Quick Logic Problem

It's like saying 1 in 5 people on the planet are Chinese, but wondering if that could be right because all the people you meet are American...
 
Yurt said:
but does math take a survey....
Click to expand...
No, if we have a survey on the answer to a long division problem, the chances are many would get the wrong answer, but would that change what the answer actually is?

The sampling of the survey is the "missing dollar" in his scenario.
 
Damocles said:
No, if we have a survey on the answer to a long division problem, the chances are many would get the wrong answer, but would that change what the answer actually is?
Click to expand...

nope

famfeud.jpg
 
KingCondanomation said:
Super, it's right there.
"Finally vos Savant started a survey, calling on women readers with exactly two children and at least one boy to tell her the sex of both children. With almost eighteen thousand responses, the results showed 35.9% (a little over 1 in 3) with two boys."

And remember this is 18,000 people, political surveys are fairly close to correct with under a 1000 people answering.

The survey is not really the point here anyway, the logic is sound, it was just used to convince those who were still sceptical.
Click to expand...

again, show me the actual survey questions and results, then we can discuss her survey.

The laws of probability do NOT support the results.
 
If we call 1,000,000 people in a given area in the US and asked how many of them were Chinese, the result would not support the actual statistic of how many people in the world were Chinese.

No survey, no matter how it was taken would change the math.

Just like if we had 5,000 people flipping coins with various results, it would not change the probability of my previous scenario one iota. It is not salient to the actual math.
 
Damocles said:
If we call 1,000,000 people in a given area in the US and asked how many of them were Chinese, the result would not support the actual statistic of how many people in the world were Chinese.

No survey, no matter how it was taken would change the math.

Just like if we had 5,000 people flipping coins with various results, it would not change the probability of my previous scenario one iota. It is not salient to the actual math.
Click to expand...

People vary in ethnicity by geography, but the ratio of boys to girls by geography largely does not.
I would find far more Chinese people in California than Tennessee, but I would still see roughly the same ratio of boys/girls in Cali that I see in Tennessee.
The people in the survey are the readers of her column, they LIKELY have a higher IQ average than the norm, but there is no bias reason whatsoever they would have children where more would be skewed to having a boy and a girl instead of 2 boys.
They do so because the probability was higher.
 
You must log in or register to reply here.
Back
Top