Odds Ratio with melogit

[Adriana Martinez]

Hello,

I'm trying to run a null model (intercepts only) with an ordinal outcome. I'm using the melogit command and I've asked it for odds ratio, however, the output is giving my a negative odds ratio. Since odds ratios cannot be negative I am trying to understand how to make sense of this output, or how to identify the problem and see if I can fix it.

[Rich Goldstein]

you did not show your exact command but I believe that this is -meOlogit- not the -melogit- you say it is; in ordered models there are cutpoints and those are not what you apparently think they are; the manual entry for -ologit- has a discussion and you should look at that (or at a good text)