Introduction

The previous chapter was concerned with what can be described as the classical theory of sampling finite populations. This theory covers many of the sampling problems that are likely to arise in a research study, but there are situations where different approaches are required. In particular, cases sometimes arise where it is not possible to decide in advance where and when the items in a population will be sampled. Instead a sampling scheme must be devised which allows items to be encountered with a certain probability. The analysis of data must then take into account the nature of this sampling scheme.

A number of different types of encounter sampling are reviewed in this chapter. First, some methods for estimating the number of individuals in a population are discussed. These are mainly of value in a biological setting, for example for estimating the number of mice living in a certain region. Next, size-biased sampling is considered, where the probability of sampling a unit in the population of interest depends on a measure of the size of that unit.

This chapter also provides an introduction to the handling of spatial data. This includes distance-based methods, involving the measurement of distances between the items in a population or between random points in space and the nearest items, and also methods for analysing quadrat counts. With these types of data there are two questions that are often of interest: