Missing values in questionnaire: What's next? I'm doing a quantitative research for my paper. Recently, I went to fieldwork for collecting data using questionnaire form. 
But, when I'm doing a data entry, I faced a missing value for a few questions. I do not want to remove the sample for analysis, but what alternative I have?
Should we accept the missing values as a "Neutral", "Don't know", "No idea" responses or remove the sample? 
 A: You have no data other than that the person "declined to answer". But your sampling plan should already have decided what to do in this case. It is a bit late to decide how you will handle data after you have gathered it. 
A research plan should be clearly stated and take account of such things. 
Talk to your advisor about how you should recover from this. 
There are many ways to design a study. One way would be to decide to reject all incomplete responses. That is pretty radical and probably rare, I think. But you need to include it in the study design. Otherwise all your results become suspect. 
A: *

*This article explains the issue and some options very well.

*One thing the article does not address is that whatever method you use, I would show the sentsivity to it. Probably look at deletion and infilling as two options and then see how much derived models differ.  Hopefully not much.  But even if they do, you are being honest.  And can recommend followup studies or different methods (e.g. ethnography or case studies versus surveys) to help check on troublesome missing variables.

*I also recommend to get a statistician involved for a consult.  Talk to your stats department.  It's not just that you might get a better answer but this is an "excuse" to work with an expert.
