Hi out there!
I have decided to "center" all independent variables entered into the
regression equation, not just the variables involved in the
interaction term as this gives better Collinearity Diagnostics,
specifically Condition Indices.
Compared to just centering variables (main effects) that are involved
in the interaction term the only difference apart from the Condition
Indices is the CONSTANT term. This means the equations will be
different. Are there any reasons why I should not center all
independent variables? How do I explain the different equations that
result from centering all, or only centering the variables involved
in the interaction term?
Karen
=====================
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
