Tilted Forum Project Discussion Community - View Single Post

jorgelito · 12-12-2004, 02:04 AM

Knifemissile, I think you might have misunderstood ConsoleMaster.

I get your point about the Japanese sample etc. By itself sure, sounds about right. There's more to it also. I think anthropologists or sociologists can explain better.

It's like, Japan is a relatively homogenous population with relatively low genetic variation ( I guess, "purer", so to speak) so the 1000 statistic is fairly viable.

BUT, I think this is ConsoleMaster's point: You can't use Japan's statistic to describe all of Asia or whatever or to represent China, India. Something like that.

I'm going to go out on a limb here but maybe we could use the US election as an example of statistical analysis. FOr example: I don't think we could use one state's stats to describe all of the US. If 1000 in California tend to lean Kerry, would that mean all of the US leans Kerry?

Or maybe this example:
If we could say, if 1000 men in Canada were shorter than Swedes, could we say that all North American men are shorter than Swedes?

Interesting stat problem...

12-12-2004, 02:04 AM	#72 (permalink)
jorgelito All important elusive independent swing voter... Location: People's Republic of KKKalifornia	Knifemissile, I think you might have misunderstood ConsoleMaster. I get your point about the Japanese sample etc. By itself sure, sounds about right. There's more to it also. I think anthropologists or sociologists can explain better. It's like, Japan is a relatively homogenous population with relatively low genetic variation ( I guess, "purer", so to speak) so the 1000 statistic is fairly viable. BUT, I think this is ConsoleMaster's point: You can't use Japan's statistic to describe all of Asia or whatever or to represent China, India. Something like that. I'm going to go out on a limb here but maybe we could use the US election as an example of statistical analysis. FOr example: I don't think we could use one state's stats to describe all of the US. If 1000 in California tend to lean Kerry, would that mean all of the US leans Kerry? Or maybe this example: If we could say, if 1000 men in Canada were shorter than Swedes, could we say that all North American men are shorter than Swedes? Interesting stat problem...