Significance Testing Between Different Variables

[Deirdre Colburn]

Hey all not sure if anyone can help me think through this problem...... I'm trying to conduct significance testing between several incident characteristic variables (table rows) for a handful of different aggregate incident types (table columns). If the rows represented different values of one variable, this would be straight forward. BUT, I've created a separate variable for each row because the categories are not mutually exclusive (i.e. a respondent could fall into more than 1 row ). I've used the -csti- command in Stata to do this sort of analysis before, but I don't think that is how it is normally used and didn't know if anyone had faced a similar problem before.

[Carlo Lazzaro]

Deidre:
welcome to this forum.
Please read and act on the FAQ on how to post more effectively. Thanks.
No example/excerpt of your data does not help.
That said, in the meantime you can try to gather together different features via the -group- function available from -egen-.