Statalist


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

st: How to interpret a propensity matching result (log-transformation)?


From   Kristian Jakobsen <pisfin@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: How to interpret a propensity matching result (log-transformation)?
Date   Wed, 10 Jun 2009 18:04:26 +0000

I have used pscore and psmatch2 in order to evaluate a given treatment on households' consumption. Before estimating the effect with psmatch2 I took the log to the consumption variable. And I got the following result:
 
Variable Sample Treated Controls Difference S.E. T-stat
      
Sample        Treated     Controls     Difference  S.E.      T-stat
Unmatched     5.96657475  6.11893931 -.152364559 .030307387 -5.03
ATT           5.96478846  6.04441313 -.079624672 .037496237 -2.12
 
As far as I can see the difference is significant on a 95% level, but how do I interpret the coefficient of the treatment when I have used log consumption as the dependent variable. Will this mean that the treatment lowers the consumption by 8% or does that not make any sense? Perhaps it is too basic, but I really hope for an answer.
 
Cheers,
Kristian
      
_________________________________________________________________
Show them the way! Add maps and directions to your party invites. 
http://www.microsoft.com/windows/windowslive/products/events.aspx
*
*   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