Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Fwd: Interpreting dummy coded interactions with 3 levels


From   Priscilla Carver <prou1129@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Fwd: Interpreting dummy coded interactions with 3 levels
Date   Wed, 18 Feb 2009 11:02:25 -0500

I am wondering if I can get advice on how to determine if an
interaction is significant in STATA that has 2 dummy coded variables
and one dummy coded variable has 3 levels.  Using Ologit (by the way).
 Years and years ago I used SPSS. In SPSS, blocks could be specified.
In the first block I would enter the main effect; in the 2nd block I
would enter the interaction terms.  The interaction was only
considered significant if the Rsquare change was significant for the
2nd block.  (btw, I do not have a license for SPSS now--so I can not
use that).  Is there a way to do that in STATA?  if not, how can I
interpret whether the below interaction terms are significant?

Here is what the simplified model looks like:

Dependent Variable:  Education spending (response coded as 1, 2 or 3).

Independent variables:
Nat_security1 spending dummy 1 (recoded as follows: 1=1 & 2 or 3=0)
Nat_security2 spending dummy 2 (recoded as follows: 2=1 & 1 or 3=0)
Gender (female=1; male=0)
Nat_security1*gender
Nat_security2*gender

Thanks much!
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index