Complex Samples Ordinal Regression - Estimated means?

[Sid Sarinopoulos]

Is there a way to get the program to output estimated means?
I've looked at manual but couldn't see how

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

NOTE that the same question was cross-posted to comp.soft-sys.stat.spss.  You can view that thread here:  https://groups.google.com/forum/#!topic/comp.soft-sys.stat.spss/mha6siHSrbU

----------------------------------------------------------------------------------

Hi Sid.  I have never used CSORDINAL, but I see that the OUTFILE with MODEL lets you save an XML file containing the parameter estimates and the parameter covariance matrix (first link below).  After that, using an input file that has the desired combinations of the explanatory variables, you should be able to get the fitted values you want by "applying the model" to the new dataset.  See the second link below.

The third link below is for a YouTube video showing how to do this for ordinary logistic regression.  Again, it might help give you the general idea.

https://www.ibm.com/support/knowledgecenter/SSLVMB_21.0.0/com.ibm.spss.statistics.help/syn_csordinal_outfile.htm

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_23.0.0/spss/tutorials/scoring_applying_model.html

https://www.youtube.com/watch?v=eboTREc3uSk

HTH.

[Art Kendall]

What is  the meaning of your DV?  How many values can it have?

Why do you believe that the intervals are severely discrepant from being equal?