Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: RE: Generating new variable conditional to another
From
Theocharis Kromydas <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: RE: Generating new variable conditional to another
Date
Sat, 23 Nov 2013 10:46:50 +0000
Dear Maarten
Thanks a lot for your reply
I reckon all the cases you mention before but there are some very extreme cases as there are participants who responded as having 0 years of education and a tertiary degree. (ISCED 5 to 6). Well this is impossible except the honorary degrees I guess,but again even a honorary degree holder should have some years of education obtained, I guess. What I have managed to do up until now is to replace all values from 1 (tertiary degree) to 0 (no-tertiary degree) and vice versa for all "extreme" cases for all countries, but again I am not getting any big differences in the final results. I am also intending to see how this works if I delete all these extreme cases (which cases are extreme of course is something I have to carefully think of) from my model and see the differences in the results. So technically speaking I think I know now how to do it but again it is a matter of argument if I delete, replace or keep the variables as they are now.
I am using a logit model, using employment status (0 unemployment 1 employment) as my dependent variable, regressed to educational years, degrees, other demographics, work experience and some other control variables for two periods, 2004 and 2010. Standard errors are clustered to the country level and the model is weighted using PWIGLS weights command. This is the first level model and I am intending to compare it with a two-level one using stata gllamm command.
Thanks again.
Harry
*
* 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/