Bookmark and Share

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]

Re: st: DHS svy questions on weights and merged datasets


From   Steve Samuels <[email protected]>
To   [email protected]
Subject   Re: st: DHS svy questions on weights and merged datasets
Date   Fri, 22 Jun 2012 11:15:05 -0400



1. I think that you need to acquire some basic knowledge of sampling before
going any further.  I recommend, at a minimum
http://www.statcan.gc.ca/edu/power-pouvoir/ch13/5214895-eng.htm. See especially:
http://www.statcan.gc.ca/edu/power-pouvoir/ch13/estimation/5214893-eng.htm. Once
you understand what a sampling weight is, the answer to your first question
should be clear.  Also, try your -egen- statement and take a look at the
resulting variable.

2. Stas's regression line was a generic snippet and has nothing to do with your
problem.


Steve
[email protected]




http://www.statcan.gc.ca/edu/power-pouvoir/ch13/estimation/5214893-eng.htm



On Jun 22, 2012, at 10:17 AM, Julian Doczi wrote:

Dear Mr Samuels / Statalist,

Thank you very much for the response - your advice is incredibly useful!

Just a couple quick follow-ups to clarify (either for you or anyone
else on Statalist):

First, in the link you posted from Stas
(http://www.stata.com/statalist/archive/2008-10/msg00521.html), do I
need to -egen group- the weight variable (v005) by year as well?
(weightXyear?) Or, since both sets (in 1998 and 2008) have means of 1
million and many replicate values already, is it not necessary?

Second, again in Stas' link, he states that I should then carry out
estimation using the following code: - xi: svy : reg response i.year
other controls - . However, I plan to carry out diff-in-diff
regression, which usually takes the form: - svy: reg depvar timedummy
treatmentdummy time*treatmentinteraction - . Considering the need to
already account for the year by using -xi-, does Stas' -i.year- term
conflict with my diff-in-diff -timedummy- term? Do I need both due to
the 'super-strata' nature of this survey, or is one or the other
sufficient?

Thanks again for the very useful assistance!

--
Julian Doczi (Mr.)
University of East Anglia, Norwich, U.K.
[email protected]
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index