PS SAMPLING VERSUS STRATIFIED SAMPLING — COMPARISON OF EFFICIENCY IN AGRICULTURAL SURVEYS
Marcin Kozak, Robert Wieczorkowski · 2005
The paper contains considerations on efficiency of three sampling schemes, i.e. simple random sampling without replacement, stratified sampling, andps sampling, in an application to large agricultural populations. A simulation study using data from Agricultural Census 2002 was carried out to make a comparison of the schemes. Stratification of the populations was carried out using a method of Rivest (2002), and Hartley-Rao scheme with randomly permutated units was chosen as the most popularps sampling scheme.