Absolutely, the sample has to be completely random (I used the word representative - perhaps random would have been better!)
The point I was making was that 31 out of 2000, 31 out of 2 million doesn't matter - if the 31 is random. David >-----Original Message----- >From: [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED]] On Behalf Of >Peter Rundle >Sent: Friday, 21 June 2002 5:00 PM >To: [EMAIL PROTECTED] >Subject: Re: [aus-soaring] RPL results > > > > >David Conway wrote: > >>Don't call the kettle black (if you are a pot)! >> >[lots of good rules about statistics snipped] > >David, what happens to your rules about statistical samples if >the sample is not random? After all in this straw poll the >voters selected themselves. > >No serious question, isn't the fact that the voting isn't a >random sample make those extrapolations invalid? > >rgds > >Pete > > > > > > > > >-- > * You are subscribed to the aus-soaring mailing list. > * To Unsubscribe: send email to [EMAIL PROTECTED] > * with "unsubscribe aus-soaring" in the body of the message > * or with "help" in the body of the message for more information. > -- * You are subscribed to the aus-soaring mailing list. * To Unsubscribe: send email to [EMAIL PROTECTED] * with "unsubscribe aus-soaring" in the body of the message * or with "help" in the body of the message for more information.