WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: System Usability Scale with Blind/VI users

for

From: Whitney Quesenbery
Date: Sep 12, 2015 2:26PM

Agree with Tim's clarification and what he said about usability and
accessibility. I fall back on the ISO definition of accessibility as
usability for people with the widest range of capabilities. And usability
as efficiency, effectiveness, and satisfaction in reaching a goal.

So it's not enough to have a site (page, group of pages, function...) that
is technically correct. It must support users in doing something - finding
info or taking an action.

The importance of satisfaction in this equation is that it can balance
absolute measures, especially for efficiency. If all you measured was time
on task, you would never know that even though a task might take a while to
complete (even longer for one user than another), it can still be
satisfying if the user had previously not been able to do it at all.

And it might be quick but if it doesn't get the job done, it not only fails
effectiveness, but is also likely to produce low satisfaction.

Just remember that there is a recency bias, and that satisfaction is mixed
up with expectations for ease. Example that I see all the time: a user is
learning how to use an interface for a task they think ought to be easy.
After a few tries they figure it out, and rapidly increase in efficiency
and effectiveness. Satisfaction is likely to be rates as high, if they
learn how to use the interface in a length of time they feel is appropriate
- and early struggles tend to be forgotten. So timing of satisfaction
questions make a difference.

Others reasons to measure perceived satisfaction (and related, trust) are:
- Increases in satisfaction can often accompany improvements in the
usability of the most common tasks
- Users are more likely to persist in working through problems on sites
which they perceive as being good




On Tue, Sep 8, 2015 at 11:52 PM Tim Harshbarger <
<EMAIL REMOVED> > wrote:

> SUS is definitely something that could be used as part of usability
> testing with people of any type of disability.
>
> However, the one thing I would urge that you keep in mind is that it
> measures the user's perception of the system's usability--which is not
> necessarily the same as the reality of how usable the system is. That
> isn't to say that there is no value in using SUS. It is just important to
> keep in mind what SUS really measures.
>
> I tend to think of accessibility and usability as a weird type of
> pyramid. At the bottom you have accessibility. If you don't have
> accessibility, you can't even really begin to address usability. From a
> task oriented perspective, until the tasks are accessible, you can't really
> even start to address issues such as accuracy or efficiency.
>
> The weird aspect to the pyramid is that it isn't built of bricks. There
> isn't a layer of accessibility bricks at the bottom with usability bricks
> placed on the layer above. There isn't a clear cut division between
> usability and accessibility.
>
> It is kind of like dawn. There is a point when it is clearly night. And
> there is a point when it is clearly day. But there is a transition between
> when it might be argued whether it is still day or night. I hope that
> makes some kind of sense.
>
> Hopefully something I wrote is useful food for thought.
>
>