I suggest you use proportional stratified random sampling techniques to determine the sample size and then use simple random sampling or systematic sampling to identify who/what/which are your samples/participants.

You should Google "sample size calculators." One approach will give sample sizes based on the precision of the estimate you want. Another approach (g*power) is based on the size of the effect you want to represent.

I think the best thing to do is first of all know the confidence interval within which you want to estimate your sample size. And then read literature to see similar works that has been done and use it to guide your choice of sampling and sample size.

Hi

I think, since your study is a descriptive or cross-sectional study, consideration of whether a previous study in your area has taken place can be effective in estimating sample size. However I would recommend a specialist in the field of Epidemiology, Biostatistics or to enter the study.