Re: Visualization for Categorical DV and Continuous IVs
Posted by
Rich Ulrich on
URL: http://spssx-discussion.165.s1.nabble.com/Visualization-for-Categorical-DV-and-Continuous-IVs-tp5740587p5740601.html
I haven't done the multinomial regression that Jon suggests, but
what I have used when the categories are relevant is the
discriminant function for multiple groups. It has nice ancillary
statistics and displays.
The plot of category centroids on the first two discriminant factors
shows the group separation, and the factors can be interpreted by
their loadings like a factor analysis. The "within-group" correlations
are also available.
--
Rich Ulrich
It really depends on what you want to show, but one possibility would be to fit a multinomial regression and then plot the category probabiliities against the regressors or something like emm's,
Can you suggest a graphical presentation for a Categorical DV and 4 Continuous IVs.
Thank you.
Lema
--
===================== 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
=====================
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