The is of course - the time and money for collection and . In the light of this, sensible decisions have to be made about, for example, the number of Health Districts in Britain to be included in the frame; or it may be necessary to limit the inquiry to children born in four months in the year instead of a complete year. In this example, the sampling frame between the population of interest (which is the birth weights of all children born in Britain in 1984) and the sample, and allows us to generalise from the sample values to those in the population of interest.

If we now return to an earlier linguistic example, we can see how the sampling frame would us to our sample with a population of interest. Take word- VOTs. Our interest will always be in the of a relatively large group and in the measurements we from their behaviour. In the present case we are likely to be concerned with English children between 1;6 and 2;6, because this seems to be the time when they are learning to voiceless from voiced stops using VOT as a phonetic . Our will be limited. We should, however, at least have a sampling frame which sets time limits (for , we could choose for the lower limit of our age- children who are 1;6 in a particular week in 1984) ; we would like it to be geographically well-distributed (we might again use Health Districts) ; within the sampling frame we must a sample of a reasonable size. (3)

That is how we might go about children for such a study. But how are language samples to be from a child? Changing the example, consider the problem of utterances from a young child to measure his mean length of utterance (mlu - see 13). Again it is possible to devise a sampling frame. One would be to a radio microphone to the child, which would and record every single utterance he makes over some of time to a tape-recorder. Let us say we record all his utterances over a three-month . We could then a number to each utterance and choose a simple sample ( 5) of utterances. This is clearly neither sensible nor feasible - it would an unrealistic expenditure of . , and more reasonably, we could divide each month into days and each day into hours, a few days at and a few hours within each day and record all the utterances made during the hours. (See Wells 1985: 1 , for a study of this kind.) If this of were to be used it would be better to that that child is active only between, say, 7 am. and 8 p.m. and hours from that time .

In a way, it will always be possible to imagine how a sampling frame could be drawn up for any population if time and other were unlimited. The which all the usual that, if the results from a sample are to be generalised to a wider population, a suitable sampling frame has been and the sample chosen from the frame. In practice, however, it is frequently impossible to draw up an acceptable sampling frame - so what, then, can be done?