Homosexuals

Dnia 2006-10-25 DZ napisa³(a):
> Andrzej Rosa <bakters@yahoo.com> wrote:
>> Dnia 2006-10-25 DZ napisa³(a):
>>>> Actually better. Up to now, more intelligent people had more
>>>> children. Now it is the other way around. Well, doesn't it close
>>>> the argument? Intelligent people populate worse then average, so
>>>> natural selection obviously prefers moderate stupidity. At least at
>>>> the moment...
>>>
>>> I'm well aware of this argument, and sometimes I'd play along. But
>>> that would only be true if we were a one big panmictic population.
>>> However, there are local trends and that is important. The average
>>> intelligence in Hong Kong is higher than in the rest of China, which
>>> is still quite high, and things like that don't seem to homogenize.
>>
>> The average intelligence is supposed to be growing more or less
>> everywhere. That's not the point. The point would be if intelligent
>> people in Hong Kong populated better than average (also in Hong Kong).
>
> That might well be the case.

If it is, that would be interesting.

> But take my argument to the extreme (not
> because the extreme is realistic, but for clarity). Suppose the
> heterogeneity is so high that there is no inter-mating across the
> segments. In such an extreme case, it doesn't matter if there are any
> differences in reproduction rates. There are more bacteria than
> humans, and they reproduce far more efficiently, but we don't compete
> for the same niche.

I understand the example.

You mean that standard deviation of intelligence (or whatever) for Homo
Sapiens can't shrink over time? As far as I've seen for US, it seems to
be shrinking.

I mean that average SAT score went down at the same time as average IQ
went up, so lets assume that is the main tendency everywhere besides
Hong Kong.

So what now? If you break any barriers between Hong Kong and the rest
of the world their genes get diluted and thats it. Whatever factor was
promoting this adaptation was local, not general, and local factors
will not promote "large scale" adaptation.

--
Andrzej Rosa 1127R

On Wed, 25 Oct 2006 21:57:11 +0000 (UTC), DZ
<14400@30382652.716624270.25741.29750.15046> wrote:

>Andrzej Rosa <bakters@yahoo.com> wrote:
>> Dnia 2006-10-25 DZ napisa³(a):
>>> Andrzej Rosa <bakters@yahoo.com> wrote:
>>>> DZ napisa³(a):
>>>>> Andrzej:
>>>>>> But it doesn't work like that for the rest of the population. Highly
>>>>>> successful and often very attractive women have a hard time to find a
>>>>>> guy for themselves.
>>>>>
>>>>> But how well such a woman would do 5000 years ago? If worse than now,
>>>>> then I win the argument :-)
>>>>
>>>> Actually better. Up to now, more intelligent people had more
>>>> children. Now it is the other way around. Well, doesn't it close
>>>> the argument? Intelligent people populate worse then average, so
>>>> natural selection obviously prefers moderate stupidity. At least at
>>>> the moment...
>>>
>>> I'm well aware of this argument, and sometimes I'd play along. But
>>> that would only be true if we were a one big panmictic population.
>>> However, there are local trends and that is important. The average
>>> intelligence in Hong Kong is higher than in the rest of China, which
>>> is still quite high, and things like that don't seem to homogenize.
>>
>> The average intelligence is supposed to be growing more or less
>> everywhere. That's not the point. The point would be if intelligent
>> people in Hong Kong populated better than average (also in Hong Kong).
>
>That might well be the case. But take my argument to the extreme (not
>because the extreme is realistic, but for clarity). Suppose the
>heterogeneity is so high that there is no inter-mating across the
>segments. In such an extreme case, it doesn't matter if there are any
>differences in reproduction rates. There are more bacteria than
>humans, and they reproduce far more efficiently, but we don't compete
>for the same niche.

Now you have me thinking about marketing segmentation for the
population of bacteria. oi...what kind of survey can one administer
for THAT?

--
Is this thing on?

