Re: st: significant to non significant variable

[daniel klein <klein.daniel.81@gmail.com>]

Might it just be that the effect of X2 is completely mediated by X3?

Or, more general, would it perhaps make more sense to approach this
from a theoretical point of view first, rather than assuming some
methodological artifact here, as your question implies? For this you
will have to share a lot more about your theoretical background and
the research question you are trying to answer with these models.

If you chose to stay on a rather abstract level, please provide at
least more information on your sample size, the change in coefficients
vs.the change in standard errors etc. to get better answers.

Although, if there is really no collinearity in the sense that the X
are uncorrelated, this begs the question of why you are running a
regression model instead of just looking at bivariate statistics?

Best
Daniel

-- 
[...]
xtmixed y X1 X2 || country: || Microfinance . In this regression  X1
and X2 are statistically significant

Now I add  X3  the regression becomes:

xtmixed y X1 X2 X3|| country: || Microfinance

In this regression X2 is no longer significant. The same problem
arised when I add more than one variable

I've checked for mulicollinearity among the variables with  VIF and
Collin and all the VIF . There is no  collinearity . All the ViF are
less than 2. So there is no mucollinearity
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/