W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2002

RE: statistics - for differences between accessible and non-acce ssible sites?

From: Simon White <simon.white@jkd.co.uk>
Date: Fri, 11 Jan 2002 16:40:26 -0000
Message-ID: <D1EFBFDCD178C24DA607A306D6E3A7112DBD5D@URANUs>
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
"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.


VirusChecked by the Incepta Group plc
Received on Friday, 11 January 2002 11:40:44 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:36:06 UTC