Andrzej Rosa wrote:
> DZ:
>> Andrzej Rosa:
>>> DZ:
>>>> Andrzej Rosa:
>>>>> Up to now, more intelligent people had more children. Now it is
>>>>> the other way around. Well, doesn't it close the argument?
>>>>> Intelligent people populate worse then average, so natural
>>>>> selection obviously prefers moderate stupidity. At least at the
>>>>> moment...
>>>>
>>>> I'm well aware of this argument, and sometimes I'd play along. But
>>>> that would only be true if we were a one big panmictic population.
>>>> However, there are local trends and that is important. The average
>>>> intelligence in Hong Kong is higher than in the rest of China, which
>>>> is still quite high, and things like that don't seem to homogenize.
>>>
>>> The average intelligence is supposed to be growing more or less
>>> everywhere. That's not the point. The point would be if intelligent
>>> people in Hong Kong populated better than average (also in Hong Kong).
>>
>> That might well be the case.
>
> If it is, that would be interesting.
>
>> But take my argument to the extreme (not because the extreme is
>> realistic, but for clarity). Suppose the heterogeneity is so high
>> that there is no inter-mating across the segments. In such an
>> extreme case, it doesn't matter if there are any differences in
>> reproduction rates. There are more bacteria than humans, and they
>> reproduce far more efficiently, but we don't compete for the same
>> niche.
>
> I understand the example. You mean that standard deviation of
> intelligence (or whatever) for Homo Sapiens can't shrink over time?
> As far as I've seen for US, it seems to be shrinking. I mean that
> average SAT score went down at the same time as average IQ went up,
> so lets assume that is the main tendency everywhere besides Hong
> Kong.

But IQ is supposed to be much more affected by genetics than SAT
(besides, the decrease in the average would NOT imply shrinking of the
standard deviation). The drop in SAT is most likely attributed to the
failure of the education system. There has been many recent claims
that the human brain evolution is ongoing, with positively selected
mutations arising very recently, one such at ~6000 years ago, that now
reached appreciable frequency. There have been an increase in brain
size too, compared to what it was even only 500 years ago. I recall
that was studied in Northern Caucasians.

> So what now? If you break any barriers between Hong Kong and the
> rest of the world their genes get diluted and thats it. Whatever
> factor was promoting this adaptation was local, not general, and
> local factors will not promote "large scale" adaptation.

I don't think it's just Hong Kong. The barriers are not economic or
geographic. The fluffy Dutchmen for example are documented to be
getting bigger AND smarter, which is sort of scary in a solid
teddybear kinda way.

Dnia 2006-10-26 DZ napisa³(a):
> Andrzej Rosa wrote:
>> I understand the example. You mean that standard deviation of
>> intelligence (or whatever) for Homo Sapiens can't shrink over time?
>> As far as I've seen for US, it seems to be shrinking. I mean that
>> average SAT score went down at the same time as average IQ went up,
>> so lets assume that is the main tendency everywhere besides Hong
>> Kong.
>
> But IQ is supposed to be much more affected by genetics than SAT
> (besides, the decrease in the average would NOT imply shrinking of the
> standard deviation).

Sure. I wrote "it seems" after all.

> The drop in SAT is most likely attributed to the
> failure of the education system.

Or something else. Sure.

> There has been many recent claims
> that the human brain evolution is ongoing, with positively selected
> mutations arising very recently, one such at ~6000 years ago, that now
> reached appreciable frequency. There have been an increase in brain
> size too, compared to what it was even only 500 years ago. I recall
> that was studied in Northern Caucasians.

I do not understand it. There is much less selection going on, part of
what actually happens is negative selection, yet somehow we as a specie
go through a surge of brainbuilding.

Strange.

>> So what now? If you break any barriers between Hong Kong and the
>> rest of the world their genes get diluted and thats it. Whatever
>> factor was promoting this adaptation was local, not general, and
>> local factors will not promote "large scale" adaptation.
>
> I don't think it's just Hong Kong. The barriers are not economic or
> geographic. The fluffy Dutchmen for example are documented to be
> getting bigger AND smarter, which is sort of scary in a solid
> teddybear kinda way.

We all are getting bigger and smarter, and for a good while too.
However it doesn't mean that genetic potential for intelligence is also
improving.

