- From: Ian Hickson <ian@hixie.ch>
- Date: Thu, 26 Feb 2009 23:14:28 +0000 (UTC)
- To: Philip TAYLOR <P.Taylor@Rhul.Ac.Uk>
- Cc: Steven Faulkner <faulkner.steve@gmail.com>, HTMLWG <public-html@w3.org>, W3C WAI Protocols & Formats <w3c-wai-pf@w3.org>
On Thu, 26 Feb 2009, Philip TAYLOR wrote: > Ian Hickson wrote: > > > > For example, if one were surveying ice cream, and one survey said that > > 30% of ice cream vendors had cones and 50% of ice cream vendors had > > lemon, and another survey said that 31% of vendors had cones, and 49% > > of vendors had lemon, then it is likely that if the second survey said > > that 40% of vendors had apples, that about 40% of the vendors surveyed > > in the first survey also had apples. > > Doesn't that require that both samples be taken from the same > homogeneous population ? The fact that first two points are so closely correlated means that the two samples probably are both proportional samples of the same overall population, which is why we can draw the conclusion from the third data point from only one sample. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Thursday, 26 February 2009 23:15:24 UTC