Hi folks,
I'm not a statistician, so please forgive my naivete. I am trying to design
a secondary analysis of an existing database with two categorical
variables--the problem behavior noted in the child, which is a checklist of
a dozen or so symptoms that are not mutually exclusive, and the
categorization applied to the case by child protection authorities--i.e.
physical abuse, sexual abuse, neglect, abandonment, parental incapacity,
complaint not warranted--the latter being mutually exclusive categories.
What I have in mind is a Chi-Square test of independence for an R X C
table, but this appears more complex.
TIA,
David Klein
|