Sampling design

In the theory of finite population sampling, a sampling design specifies for every possible sample its probability of being drawn.

Mathematically, a sampling design is denoted by the function $P(S)$ which gives the probability of drawing a sample $S.$

An example of a sampling design

During Bernoulli sampling, $P(S)$ is given by

:$P(S)\; =\; q^\{N\_\; ext\{sample\}(S)\}\; imes\; (1-q)^\{(N\_\; ext\{pop\}\; -\; N\_\; ext\{sample\}(S))\}$

where for each element $q$ is the probability of being included in the sample and $N\_\; ext\{sample\}(S)$ is the total number of elements in the sample $S$ and $N\_\; ext\{pop\}$ is the total number of elements in the population (before sampling commenced).

See also

*Statistical sampling
*Bernoulli sampling

Further reading

* Sarndal, Swenson, and Wretman (1992), Model Assisted Survey Sampling, Springer-Verlag, ISBN 0-387-40620-4