| Re: Mismatch of file names within the same survey [message #24303 is a reply to message #24247] |
Fri, 15 April 2022 16:01  |
admin
Messages: 50
Registered: November 2012
|
Senior Member
Administrator |
|
|
Following is response from DHS Research & Data Analysis Director, Tom Pullum:
You have identified a nuisance that we often have to deal with. Column 6 of the filename is a "version" code. Sometimes (for a range of reasons) the DHS data processing staff will update a file and will increment the version code (for example from 1 to 2) for that file. If the other files for that survey are not affected, their version codes will not be changed. If you previously would have merged BRxx51 with PRxx51, you would now merge BRxx51 with PRxx52, for example.
When looping through lots of surveys, you may hit a problem like this. A related problem comes up if both files were 51 and now both are 52. You have to have enough flexibility to handle such changes/updates in filenames.
Note that you need the version number because the phase identifier in column 5 is not sufficient to identify a survey. For example, Egypt had both 51 and 5A surveys--two surveys within DHS-5.
We now have repositories of code written in Stata and SPSS available on Github. Please reference these code repositories as a resource for code for matching or calculating DHS indicators. The code repositories can be found at:
https://github.com/DHSProgram/DHS-Indicators-Stata
https://github.com/DHSProgram/DHS-Indicators-SPSS
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
