Hello,
I have read that to interpret a b coefficient when the DV is log-
transformed (using the natural log), I can simply multiply 100*b to obtain
the proportional change in the DV given a one-unit change in the
predictor, e.g., b=.045 means a 4.5% increase in the DV for a one-unit
change in this control variable. HOwever, is this quick computation valid
only for b's that are less than .1? Can this logic apply to b's of,
say, .2?
Thanks for any insight!
Shannon
=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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