--
Andrzej Rosa 1127R

Andrzej Rosa wrote:
> DZ:
>> There has been many recent claims that the human brain evolution is
>> ongoing, with positively selected mutations arising very recently,
>> one such at ~6000 years ago, that now reached appreciable
>> frequency. There have been an increase in brain size too, compared
>> to what it was even only 500 years ago. I recall that was studied
>> in Northern Caucasians.
>
> I do not understand it. There is much less selection going on, part
> of what actually happens is negative selection, yet somehow we as a
> specie go through a surge of brainbuilding. Strange.

"Part of" is the key. The negative selection for intelligence it is
pretty much a speculation. Sure, it's fun to talk about somebody who
can't even properly operate a foodstamp. The fact is, intelligence of
the mental retardation category, when a person is barely able to
function in a modern society is more than sufficient to survive in a
primitive society (where IQ of 60 could be the norm). The way
selection operates in this case is like follows. It is reasonable to
assume a large number of genes affecting intelligence. The
recombination constantly shuffles these alleles thus producing a
certain proportion of unlucky individuals with the intelligence low
enough to be at disadvantage (in the sense of reproductive potential).
These are truncated. All you need for the population mean of IQ to
increase is to move this threshold for the modern society slightly up
compared to living in a jungle. The math of this process was first
studied by Haldane (http://en.wikipedia.org/wiki/J._B._S._Haldane) in
the context of advantages of sexual reproduction with recombination.

>>> So what now? If you break any barriers between Hong Kong and the
>>> rest of the world their genes get diluted and thats it. Whatever
>>> factor was promoting this adaptation was local, not general, and
>>> local factors will not promote "large scale" adaptation.
>>
>> I don't think it's just Hong Kong. The barriers are not economic or
>> geographic. The fluffy Dutchmen for example are documented to be
>> getting bigger AND smarter, which is sort of scary in a solid
>> teddybear kinda way.
>
> We all are getting bigger and smarter, and for a good while too.
> However it doesn't mean that genetic potential for intelligence is
> also improving.

Intelligence is highly heritable (take the intelligence of twins) -
the controversy is not really about that. It's about issues when
people are trying to compare IQs between groups, for example the
coefficient is not invariant with respect to the properties of the
populations. There is also the issue of twins being similar, while
offspring and parents might be not - a trait like that can still be
highly heritable.

Dnia 2006-10-26 DZ napisa³(a):
> Andrzej Rosa wrote:
>> DZ:
>>> There has been many recent claims that the human brain evolution is
>>> ongoing, with positively selected mutations arising very recently,
>>> one such at ~6000 years ago, that now reached appreciable
>>> frequency. There have been an increase in brain size too, compared
>>> to what it was even only 500 years ago. I recall that was studied
>>> in Northern Caucasians.
>>
>> I do not understand it. There is much less selection going on, part
>> of what actually happens is negative selection, yet somehow we as a
>> specie go through a surge of brainbuilding. Strange.
>
> "Part of" is the key. The negative selection for intelligence it is
> pretty much a speculation.

If we are talking about above average intelligence, those people surely
do populate worse than right in the middle crowd.

> Sure, it's fun to talk about somebody who
> can't even properly operate a foodstamp. The fact is, intelligence of
> the mental retardation category, when a person is barely able to
> function in a modern society is more than sufficient to survive in a
> primitive society

I think that you underplay the role of intelligence in *survival*. One
mistake and you are dead. Also, which mother will be able to take
better care for their children.

And last but *definitely* not least point. Who will win the "hearts"
of maidens more often? Well to do brave warrior and successful hunter
or a village idiot?

I do not buy this "an idiot can do well in a jungle but will be lost in
a concrete jungle". An idiot who is not able to hunt buffaloes can
still roam through a trashcan.

> (where IQ of 60 could be the norm).

I'm not impressed. I've seen estimations that the generation of my
grandfather was around 80, and he was Caucasian grown in XX century
Europe.

> The way
> selection operates in this case is like follows. It is reasonable to
> assume a large number of genes affecting intelligence. The
> recombination constantly shuffles these alleles thus producing a
> certain proportion of unlucky individuals with the intelligence low
> enough to be at disadvantage (in the sense of reproductive potential).
> These are truncated.

Sure.

> All you need for the population mean of IQ to
> increase is to move this threshold for the modern society slightly up
> compared to living in a jungle.

