Date:         Fri, 2 Dec 2011 14:10:01 -0800
Reply-To:     J McClure <mc006@PACBELL.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         J McClure <mc006@PACBELL.NET>
Subject:      coding new mutually exclusive multilevel variables
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi,

I have a dataset with 5 variables representing progressively more severe
levels of suicidal ideation and behavior, each has the same 4 possible
responses: 0=not at all  1=once    2=2-3 x  and 3= more than three times.
Because individuals can have multiple levels of severity at the same
time, participants could positively endorse more than one of the variables.

I now want to create 5 new mutually exclusive variables representing the
same levels of severity and the same 4 possible responses.
For example, a participant who endorsed passive ideation=0  active
ideation >3x week, and making a suicide plan 2-3xwk, and preparing that
plan 1xwk on the survey would have the new variables coded as 0 for
passive, 0 for active, 0 for plan, and a 1 for preparing plan.

The goal is to answer the question: what is the highest level of
suicidal ideation and behaviors experienced in the past week  and how
often did it occur?
Any help will be greatly appreciated,
Jan