Geoscience Reference
In-Depth Information
2
Standard Sa
mpling Methods and Analyses
Bryan Manly
2.1 Introduction
Often, the goal of ecological sampling is to summarize the characteristics
of the individual units in a biological population. For example, the charac-
teristics of interest might be the weight and sex of the individual animals in
a population in a particular area. A summary of these characteristics then
consists of estimates of the mean and standard deviation of the weight of
animals and an estimate of the proportion of females.
In the terminology of statistics, the
population
is defined to be the collec-
tion of all items that are of interest in an investigation. Thus, the items might
be individual animals or plants, but they could also be small plots of land,
pieces of rock, or groups of animals. As far as statistical theory is concerned,
it is crucial that the items that make up the population are sampled using an
appropriate procedure. For this reason, the items are often called the sample
units. Sometimes, population sizes are small enough to allow every item
to be examined. This then provides a census. However, the populations of
interest in ecology are usually large enough to make a census impractical.
The measures that are used to summarize a population are called popu-
lation parameters, and the corresponding sample values are called statis-
tics. For example, the population mean (a parameter) might be estimated
by a sample mean (a statistic). Similarly, a population proportion of females
(a parameter) might be estimated by a sample proportion (a statistic).
2.2 Simple Random Sampling
The use of random sampling is important whenever inferences are to be
made about population parameters on the basis of the sample result because
7
