Re: st: Is multilevel model the correct model?

[Marcello Pagano <pagano@hsph.harvard.edu>]

One of the constants in the bleak winter is that it must come to an 
end.  But that does not detract from the value of the signs of Spring; 
they are surely welcome even if they predictably come each year. So the 
singing of birds, the day getting longer, the crocuses sprouting through 
the snow, these are all signs we look for in the North at this time of 
year. Other signs that also trumpets the end of the scholastic year are 
the exams and term projects that need to be completed. The tell tale 
signs of "x and y" that are common to many homework and exam questions. 
We sometimes see these on Statalist, even though the practice is 
discouraged.

Ah Spring!

m.p.



On 2/9/2011 8:34 AM, Shikha Sinha wrote:
> Hi there,
>
> I have individual level data for 30 countries, and I am exploring the
> effect of x on y. I estimate an OLS in the pooled sample with country
> fixed effect and then estimate the same model for each country
> separately. The aim of the analysis is to explore whether the
> relationships between x and y are associated in ways that vary between
> countries. Should multi-level models be used to demonstrate the extent
> to which there is national variation in these relationships? Kindly
> advise if the correct model is random intercept or random coefficient
> or may be an interaction model will do the trick?
>
> Thanks,
> Shikha
> *
> *   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/
*
*   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/