|
The first step is the collection of samples. Statistically, a sample can be defined as an individual or group of individuals drawn from a large or infinite population, Information obtained from samples is only as representative of the material as the samples on which they are performed. If observations reveal little variation and there has been no bias in collecting, then a small sample or small number of samples may be highly representative of a population. If the variation is large then more and/or larger samples will be required Representative sampling, however, is perhaps the most difficult of the control operations to perform satisfactorily. Sampling, as with all types of test, introduces sources of variation and error, so that judgements of materials based on infrequent random tests are fraught with difficulties. In this connection, see also the remarks on sampling in relation to classification given in Chapter 6.
...
This 250-word extract was created in the absence of an abstract.
