Quote:
Originally Posted by nucleardonkey
Don't the calculations of sample size suppose spatially homogeneity to a certain degree? Miami is blue but with very localized sharp red peaks. That variation could be missed by insufficiently fine sampling.
There's a big difference between sampling from a uniform distribution and sampling from an al lost uniform distribution with highly localized concentrations of mass.
Hi donkey:
You bring up a valid point. Back in my Census Bureau survey design days, we used "systematic random sampling" to account for this and the variance calculation is a little different.
If you know that in certain areas, such as your Miami example, there are pockets which differ from the overall population, you can go to what is known as stratified random sampling where you sample the different (but smaller) areas at a higher rate and then adjust the weighting to account for this. This has the effect of incrfeasing the overall variance by a little.
Whether any of the pollsters do anything like this I don't know. But I suspect that systematic random sampling is used. Also, definitions and explanations of systematic random sampling and stratified random sampling can be found on the Internet for anyone interested.
Best wishes,
Mason