This then seems a reasonable way to . Judge the results as though they were based on samples and then look at the possibility that they may be by the way the sample was, in fact, . However, this on the inescapable duty of describing carefully how their experimental material - including subjects - was actually . It is also good practice to attempt to foresee some of the objections that might be made about the quality of that material and either attempt to forestall criticism or admit openly to any serious defects.

When the subjects themselves determine to which experimental group they belong, whether deliberately or accidentally, the sampling needs to be done rather more carefully. An important of the Fletcher & Peters (1984) study mentioned earlier was to compare the speech of language- with that of language-impaired children. In this case the could not children to one of these groups - they had already been classified before they were . It is important in this kind of study to try to avoid choosing either of the samples such that they belong to some special subgroup.

There is one type of for which proper sampling is absolutely essential. If a reference test of some kind is to be , perhaps to lack of language development in young children, then the test must be applicable to the whole population and not just to a particular sample. in determining cut-off levels for children who should be given special have (e.g. too great a demand on ) and social (language-impaired children not being ). For studies of this nature, a should be recruited before any is collected and before a sampling frame has been .

With this introduction to some of the problems of the relation between sample and population, we now turn in 5 to the of probability as a in providing a between the properties of a sample and the of its parent population. In the of that we outline a for sampling. 6 deals with the modelling of populations, and introduces the , an important model for our purposes in characterising the relation between sample and population.

(1) We have passed over here an which we have to postpone for the moment, but which is of importance for much of the done in language studies. Imagine the case where the population of interest is utterance lengths of British English-speaking pre-school children. We have to consider whether it is better to a sample which of many utterances from a few children, or one which of a small number of utterances from each of many children. We will return to this question, and the general of sample size, in 7.

(2) The of the gathered by such sampling can become quite complicated and we will not deal with it in this book. Interested readers should see on sampling or or an experienced survey .

(3) The raised in the first footnote crop up again here; measurements on linguistic are more than birth weights. We could again ask whether we should collect many word- plosives from few children, or few plosives from many children (see 7). A problem arises with the sample chosen by several stages. Is it better to choose many and a few Health Districts in each or vice versa?