ABSTRACT
In Sect ion 1.9 we discussed a simple default strategy for use when subject-matter knowledge and experience of s imilar sets of da ta suggest a na tura l s tar t ing point for the analysis. There are a num ber of reasons w h y such a strategy is inappl icable to the k ind of example described i n Chap te r 6. Some of these reasons are con cerned w i t h the prac t ica l i ty of handl ing complex sequences of ana lysis and others are broader conceptual issues.
