Table 3. Eight different scaling models tested for the 12 JCSG data sets, corresponding to the normalized merging residuals shown in Fig. 6 ▶ .

The first run corresponds to a very simple scaling model, while the last includes all corrections and run 4 corresponds to the CCP4i default.

Scaling run	Partiality correct (i.e. ‘tails’)	Decay correction (i.e. ‘bfactor on’)	Absorption correction (i.e. ‘secondary 6’)
1	No	No	No
2	No	No	Yes
3	No	Yes	No
4	No	Yes	Yes
5	Yes	No	No
6	Yes	No	Yes
7	Yes	Yes	No
8	Yes	Yes	Yes