Content validity refers to how accurately a measurement tool covers all aspects of the specific construct. I believe it is necessary to do content validity tests before doing some statistical analysis.
As Mahfuz Judeh mentioned, content validity refers to the extent that a measure represents all facets of a particular construct. On the other hand, reliability simply represents (depending on how it is assessed) the repeatability of a measure. Of course, something that is measured with consistency does not mean that it is measuring what is desired. This is where the various types of validity measures come into play as validity is assessing that an assessment is measuring what it is supposed to be measuring.
using the classic example of a dart bored, hopefully I can explain this more clearly. On the dart bored you have the bullseye which can be though of as validity. On the other hand, reliability, as is merely pertains to consistency in measurement though what’s being measured may not be relevant can be thought as darts being grouped tightly together but not near the bullsey. Therefore you want a measure To be both reliable and valid in which there is a substantial grouping on the bullseye. Hopefully this explanation is not too fickle and provides some type of clarification on the necessity of assessing validity.
Let me know if you would like a better explanation and I will gladly provide one.
For content validity give questionnaires to subject experts.preferably 6 to 8 experts feed back on content and you can update.verify what ever criteria used for assessment is valid even in this experts opinion and review literature is at most important.
Besides the excellent answer of Logan briefly introduced early by Mahfuz I want to add that it depends very much on how you design the questionnaire. It is all derived from your creative work or it rest on other questionnaires and theories? It is derived by yourself or it was developed on the rationality of brainstorming with experts or focus group with key informants? For instance, John Rossiter work from marketing is against of content validity with some compelling arguments.
I agree with the need of be confident about what is it measuring before measure anything. Ravishankar post one way to do content validity, I am witness of satisfactory content validity analysis with only three experts but in any case the very big question is How do you choose your experts and how do you assess their expertise?
Rossiter's mos complete and simple reference is his latest book published by Springer:
Rossiter J. R., 2011. Measurement for the Social Sciences. The C-OAR-SE Method and Why It Must Replace Psychometrics.
the statistical analysis will not be accurate if the value to be statistically calculated is under valued or over valued. Thus, the need for content validity to ensure that all elements in the research are adequately covered.