Home » Data » Merging data files » Merging data files in Stata
Re: Merging data files in Stata [message #2125 is a reply to message #2120] |
Tue, 06 May 2014 09:33   |
Bridgette-DHS
Messages: 3230 Registered: February 2013
|
Senior Member |
|
|
Here are comments from Tom Pullum:
Dear Ahmed--I do not like unnecessary complexity either, but the choice of the model depends on many things, including the level of measurement of the outcome variable (for example, is it an interval-level variable, or binary, or a count), and the sampling design (for example, is it a multi-stage cluster sample and were weights used). OLS normally refers to an interval-level outcome and a simple random sample (srs)--that is, the simplest possible situation. For most DHS users the outcome is NOT interval level, and the sample is never srs.
Please say a little more about what you propose to do after you pool several surveys from the same country? What is your research question? What is your outcome variable?
|
|
|
Goto Forum:
Current Time: Wed Feb 26 06:01:28 Coordinated Universal Time 2025
|