Further, there is dependence on how first and second kind errors are balanced, which have been assumed equally important here (which is not to say that they are of equal size; also this topic
has been discussed in the first part and in Bossew, 2014b). The results are applicable here as well, in that higher importance of second kind errors (erroneous non-assignment as Rn prone) would
lead to more geological classes labelled Rn prone.