- From: Ian Hickson <ian@hixie.ch>
- Date: Fri, 27 Feb 2009 00:57:30 +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: > > > > > > 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. > > We are getting dangerously off-topic, but I disagree. There may well be > a correlation between the prevalence of lemons and the prevalence of > cones, yet no such correlation between (the prevalence of) either lemons > and apples or between cones and apples. Yup, that's possible too. It's just less likely. It's also possible that a random sample of the total population happens to pick the exact wrong set of files. We could look at one sample and see that there are no files that use the <tfoot> element, when in fact every single other page in the population uses it. However, that is unlikely. Statistics is all about probabilities. Again, if you are not satisfied with the methodology of the studies that have been done to date -- and I'm certainly not suggesting they are perfect -- then I urge you to do your own studies with even fairer, more random, more representative samples. The more data the better. -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Friday, 27 February 2009 00:58:16 UTC