Dear all,
I would like to draw to sample of 384 cases from the national population in
a proportionate stratified sampling scheme with 2 strata involved.
The problem is that the 1st strata has 16 categories and the 2nd strata has
2 categories. This results that in some areas i should sample very few units
e.g. 2-5 units.
Is there any method to adjust the sample size and increase it? Are there any
further suggestions, or you think that it is OK since that the final
analysis will not be analysed by region and therefore 384 cases are enough.
Any suggestions for further reading are more than wellcome and appreciated.
Evie
|