Hi,
I'm using the commands under the analyze/GLM/multivariate menu in order to
build a multivariate regression model with multiple DV's. I'm more
familiar with working with the 'regular' multiple regression commands (ie,
1 DV), and I have some questions about how the 2 compare with each other.
1 - The GLM/multivariate menu contains commands which will calculate the
interaction effect of 2 variables (under the 'model' subcommand menu).
When you're testing for interactions in multiple regression, you have to
manually center and create new interaction terms, and then enter them into
the model. Do you have to center variables, or do anything else to create
interaction variables, when using the GLM/multivariate menu?
2 - Using the GLM/multivariate options, how do you build models in order
to 'control' for certain variables, or to enter blocks of varaibles before
other blocks? The 'model' subcommands allow you to customize which effects
you want to test for, but it's not clear how or whether you can enter
variables in stepwise fashion.
Thanks
Josh
=====================
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
Locked
