It depends upon your survey design. Are you using stratified random sampling or some other randomized design? Chapters 4 and 5 in Cochran, W.G.(1977), Sampling Techniques, 3rd ed., John Wiley & Sons could help. Or some other survey statistics book.
Do you have auxiliary data? You could check out a book such as Särndal, C.-E., Swensson, B. and Wretman, J. (1992), Model Assisted Survey Sampling, Springer-Verlag.
Other good books include Brewer, KRW (2002), Combined survey sampling inference: Weighing Basu's elephants, Arnold: London and Oxford University Press, and
Lohr, S.L.(2010), Sampling: Design and Analysis, 2nd ed., Brooks/Cole.
If you are using strictly model-based methodology, the paper at the attached link might help.
I hope your project is enjoyable.
Conference Paper Projected Variance for the Model-based Classical Ratio Estim...
Note that in every case, some idea of the variance for data of interest has to be estimated, 'guessed' (Cochran, p 77), or assumed first. Cochran notes four ways to do this, including a pilot study or previous similar data. You may have to do a sensitivity analysis (worst case - best case soft type of analysis) first. Other survey statistics books, such as Kish or Lohr, might help.
Note also that this needs to be done individually for every (important) variable of interest (item/question on your survey).
Also note that nonsampling error (such as measurement error/data quality) can be more of a problem if the sample sizes you find you need to attain are too large for your resources and logistics. That will inflate the variance. Please consider researching the term "total survey error."