Dear allstaters
I am attempting to run a k-means cluster analysis of several ordinal
variables, each of which scores 1 to 4 (1-never, 2-sometimes, 3-often,
4-always). The items relate to the frequency of carrying out certain
activities.
Some of the items have very low counts for the categories of interest (i.e.
less than 10% frequency for the combination of categories 2, 3 and 4).
Is there any literature/does anyone have any knowledge on the effects of
including such low-count variables in a k-means cluster analysis?
Many thanks in advance
Roman Romero
PhD student, TCD
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|