Several years ago. Someone should have intervened and retired him that time. If you are a scientist and you start voicing out conclusions for controversy, you are just a sad old man begging for attention.

During a lecture tour in 2000, he suggested there might be links between skin color and sexual prowess, and between a person's weight and their level of ambition.  
 
For the love of God, sexual prowess? The guy has COMPLETELY lost his marbles, admit it.  
 
Person's weight vs ambition-- Oprah?

Oh, please. 
 
If you are talking about averages: 
 
The average male American mens height is 177 cm, which is 69.7 inches, which is approximately 5 foot - 10 inches tall. (for white males*). 
 
Most pedophiles are white males. 
 
Now if this comment is deleted, does this mean its not true or does this mean you cant talk about it?

See, this is exactly the big problem you'd have if you treat the literature/language graduates as "intellectuals". They just love to talk and spin lies and then print them into history books. 
 
Whoever is the we - first world, does that include the old civilizations that this "special" race destroyed. 
 
Dont even reach into history before year 1980. I believe the bombing of Baghdad just happened recently?

Our inner cities are practically matriarchal, since the men head off to jail and aren't exactly "Dad of the Year" material. They're squalid and deteriorating, not feminist paradises flourishing under the wise rule of elder Gramma Jones.  
 
Youre talking about single-parent household compared to a two-parent household. Thats just poverty.

Oh Ryan,  
If you notice, we are just having fun, considering the topic. Its like here we go again, male vs. female. 
I wouldnt argue with you using scientific terms because I dont use that in my daily life,  
I can put something up cut and pasted and borrowed from somewhere else, argue with the same arguments you've already heard from somewhere. Im just throwing ideas crappy it may sound. 
I am going back and forth to my system design and this blog, because Im bored. You need to lighten up with your "stupid-female namecalling".  
When was the last time you got laid?  
 
:P

human hermaphrodites are not capable of self-reproduction 
NOT yet. Were still evolving.  
 
Wouldn't hermaphroditism reduce species biodiversity? Then the hermaphrodites can impregnate not itself but another one. 
 
men average 5 IQ points above women Like I said, what if we design the test according to our liking, something they cant pass.  
 
:)

Anyway, what is the deal with this female-to-female communication nonsense?  
 
Typical male - anything thats female-related is nonsense. 
 
To follow darwin's logic.. if a female has lesser strength than male, how did she survive?  
Why didnt male species just turned into hermaphrodites? 
The female made up for the lack of male strength by being smarter. 
Do you think that those hermaphrodites are females who finally got rid of the males? 
 
:P

oh shut up. lol. My bad, mistypo. i mean west.

I was channel-surfing and came upon the audition of Beauty and the Geeks  
So there is this pretty girl in the spotlight, she was asked if 
right is your north, where is your west. 
 
She pointed upwards, and the auditioning group were shown 
straight face, like that deer in the headlight look, the correct answer is that she should point in front her. 
 
FUnny thing, I got what she meant. Is it a female to  
female thing? 
She was seeing the analogy in another plane, which is parallel 
to her body, not perpendicular. 
 
Also if you notice, women can jump from topic to another 
and still follow..each other, as if they didnt stop thinking  
about it 20 minutes ago.  
 
So bottomline, I think women are still smarter, in any plane. 
 
Its just that their brain process information faster they get  
lost between childrearing and female talk. 
 
What if tge test were designed to measure a male brain, not measure  
the female brain. 
 
Do you think if we design the test, the male population can pass them? 
 
:P

http://www.freep.com/apps/pbcs.dll/article?AID=/20070918/NEWS04/709180375

Women have cheaper insurance rates because they are safe drivers, not because they lack testosterone.  
Elderly drivers become bad because they cant see anymore and have slower reflexes. 
East Asians are bad drivers because most of them  
didnt have a car but a bike before.  
Im confused about this thread, I tried to read his pov and yours, its like a free for all postings.

Thank you, Razib. 
Is the remaining 1% Catholic? 
 
Although I am not religious in a strict sense, I see religion as one of the variables that mostly differentiates a culture from other cultures.

Thank you, J. 
 
They are Muslim over there, aren't they?

Is "Razib" an Indian name?

The Mood, Graybill and Boes classical textbook is not bad if you want to read about some theoretical, general aspects. Its level is intermediate.

This is just to acknowledge that I've read your reply. 
 
I am not sure that I'll comment further because our perspectives are too different. The issues that you are concerned with are of little interest to most statisticians of our times. There are more challenging problems to be solved and new techniques to be applied than to reflect on the properties of the mean. You, instead, seem to be more fascinated by the historical development of some statistical concepts. But after all, what's wrong with that?

A general observation concerning the formal structure of the post. 
 
In the introduction you correctly observe that it is not worth to include formulae in your account. I agree. In fact there are too many of them, starting from the Notation paragraph which seems to me completely useless for the economy of the discussion, just like all the equivalent ways to express the correlation coefficient in terms of the regression coefficient and other such stuff. In some cases, a simple formula can be meaningful even to the readers with little mathematical knowledge as it is, for instance, the equation of a linear model. As regards covariance, correlation, least squares estimators, instead, I doubt that such concepts can be understood from your notes.  
 
In my opinion you have to make a choice: either you write about this topic for the general public or you write about it as a statistician would do. The compromise that you?ve chosen seems to combine the disadvantages from both perspectives. Why not explain, for example, what kind of statistical measure the covariance is that is what it is supposed to measure, which is its domain and what interpretation is associated with the values it can take on. And then, starting from the observation that the covariance does not provide an indication of the strength of the association, you could have introduced the linear correlation coefficient (Pearson?s). 
 
Observations on specific points. 
 
Unless otherwise stated, these notes deal only with the linear regression and correlation of two variables. Questions of sampling and measurement error will not be covered. 
Then do you mean that you will not tackle the inferential aspects of regression? Some of your considerations, instead, implicitly assume that you?re working with a sample (see further on). 
 
Of course, we do not always literally want to estimate or predict the value of a variable. The value may already be known, in which case prediction would be unnecessary, or easily measured, in which case a mere estimate would be a poor substitute. The use of regression is more often in connection with hypotheses. We may want to formulate a hypothesis about the general relationship between two variables, or we may already have such a hypothesis and want to test it. Calculating a regression coefficient may suggest such a hypothesis, or put an existing one to the test.  
I see what you mean but it is badly stated. It would have been enough to mention the word ?model?. The purpose of regression is to model a causal relationship between an independent variable X and a dependent variable Y and therefore be able to estimate values of Y all along the line. 
 
For theoretical purposes it is generally simplest to use formulae with deviation values. 
I don't agree, actually it is quite a useless complication that is no longer found i
More....

Thanks, David, that's it. 
There are still no comments...but it is not a question of life or death.

Comment to the addition. 
I am not sure (I know those formulas so I did not pay much attention earlier) but I think the symbols were more readable in the previous version, at least with my browser. 
 
PS - Nice way to spend Sunday... and, if you are from the American continent, you have gotten up quite early too.

David and Peter, 
I'll be delighted to join the discussion. It will take me some time to formulate my thoughts so please be patient.  
 
I have to tell you from the beginning that I am rather straightforward in the way I express myself which sometimes causes a bad first impression. Also, English is my third language so forgive my syntax errors.

I have taken a quick look at your post and I would have many critical observations to make. If you're interested, I can explain my motivations. If you're not, I am not going to waste my time typing, 
 
PS - No one answered to the Parallels + Bootcamp question that one of your friends posted some time ago?