ABSTRACT
The derivation of a sample of small businesses which is representative of the population of small firms is a truly formidable problem. The small firm sector fluctuates markedly from year to year with, in the UK, more than 10 per cent of the stock changing annually. Furthermore, business failure rates are disproportionately high for the youngest firms. Of those firms which fail within ten years of starting business, 50 per cent of failures occur in the first 2½ years, 33 per cent in the next 2½ years, and only 17 per cent in the following five years (Ganguly, 1985). If the dynamics of the small firm population are to be modelled over a period of time, in order that public policy towards the sector can be made more effective, then the collection of firms analysed has to include both substantial numbers of new firms and those which have ceased trading. It is wholly misleading to concentrate upon survivors, although as we shall demonstrate ‘missing’ data do present considerable problems in making comparisons.
