W3C home > Mailing lists > Public > public-wai-evaltf@w3.org > August 2011

confidence (was Re: some initial questions from the previous thread)

From: Shadi Abou-Zahra <shadi@w3.org>
Date: Wed, 24 Aug 2011 09:08:06 +0200
Message-ID: <4E54A356.8030300@w3.org>
To: Vivienne CONWAY <v.conway@ecu.edu.au>
CC: Eval TF <public-wai-evaltf@w3.org>
Hi Vivienne,

On 24.8.2011 02:55, Vivienne CONWAY wrote:
> Confidence score: not sure how this is meant.

For example, Success Criterion 1.1.1 (Non-text content) has more level 
of subjectivity than 1.4.3 (Contrast). This means it is more likely that 
two independent evaluators of 1.4.3 come to the same result, so the 
confidence in an assertion made by a single evaluator is higher.

Same can be applied for level of difficulty rather than subjectivity. 
For example, SC 1.1.1 might be seen to require less (accessibility) 
knowledge to evaluate than for SC 2.4.1 (Bypass Blocks). Therefore it is 
more likely that two independent evaluators agree on a result.

The question is how much value does such a confidence attribute has 
versus the complexity it adds. For example, is a website that is 90% 
conformant with 60% confidence more accessible than a website that is 
60% conformant with 90% confidence? Is this result comparable at all?


Shadi Abou-Zahra - http://www.w3.org/People/shadi/
Activity Lead, W3C/WAI International Program Office
Evaluation and Repair Tools Working Group (ERT WG)
Research and Development Working Group (RDWG)
Received on Wednesday, 24 August 2011 07:08:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:40:18 UTC