No. Not the number. The percentile (which I also doubt happens).

> The math of this process was first
> studied by Haldane (http://en.wikipedia.org/wiki/J._B._S._Haldane) in
> the context of advantages of sexual reproduction with recombination.

I do not know the math you are referring to, so maybe I'm wrong, but
from my point of view it looks like this "weeding off excesses" process
at work.

Low IQ cutoff and high IQ low population rate seems to indicate lower
standard deviation.

Remember that up to now high IQ yielded more healthy children, low IQ
was similarly at disadvantage so something changed, and for worse.

>>> I don't think it's just Hong Kong. The barriers are not economic or
>>> geographic. The fluffy Dutchmen for example are documented to be
>>> getting bigger AND smarter, which is sort of scary in a solid
>>> teddybear kinda way.
>>
>> We all are getting bigger and smarter, and for a good while too.
>> However it doesn't mean that genetic potential for intelligence is
>> also improving.
>
> Intelligence is highly heritable (take the intelligence of twins) -

Knock one on the head and compare their IQ.

BTW - prenatal environment seems to have high impact. It's much better
when your mother is not starving while pregnant.

> the controversy is not really about that. It's about issues when
> people are trying to compare IQs between groups, for example the
> coefficient is not invariant with respect to the properties of the
> populations.

You mean that mono-zygotic twins can be more or less similar, depending
where they grew up? So what?

> There is also the issue of twins being similar, while
> offspring and parents might be not - a trait like that can still be
> highly heritable.

So it may, but what's your point?

--
Andrzej Rosa 1127R

Andrzej Rosa:
> DZ:
>> Sure, it's fun to talk about somebody who can't even properly
>> operate a foodstamp. The fact is, intelligence of the mental
>> retardation category, when a person is barely able to function in a
>> modern society is more than sufficient to survive in a primitive
>> society
>
> I think that you underplay the role of intelligence in *survival*.
> One mistake and you are dead. Also, which mother will be able to
> take better care for their children. And last but *definitely* not
> least point. Who will win the "hearts" of maidens more often? Well
> to do brave warrior and successful hunter or a village idiot? I do
> not buy this "an idiot can do well in a jungle but will be lost in a
> concrete jungle". An idiot who is not able to hunt buffaloes can
> still roam through a trashcan.

There are regions on this planet with average IQs of about 60. Such
average would be too low for a Western city to sustain. Also, compare
the fitness potential (expected number of offspring who will
themselves survive and reproduce) of a buffalo hunter vs. that of a
trashcan hunter.

>> (where IQ of 60 could be the norm).
>
> I'm not impressed. I've seen estimations that the generation of my
> grandfather was around 80, and he was Caucasian grown in XX century
> Europe.

These tests are routinely re-standardized to have the average of 100
and std of 15, so a good comparison with the past is difficult to
make, in terms of the score.

>> The way selection operates in this case is like follows. It is
>> reasonable to assume a large number of genes affecting
>> intelligence. The recombination constantly shuffles these alleles
>> thus producing a certain proportion of unlucky individuals with the
>> intelligence low enough to be at disadvantage (in the sense of
>> reproductive potential). These are truncated.
>
> Sure.
>
>> All you need for the population mean of IQ to increase is to move
>> this threshold for the modern society slightly up compared to
>> living in a jungle.
>
> No. Not the number. The percentile (which I also doubt happens).

Sorry, but it's just a number. One can't operate a given device such
as an aircraft, if he's below some threshold, irrespective of the
population quantiles. Here is an analogous problem: It is known that
the longevity of offsping is decreased with father's age. The reason
is that the process of sperm formation involves constant cell
divisions, accumulating slightly deletrious mutations with age (that
is not the case in females). In fact, a male at puberty has already
accumulated several divisions. There is no indication that the
fertility of ofspring is decreased with father's age though. Why the
human population life span is not decreasing in generations?

>> Intelligence is highly heritable (take the intelligence of twins) -
>
> Knock one on the head and compare their IQ. BTW - prenatal
> environment seems to have high impact. It's much better when your
> mother is not starving while pregnant.

It's not as important as you think.

