When subpopulations of the population to be studied vary considerably, it is advantageous to sample each subpopulation (stratum) independently and produce a
weighted mean that has less variability than the
arithmetic mean of a random sample. There are several possible strategies:
- Proportionate allocation uses the same sampling fraction[?] in each of the strata.
- Optimum allocation takes larger samples in the strata with the greatest variability to generate the least possible sampling variance.
Sampling equal numbers from strata varying widely in size may be used not only to improve the accuracy of estimation but also to equate the statistical power of tests[?] of differences between strata.
- cluster sampling -- multistage sampling -- simple random sampling
back to survey sampling