Question structures in CHOICE
See the question info page in the
information architecture pages.
It is very important for people to mean the same when answering,
or else comparison across people is compromised.
Clarity and simplicity are key virtues here.
The challenge is how to help people attain them.
Because most questions will be written by users,
there is always a strong possibility that the question doesn't make sense to others.
Therefore, we have, for all questions, not only the ability to refuse to answer,
but also a place to register confusion, lack of understanding, inability to answer, etc.
List of question structure types
- QSopt: option list questions
- This is a very general structure to hold a question with multiple choice answers.
There may be just one answer possible, or multiple answers.
- QSfact: factual questions
- Any factual question to which the answer is Yes or No or Don't know.
- QSlik: Likert scale questions
- The classic 5-point Likert scale about agreeing with a statement;
or other question structured in the same way.
- QSpref: alternative preference questions
- This is for A/B choices, again with a 5- or 7-point scale.
- QSrange: position / range on continuum questions
- This is not for enquiry location, which is handled separately, but for any value
effectively on a continuum, including personal height and weight.
- QSlevel: level questions
- The only difference between this an QSopt is that these options have an inherent order,
where more is better in a scalar manner.
This may be more likely to be in a predefined structure than for QSopt, so there is an explicit
possibility of having an externally-defined, reusable scheme for the answer items.
Particularly useful for ability / knowledge / skill / competence ratings.
- NOT USED: QScheck: pure binary questions
- This is the barest possible kind of question, with answers "yes" or "not yes".
But as this is purely binary, there is a clear potential for polarizing questions.
It would be suitable for gathering external input from e.g. Facebook "like" data,
and perhaps that is exactly the problem: FB tends to polarize.
Are you in or out? One of us or one of them?
"The one who is not with us is against us."
For these reasons, we will not have this kind of question, and it is listed here
to make that clear and give the reasons.
Question attributes
See the Question information structure