STATA Logistic Regression Coefficient Plot Not Working?

[Aarushi Jain]

Good Day all!
I need some urgent help because I can't seem to understand why my coefficient plot seems to be wrong.

The first two photos show the 2 logistic regressions I am trying to plot on the coefficient plot. For each IV which measures my DV, there is a dummy (base) variable, for which the value is 1 as it is an odds ratio. Now, when I plot this using the following code:
logistic occupbin i.educat age i.incomecat, baselevels
estimates store occupbin
logistic classpercbin i.educat age i.incomecat, baselevels
estimates store classpercbin
coefplot (occupbin, label(Objective Social Class)) (classpercbin, label(Subjective Social Class)), drop(_cons) xline(1) baselevels

I get the graph in the third picture. I am incredibly confused because the base variables are at 1, and the information displayed on this graph does not match the information on the tables (Pictures 1 and 2) at all. Am I doing something wrong?


I don't know why this is happening, and it has been a reoccurring issue. Thank you for taking the time to read this and I hope to get a response quickly!

[Ken Chui]

It's showing the results of -logit- rather than -logistic-. If you wish to show the odds ratios in the graph, add the option -eform- as one of the options in the -coefplot- command. If you do not like the resultant confidence intervals being so asymmetric, add another option -xscale(log)- to covert the axis to log scale.

[Aarushi Jain]

Ken Chui Thank you so much for all your help and efforts! I greatly appreciate it. My graph is showing exactly what I need it to.