Natural Resource BiometricsSampling ProbabilityIn most situations, we assume that individuals in a population have a equal probability of being sampled. However, sometimes we are more interested in certain parts of the sample that in other parts and want to sample with unequal probabilities. This document is an attempt to describe the relationship between different sample methods and when to choose each. Probability Proportional to Frequency  PPFPPF sampling is the most common sampling method. As the name implies the individual that are most frequent in the population are sampled at the highest rate. In most cases this is perfectly acceptable.
Probability Proportional to Size  PPSPPS sampling is one method of sampling with unequal probability. In this type of sampling individuals are included with different probabilities. One way to think of this is that certain individuals are more important that others so we sample the important individuals at a higher rate.
Probability Proportional to Prediction  PPPPPP is a type of sampling that allow one to estimate a value that may be very expensive or time consuming to measure for a large number of individuals in a population and then validate that estimate.
Literature Cited

Natural Resources Biometrics by David R. Larsen is licensed under a Creative Commons AttributionShareAlike 4.0 International License . Author: Dr. David R. Larsen Created: October 14, 1996 Last Updated: September 13, 2014 