Processing questionnaires: what to do with "i dont know" answers?

I could place them as a seperate catagory and give a % OR i could file them under "missings" but then they wouldnt be shown with a %. What i would like to know is if i can CHOOSE between the two ways, depending on the needs...

Processing questionnaires: what to do with "i dont know" answers?
Both of the two first answers here are informative.





You can choose between the two methods - exclude from the overall total and indicate them as missing - or include them as 'don't know' and indicate the percentage responding in this category. However, you will need to be consistent in analysing the data and reporting the results.





The best thing to do is to include the "I don't know(s) as representative of your dataset. So include them in the total (for all questions) and work out the percentage of those in this category. As someone else here comments, it is better to think of this in the questionnaire design stage and to think about what a "I don't know" response really means; is it better for example to have a sliding scale from 'strongly agree' through to 'strongly disagree'. This together with piloting your questionnaire is one of the most aspects of questionnaire design.
Reply:Sure. It depends on what the question is. "I don't know" may be a legitimate response. I would generally place them in a separate category, but it depends on the nature of the survey. "I don't know" could mean "I don't care" or "I don't have enough choices to choose from" depending on the circumstance. they key thing here would be to use a footnote on the survey results stating what you did, so others who read it won't be tricked into thinking that you are playing with the numbers a bit.
Reply:I think you need to keep your results consistant so changing how you represent the "I don't know" answers to suit your needs would probably be defeating the object.





I would give them their own category because "I don't know" can say a lot about the question you've asked or the person who has answered.

medicine