The above procedure for testing the statistical significance of performance differences between types can also be used to explore the potential benefits of combining the two typologies. It might be argued, for example, that despite (possible) structural similarities PR regions are, prima facie, very different, from Intermediate ones, and that the Structural Typology should therefore be applied to the Intermediate and PR regions, or to each of the D-P types separately. Figure 4 provides some initial indications that such a combination of the two typologies might not provide additional discrimination, since a number of the combinations which fail.
