|
One change with this edition of Sampling is that I have included sections of
computing notes for sample selection, calculation of estimates, and simulations.
These computations are illustrated using the statistical programming language R.
In doing this I have avoided the use of specialized packages for specific complex
designs, choosing instead to show simple calculations and sampling procedures
from scratch using a few basic functions. The purpose of these sections is as much
for understanding of sampling ideas as for easy ways to select samples and calculate
estimates. Other software than R can, of course, be used for the same purpose.
The advantages of R include: it is a free and open source, is widely supported by
the statistical and other research communities, is available to anyone, and is easily
installed on a computer with any of the common operating systems, including
Windows, Macintosh OS X, Linux, and other types of Unix. The syntax of R tends
to read like generic code and conveys the thinking that goes along with calculations
rather than serving as a magic box. R is interactive and has very nice graphics.
Once one learns how to select a sample with a given type of design and to produce
|