#0 - March 10, 2010, 3:44 p.m.
They are fun to take don't get me wrong and if this is the point then, cool, mission accomplished!
If you are truly trying to get usable data from these then your data will be flawed.
For instance:
---------------------------------------------------------------------------------------------------------------------------------------------
Question: "How often do you visit the barbershop?"
Answer choice: Infrequently, as once I make a change, I like to stick with it.
Problem:
Someone may visit the barber shop infrequently but it may have nothing to do with the fact that once they make a change they stick with it. Perhaps they don't visit the barber shop for various other reasons.
---------------------------------------------------------------------------------------------------------------------------------------------
Question: "How often do you visit the barbershop?"
Answer choice: Fairly often, as variety is the spice of life
Problem:
Someone may visit often but assuming that it is because they feel that "variety is the spice of life" is a fundamental flaw in survey formation.
---------------------------------------------------------------------------------------------------------------------------------------------
Overall solution is providing more response choices that only include one response per choice.
A better format would be to have many questions about the topic and put responses on a Likert or Likert-type response format. (ex. Strongly Agree to Strongly Disagree).
Then you could hire someone like me to create your survey and then statistically analyze them =)
