Hi all! I'm working on a Table indicating the frequency of 48 different words identified in a two-page text. Can you suggest a statistically sound way to set a cutoff point in order to divide these words in two groups (e.g. main and secondary words) taking into account their frequency? For instance, I was thinking of dividing the maximum frequency which is equal to 9 with two, and then setting 4.5 as the cutoff point. Words above this cutoff point could be considered as "main'' and the rest as "secondary". Does this make sense? Is there a name of this cutoff point technique I'm suggesting and any relevant literature to support this? Are there other, cutoff point techniques you could suggest? Many thanks, in advance!