There is this small arbitrary assumption bothering me a bit. We present to our respondents a questionnaire with standardized options of answers and think that they:

1. have their own opinion (on the researched subject) clear enough to be expressed

2. read/listen to the whole question

3. read/listen to all options in cafeteria

4. pick the option fitting most their own opinion (characterized above)

5. want to inform us honestly about their choice.

In other words, we assume that the respondents are rational and honest. This is basic, prior to later analysis when we use sophisticated statistics to argument that observed opinions may be generalized and comprise models etc. But these are at least five points/steps of uncertainty undermining further conclusions. Or am I just paranoid?

Similar questions and discussions