I don't know what academic field the original questioner is in, and the
question doesn't seem to have a lot to do with Stata, but for what it is
worth there is a very good and clear discussion of how to interpret
multiplicative interaction effects (in the context of the political
science literature) in a recent article by Bill Clark et al, available
at http://homepages.nyu.edu/~mrg217/interaction.html.
JY
-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Anderson,
Bradley J
Sent: Wednesday, September 20, 2006 10:34 AM
To: 'statalist@hsphsun2.harvard.edu'
Subject: st: RE: interaction between two continuous variables
The interaction means the the slope of one variable (say X1) varies as a
linear function of the second variable (say X2). Something you might do
is
to estimate the effect of X1 at selected values of X2 (e.g., at -2SD, at
-1SD, at the mean, at +1SD, at +2SD). The selected values of X2 may be
guided by theory. You might also plot the slope of X1 at selected
values of
X2. Of course, the interaction also implies that the effect of X2
varies as
a linear function of X1. But in general, substantive considerations
will
serve as a guide us here. For example, we might be interested in
describing
how the effect of depression on drug use is conditioned on age. But we
probably wouldn't be interested in how the effect of age on drug use is
conditioned by depression.
-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of stata_user
stata_user
Sent: Wednesday, September 20, 2006 11:59 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: interaction between two continuous variables
Hi,
I have the following model y continuous , x1 x2 continuous
I do a regression with
reg y cx1 cx2 cx1*cx2
with
cx1= x1 -mean (x1)
cx2= x2 -mean (x2)
both the paramters of cx1 and cx2 are signficants , which mean that the
slope of x2 is signficant for x1=mean (x1)
the paramter of cx1*cx2 is also signficant
How to interpret the parameter of cx1*cx2 ?
My first interpretation is that there are at least two values of x1 ,
where
the slopes of x2 are not parallels
but how to know, that the slope of x2 for a given value of x1=a is
signficant?
like using a lincom command
Any comments, please
Thanks
Sami
_________________________________________________________________
Be the first to hear what's new at MSN - sign up to our free
newsletters!
http://www.msn.co.uk/newsletters
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/