- From: Simon White <simon.white@jkd.co.uk>
- Date: Fri, 11 Jan 2002 16:40:26 -0000
- To: "Steven McCaffrey" <SMCCAFFR@MAIL.NYSED.GOV>, <Ian.SHARPE@cambridge.sema.slb.com>, <w3c-wai-ig@w3.org>
Agreed, but only when stats are applied to something specific. They still mean nothing when we say that an accessible site will give a 2% rise in hits or a 25% rise in revenue. That cannot be argued with... -----Original Message----- From: Steven McCaffrey [mailto:SMCCAFFR@MAIL.NYSED.GOV] Sent: Friday, January 11, 2002 16:27 To: Ian.SHARPE@cambridge.sema.slb.com; Simon White; w3c-wai-ig@w3.org Subject: RE: statistics - for differences between accessible and non-acce ssible sites? I think there are some valid points on this thread about the potential misuse or misapplication of certain specific statistics, but I think we should clarify where the problem lies. Borrowing the form, but not the content, of a well known phrase used by <unnamed>, "Statistics don't lie, people do." The point is a bit more subtle in that the "lie" is usually an implied inference from the statistics rather than any given statistic. Statistics is the best tool humans have developed precisely to avoid deceiving ourselves which is why it is the bedrock of science. Not only does it allow precise quantification of the subject matter itself, but it also gives us a way to quantify the degree to which our statements may be in error. It is, I agree, troubling that some continue to misuse this very helpful tool of thought and communication. Steve _____________________________________________________________________ VirusChecked by the Incepta Group plc _____________________________________________________________________
Received on Friday, 11 January 2002 11:40:44 UTC