CHI-WEB Archives

ACM SIGCHI WWW Human Factors (Open Discussion)


Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Juan Lanus <[log in to unmask]>
Reply To:
Juan Lanus <[log in to unmask]>
Mon, 16 May 2005 08:51:48 -0300
text/plain (104 lines)
THe form at
is very simple, there is not much headroom for improvement. But anyway ...

"Work" might lead to confussion, somebody could answer "yes".
Also, IŽd put together "Gender", "Single or maried" and "Number of children".
Also, "Single or married" seem too few options to me. It lacks
"Widow", "Divorced".
Also, "Work" and "hobbies" go together.
The grouping of related questions is to make the user put his mind in
one subject at a time instead of moving like a hummingbird. Obviously
this is so here because this is a test form, but it' an example of
what to look for.
In this section some errors are reported under the field while others
are reported over the field. As this form has borders the information
conveyed is not ambiguous, but anyway this requires a mental
adaptation of the user. Any lack of coherency betrays a contract
between the form and the user ans has a cost. The user might adapt to
these small chances without even noticing it, but in all it's teasing
and consumes some of the "patience units" the user has alloted to this
task. When there are no more PUs left the user leaves. The initial
n.munber of PUs depends on the user's interest for what the site is

Section 2:
Should this be a real-life form, the same could be said as for section
1 about mixed data.
Also, mi preference, I'd reword the massage "A non-digit was found in
the field for number of cars" to "Number of cars must be a number". It
sounds too IT to me bacause of "non-digit", "found" and "field".
"Name of high school", what happens if the user went to more than one?
There is room for only one, should the ending one be informed? Or all?
Same as above about erratic error message positioning.

In the third part, after "Thank you for completing the test.", hhe
question "Thank you for completing the test." deserves some
Blind users might enter tne name of the access program i.e. it's
brand, or a generic type like "screen reader".
Non-blind users might get lost at this point.

About the form's design, it's bvious that it has no design being as it
is a test. It's apparent that the intend id to leave is as simple as
possible, isn't it?
This led to those double borders in the tables that contain the data. 
This makes the form look more complex than it is maybe consuming some
PUs at the forst glance.
Try looking "fuzyly" near the field containing "Berkeley". Your ming
will have to process all those horizontal lines that sem to be more
that the actually are.
Should the designer have used single lines for the cell boundaries
(cellspacing="0" in HTML) then the form would look simpler at first
Should the lines be gray instead of the default black color the the
form would be much more readable. The lines do their duty but somehow
disappear from the conscious level by making the prompts glow in

Also I'd try to compress the general design in it's height to make the
form look shorter, faster to fill. The visualisation of a long form
may make the user drop some PUs.
In this form there is some unneeded vertical spacing. The presence of
the horizontal lines makes unnecesary all that margin at the top AND
at the bottom of each field. And specially, the space between the
field in error and the corresponding error message is undesirable
because it divorces one from the other.

Well Michele, this was a sample "expert rewiew" on a form surely not
intended to be reviewed.

Please heep in touch!
Juan lanus

On 5/15/05, michelle <[log in to unmask]> wrote:
> This is a great discussion, one that I'm highly enjoying!
> One question - how would you redesign this form (or a different one of
> your choice) to make it more usable, thereby decreasing the likelihood
> that someone would make multiple mistakes:
>  (i.e., the only
> concrete example in this thread)
> Let's say the purpose of this form is to provide a user researcher
> with information about a potential participant.  To make the problem
> simpler, discard the fields that expect numerical values.
> I'm asking because it's hard to reach consensus on a problem without
> having a specific example to evaluate...
> Michelle

    Tip of the Day: Suspend your subscription if using auto replies
     CHI-WEB: POSTINGS: mailto:[log in to unmask]
              MODERATORS: mailto:[log in to unmask]