The mapping options controls how the sort data are aggregated. You have two filtering options:

Filter Method (Select one of the following):

1. No filter (the default). The data are aggregated as described in Constructing the Input.

2. Specify similarity cutoff. The value that you specify in the similarity cutoff box is applied to the final total similarity matrix. Essentially, values in the similarity matrix at or below the cutoff value are set to zero for the analysis. if you specify a value of 2, this means that if only 2 or fewer users placed two statements together in a category the analysis will treat it as though no users placed the statements together. This effectively filters out spurious relationships between statements. In a project where the subject matter is very ambiguous or where you suspect that there is lots of noise in the sorts, you might try recomputing maps with a cutoff value of 1 or 2. It would be extremely rare to use a cutoff value greater than 2, unless you have a large number of available sorts for the analysis.