>> the controversy is not really about that. It's about issues when
>> people are trying to compare

.... the genetic component of

>> IQs between groups, for example the
>> coefficient

.... measuring heritability

>> is not invariant with respect to the properties of the
>> populations.
>
> You mean that mono-zygotic twins can be more or less similar, depending
> where they grew up? So what?

No, that's not what I mean. The coefficients like heritability are not
invariant with respect to the population variance. The point is that
there are difficulties with interpretation of those studies that
produced a lot of outrage in the press. I should add that I'm not a
big fan of studies on IQ. But some people fell victims of that debate.

>> There is also the issue of twins being similar, while
>> offspring and parents might be not - a trait like that can still be
>> highly heritable.
>
> So it may, but what's your point?

Arguments about statistical issues of estimating and comparing
heritability had one bad outcome - many people interpreted these
discussions as implying that heritability of intelligence might be
low. This is not the case.

Dnia 2006-10-27 DZ napisa³(a):
> Andrzej Rosa:
>> DZ:
>> I think that you underplay the role of intelligence in *survival*.
>> One mistake and you are dead. Also, which mother will be able to
>> take better care for their children. And last but *definitely* not
>> least point. Who will win the "hearts" of maidens more often? Well
>> to do brave warrior and successful hunter or a village idiot? I do
>> not buy this "an idiot can do well in a jungle but will be lost in a
>> concrete jungle". An idiot who is not able to hunt buffaloes can
>> still roam through a trashcan.
>
> There are regions on this planet with average IQs of about 60.

Give them food and they will get smarter. I mean, they can reach this
80 which our grandfathers had and managed in the cities.

> Such
> average would be too low for a Western city to sustain. Also, compare
> the fitness potential (expected number of offspring who will
> themselves survive and reproduce) of a buffalo hunter vs. that of a
> trashcan hunter.

Let's talk about equivalent of trashcan hunters in buffalo hunter
society. They wouldn't survive till puberty.

>>> (where IQ of 60 could be the norm).
>>
>> I'm not impressed. I've seen estimations that the generation of my
>> grandfather was around 80, and he was Caucasian grown in XX century
>> Europe.
>
> These tests are routinely re-standardized to have the average of 100
> and std of 15, so a good comparison with the past is difficult to
> make, in terms of the score.

That's why it was called estimation.

>>> All you need for the population mean of IQ to increase is to move
>>> this threshold for the modern society slightly up compared to
>>> living in a jungle.
>>
>> No. Not the number. The percentile (which I also doubt happens).
>
> Sorry, but it's just a number. One can't operate a given device such
> as an aircraft, if he's below some threshold, irrespective of the
> population quantiles.

But it says nothing about genetic potential for intelligence.
Primitive societies were less intelligent then we are, but they had at
least "proper" selection.

> Here is an analogous problem: It is known that
> the longevity of offsping is decreased with father's age. The reason
> is that the process of sperm formation involves constant cell
> divisions, accumulating slightly deletrious mutations with age (that
> is not the case in females). In fact, a male at puberty has already
> accumulated several divisions. There is no indication that the
> fertility of ofspring is decreased with father's age though. Why the
> human population life span is not decreasing in generations?

Having children early is good strategy for fast population. Short
living parents do not help raising their offspring.

Probably something else.

>>> Intelligence is highly heritable (take the intelligence of twins) -
>>
>> Knock one on the head and compare their IQ. BTW - prenatal
>> environment seems to have high impact. It's much better when your
>> mother is not starving while pregnant.
>
> It's not as important as you think.

Based on uniform American society where most everyone has enough food.

Are there some good data on poor societies (where substantial number of
people suffer from undernourishment)?

>>> the controversy is not really about that. It's about issues when
>>> people are trying to compare
>
> ... the genetic component of
>
>>> IQs between groups, for example the
>>> coefficient
>
> ... measuring heritability
>
>>> is not invariant with respect to the properties of the
>>> populations.
>>
>> You mean that mono-zygotic twins can be more or less similar, depending
>> where they grew up? So what?
>
> No, that's not what I mean. The coefficients like heritability are not
> invariant with respect to the population variance.

Damn. Either it boils down to what I wrote, or you lost me here.

> The point is that
> there are difficulties with interpretation of those studies that
> produced a lot of outrage in the press. I should add that I'm not a
> big fan of studies on IQ. But some people fell victims of that debate.

So you say it's difficult to interpret, but we better not know it than
know it.

Not my opinion.

>>> There is also the issue of twins being similar, while
>>> offspring and parents might be not - a trait like that can still be
>>> highly heritable.
>>
>> So it may, but what's your point?
>
> Arguments about statistical issues of estimating and comparing
> heritability had one bad outcome - many people interpreted these
> discussions as implying that heritability of intelligence might be
> low. This is not the case.

Of course it is not the case. It's very high in uniform environments.

Beside the fact that even raised apart twins share prenatal
environment, so they "show" a bit more heritability, but that's minor.

--
Andrzej Rosa 1127R

Andrzej Rosa <bakters@yahoo.com> wrote:
> DZ :
>> All you need for the population mean of IQ to increase is to move
>> this threshold for the modern society slightly up compared to
>> living in a jungle. One can't operate a given device such as an
>> aircraft, if he's below some threshold, irrespective of the
>> population quantiles.
>
> But it says nothing about genetic potential for intelligence.
> Primitive societies were less intelligent then we are, but they had
> at least "proper" selection.

It's going nowhere. Let me try one more time - I have to go to my
10 to 11 pm workout. Threshold selection will increase the average
intelligence of a population if that threshold - however low it may
be - is moved up (in the absolute value; forget the quantiles). Plz
marinade on this concept while I'm away :-)

Dnia 2006-10-27 DZ napisa³(a):
> Andrzej Rosa <bakters@yahoo.com> wrote:
>> DZ :
>>> All you need for the population mean of IQ to increase is to move
>>> this threshold for the modern society slightly up compared to
>>> living in a jungle. One can't operate a given device such as an
>>> aircraft, if he's below some threshold, irrespective of the
>>> population quantiles.
>>
>> But it says nothing about genetic potential for intelligence.
>> Primitive societies were less intelligent then we are, but they had
>> at least "proper" selection.
>
> It's going nowhere. Let me try one more time - I have to go to my
> 10 to 11 pm workout. Threshold selection will increase the average
> intelligence of a population if that threshold - however low it may
> be - is moved up (in the absolute value; forget the quantiles).

Yeah. If I forget about quantiles, it works. But I won't forget. ;-)

> Plz
> marinade on this concept while I'm away :-)

Sure. I'm done. What now? ;-)

--
Andrzej Rosa 1127R

Andrzej Rosa:
> DZ:
>>>> the coefficient measuring heritability is not invariant with
>>>> respect to the properties of the populations.
>>>
>>> You mean that mono-zygotic twins can be more or less similar, depending
>>> where they grew up? So what?
>>
>> No, that's not what I mean. The coefficients like heritability are not
>> invariant with respect to the population variance.
>
> Damn. Either it boils down to what I wrote, or you lost me here.

You seem to be thinking that "heritability is heritability" as long as
the environment is uniform. The bounds on heritability also depend on
the amount and the type of genetic variation in a given
population. So, even in the uniform environments, the heritability is
population-specific.

A more familiar to you might be the concept of correlation, which is
similar in that it's also a ratio of measures of variability. Suppose
you measure the correlation between 2 variables (X,Y) in the
population 1, and it is R1. In the population 2, the correlation is
R2.

Suppose R2 > R1. Are X and Y more correlated in population 2? It's a
trick question :-)

Facts like that (or some would say, nitpicks) had been exploited by
some very shrewd people to discredit much of research on genetics of
intelligence. The result was that all of it looks kinda bad to a
bystander.

>> The point is that there are difficulties with interpretation of
>> those studies that produced a lot of outrage in the press. I should
>> add that I'm not a big fan of studies on IQ. But some people fell
>> victims of that debate.
>
> So you say it's difficult to interpret, but we better not know it
> than know it.

That's not what I'm saying. Lots of that research had been done by
people who know psychometrics, but not population genetics. So they
were rightfully called on their BS. There is also a considerable
amount of politically influenced opinions. So, what I'm saying is that
a lot of that research is not very good.

