Questionnaires.

[Seth A Maislin <smaislin -at- WORLD -dot- STD -dot- COM>]

Janet writes, "I have found that too often these forms offer no place
obvious or suitable to make any type of negative remarks.  I get asked
what I liked a lot.  I don't get asked what part of the documentation was
incorrect, useless, not there, etc."


A good questionnaire doesn't tell you what you want to here; it gives you
what you can USE.  Many times even a yes/no survey or a list of question
where the survey-taker gets to rate each category from 1 to 5 is a
useless survey.  So what if accessibility is rated at its lowest; does
this mean that the index of the manual needs work, that the langauge
must be "toned-down" for a nontechnical reader, or that the on-line
manual just takes too darned long to show up on the monitor?  The
question should be something that asks the survey-taker what is desired
or needed, not "what is lacking"; in other words, ask how the existing
documentation (or other product) can be IMPROVED, because that's useful.
And yes/no questions are sometimes too specific, since you would want the
survey-taker to have the freedom to answser something you mightn't have
dreamed of.  If you ask "How could this product be more accessible?,"
then you will get a useful answer.  And questions like, "Do you like
using this product?" are COMPLETELY USELESS -- what could you possibly
hope to do with that kind of information?

I have seen too many ridiculous surveys in my time.  At the point that
maybe 5% of all surveys are answered and returned (through the mail, the
statistic actually less than one percent), don't you think the questions
should be set up so as to give you useful answers?  What do YOU want to
know, and then ask.

Seth Maislin
<smaislin -at- world -dot- std -dot- com>