Eamonn O'Brien wrote:
I have variables var1, var2, var3 & var4. Variables var1 var2 & var3
are three independent results of a screening test. The independent
result (ordered categorical) is either normal, borderline, mild, mod or
severe. Variable Var4 is the final result and it is meant to be based on
the majority decision. So if var1 & var3 are borderline for example,
var4 should be borderline, regardless of var2. Another example, if var2
& var3 are severe, var4 should be severe regardless of var1. Where all
three differ I would like to check that var4 is the worst of var1, var2,
var3. If I can get started I think I can develop it to check for missing
values, but I am not sure if I should approach this problem with if
statements or -assert- or something else?
--------------------------------------------------
Let the values be 1-5, with 1 meaning normal and 5 severe. Now,
according to your rules:
var1 var2 var3 var4
3 3 4 3
2 3 4 4
In the first case, the result is 3 (majority decision). In the second
case, the result is 4 (all three differ; take the worst). This does not
look consistent to me. A simple, but less inconsistent solution is to
take the mean:
egen v4=rowmean(var1-var3)
generate var4=round(v4)
You should decide what to do with 2.5; should it be rounded up or down?
Svend
__________________________________________
Svend Juul
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6
DK-8000 Aarhus C, Denmark
Phone: +45 8942 6090
Home: +45 8693 7796
Email: [email protected]
__________________________________________
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/