Dnia 2006-10-27 DZ napisa³(a):
> Andrzej Rosa:
>> DZ:
>>>>> the coefficient measuring heritability is not invariant with
>>>>> respect to the properties of the populations.
>>>>
>>>> You mean that mono-zygotic twins can be more or less similar, depending
>>>> where they grew up? So what?
>>>
>>> No, that's not what I mean. The coefficients like heritability are not
>>> invariant with respect to the population variance.
>>
>> Damn. Either it boils down to what I wrote, or you lost me here.
>
> You seem to be thinking that "heritability is heritability" as long as
> the environment is uniform.

Yeah. That was my line of thought.

> The bounds on heritability also depend on
> the amount and the type of genetic variation in a given
> population. So, even in the uniform environments, the heritability is
> population-specific.

I can see it. Inbreed families have higher heritability coefficient
than normal families.

Kind of obvious. Smaller gene pool so higher chance to hit the same
set of genes.

> A more familiar to you might be the concept of correlation, which is
> similar in that it's also a ratio of measures of variability. Suppose
> you measure the correlation between 2 variables (X,Y) in the
> population 1, and it is R1. In the population 2, the correlation is
> R2.
>
> Suppose R2 > R1. Are X and Y more correlated in population 2? It's a
> trick question :-)

I see. It depends on population too. All right.

> Facts like that (or some would say, nitpicks) had been exploited by
> some very shrewd people to discredit much of research on genetics of
> intelligence. The result was that all of it looks kinda bad to a
> bystander.

Not to me. It looks solid enough. People react badly to learn that
not everyone is created equal, but it's to be expected.

>>> The point is that there are difficulties with interpretation of
>>> those studies that produced a lot of outrage in the press. I should
>>> add that I'm not a big fan of studies on IQ. But some people fell
>>> victims of that debate.
>>
>> So you say it's difficult to interpret, but we better not know it
>> than know it.
>
> That's not what I'm saying. Lots of that research had been done by
> people who know psychometrics, but not population genetics. So they
> were rightfully called on their BS. There is also a considerable
> amount of politically influenced opinions. So, what I'm saying is that
> a lot of that research is not very good.

Most of what we were discussing was based on quite obvious info from
demographics and so on. But I can see how bad research can be twisted
further to make some hit in the media.

To tell the truth, my background on this topic is quite old news. I'm
not in touch with what's going on recently, so maybe I'm missing
something.

However, I still can't see how present demographical dynamics will make
us all smarter in the future.

--
Andrzej Rosa 1127R

Andrzej Rosa:
> DZ:
>> Suppose you measure the correlation between 2 variables (X,Y) in
>> the population 1, and it is R1. In the population 2, the
>> correlation is R2. Suppose R2 > R1. Are X and Y more correlated in
>> population 2?
>
> I see. It depends on population too. All right.

If the second population differs from the first by (a+b*X, c+d*Y)
where lowercase are some constants, then R1=R2 only for some specific
types of distributions, like when (X,Y) is bivariate normal. So, the
correlation may not be able to reach the extreme value of -1 or 1 in a
population even if both X and Y co-vary perfectly together. A lower
value may simply be a reflection of some aspect of the underlying
distribution (like the mean value of X) being different in one of the
populations.

"DZ" <1553@2577127584.1206928762.3934.20258.15735> wrote in message
news:7913@244729345.1318421292.9579.19468.7066...
> Andrzej Rosa:
>> DZ:
>>> Suppose you measure the correlation between 2 variables (X,Y) in
>>> the population 1, and it is R1. In the population 2, the
>>> correlation is R2. Suppose R2 > R1. Are X and Y more correlated in
>>> population 2?
>>
>> I see. It depends on population too. All right.
>
> If the second population differs from the first by (a+b*X, c+d*Y)
> where lowercase are some constants, then R1=R2 only for some specific
> types of distributions, like when (X,Y) is bivariate normal. So, the
> correlation may not be able to reach the extreme value of -1 or 1 in a
> population even if both X and Y co-vary perfectly together. A lower
> value may simply be a reflection of some aspect of the underlying
> distribution (like the mean value of X) being different in one of the
> populations.

The mind boggles.....