haired versus black etc. So why would it be impossible to find commonalities in a group that shared a common and easily identifiable physical look and group of attributes. perhaps people who look different have no other differences but perhaps there are also unseen differences as well.

[Ecowiz Returns]

Originally posted by GP
BTW, this is also the only way to identify a "DNA race detector" to start with. (Think about it). )

In fact, the Human Genome Project did prove that there is no correlation with any part of our genetic constitution (being that number of genes and it's combination) and "race". It was proven that people from diferent "races" had more genetic similarities with people from other "races" then with others from their own.
This means that no "DNA race detector" was found. If it was there to be found, it would most likelly be found, but it wasn't.

Doesn't it stricke you as odd that no particular gene or combination of genes were consistently found on the people of one "race"?

What you are proposing is a wild goose chase: try and find something that just isn't there.

Now, could you tell me what purpose exactly would like that investigation to serve?

Again, I bet that you are as likelly to get diferent results from the test you propose as of getting if you were to take the fans of different football teams. That is, no statistically relevant diference.

[TCO]

Originally posted by Ecowiz Returns

Now that's where you're wrong.
As any statistician knows almost without thinking is that, the moment you have high variance, mean loses statistical significance. The higher the variance, the more likelly it is you get any value other that one near the mean. In those cases it is said that the mean is insuficient to explain the population.

It depends on what type of analysis you are doing. High variation within classes does not meant that the comparison of means is irrelevant statistically. You can do tests easily to show statistical significance. If you want a more "everyday example" look at what I wrote about car engines...or think for a minute about betting on horse races. Large variability does not eliminate the significance of different odds for different horses.

[TCO]

Originally posted by Ecowiz Returns

The problem of picking an irrelevant characteristic to stratify your sample with is that you may end up believing that it was that strata that explained the phenomena, when, in fact, a multitude of other factores may be the ones expalining it. The only thing you need is that the determining factors are themselves highly correlated with the stratification you chose.
An example:
You divide the world population by "races": you end up saying that race is the reason from some diference in, say, height, while overlooking some factors that are highly correlated with the "race" stratifications (nutrition, income, health conditions and sorts).
That's why it should be moro productive to start by stratifying according to those factors first, instead of some hystorical mith.

Agreed that this is a possibile danger in any analysis. The answer is to

1. do multiple regression.
2. look for mechanisms.

The answer is not to "stop doing analysis". That's not scientific.

BTW, one should have the same caution about stratifying by income, nutrition, etc. (what if those are correlated with other factors.)

[Rex Little]

As any statistician knows almost without thinking is that, the moment you have high variance, mean loses statistical significance. The higher the variance, the more likelly it is you get any value other that one near the mean. In those cases it is said that the mean is insuficient to explain the population.

I'll have to remember this argument next time I'm in a debate about discrimination and affirmative action. The variance in income level among whites or among blacks is much greater than the difference in the mean income of the two groups, but the mean difference is commonly used as evidence of the existence of discrimination and the need for affirmative action.

[Ecowiz Returns]

Originally posted by Flubber
All true but as the sample size increases and increases it becomes harder and harder to discount the mean as an accurate indicator of a trait in the population. The man/ woman height thing is an example of this.

You are calling upon assymptotic theory.
The problem with your statement is that it is only valid in Gaussian Populations. Although it is said that all distributions tend, at some high enough number to a Gaussian one, the problem is with how high that number must be.
I actually finished working in a very interesting subject (Economic Growth) and found out that the mean growth rate can be pretty deceptive! Another known example, at leat in Economics is what is known as the Union Premium (the diference in wage one gets from being unionized).

"Other things
1. Do you acknowledge that there are some physical characteristics that are shared by groups when divided into groupings based on ethnicity or do you see this as a circular argument in that the physical differences exist because you have divided people based on their physical differences."
Some physical diferences can be perceived between different ethnic groups. As far as I know, however, phisical diferences is not a suficient criteria do define an ethnic group.

"2. Is it your claim that there are NO genetic commonalities shared by ethnic groups as compared to other groups (on average)?"
Yes, it is my claim.
It wasn't, prior to the conclusions of the HGP, but now it is.

"3. Lets forget the buzzwords. It may be interesting and informative to study larger people versus smaller, righthanders versus lefthanders, red haired versus black etc. So why would it be impossible to find commonalities in a group that shared a common and easily identifiable physical look and group of attributes. perhaps people who look different have no other differences but perhaps there are also unseen differences as well."
Actually, I couldn't find any interest in any of the studies you "proposed". Go figure!

[Ecowiz Returns]

Originally posted by Rex Little
I'll have to remember this argument next time I'm in a debate about discrimination and affirmative action. The variance in income level among whites or among blacks is much greater than the difference in the mean income of the two groups, but the mean difference is commonly used as evidence of the existence of discrimination and the need for affirmative action.

In fact, it was lately proven that the diferences occur mostly on the extremes of both distributions, ie, the richer and the poorer of both groups.

So the result pretty much remains.