P Floding wrote:
> Negative results:
> 
> 1. We would like to prove that there exists red cars.
> 2. We observe cars.
> 3a. We see red car. We have proven that at least one red car exists.
> or
> 3b. We never see a red car. We didn't prove a single thing. This is
> what's called a "negative result".

Pretty silly example. All you have to do is find one of anything to 
prove that X exists.

A better hypothesis is that there are so many red cars that you can see 
one if you sit at the Holland tunnel for even one hour, you will see 
one, and from that you can extrapolate that there could be as many as Y 
zillion red cars.

You could start with an analysis that 17 million cars are sold in the US 
every year, and they last about 20 years, so there are about 340 million 
cars on the road. If 2% of cars are red, and you sample ten thousand of 
them (sitting at the front of the Holland Tunnel for Z minutes) then
you can test the hypothesis that red cars make up A percent of the cars, 
and you should see one at the Holland Tunnel.

If you sat there measuring for a year of rush hours, and never saw a red 
car, then you could claim that the probability of there being any red 
cars is low. And you could back that up.

What you will not know that all the red cars are in California and 
Florida....



-- 
Pat
http://www.pfarrell.com/music/slimserver/slimsoftware.html

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/audiophiles

Reply via email to