| Pooling 3 rounds of DHS dataset [message #13513] |
Sun, 12 November 2017 17:25  |
mianrashid
Messages: 13
Registered: February 2016
Location: Pau, France
|
Member |
|
|
Dear members, I want to pool the three rounds of DHS Pakistan carried out in 1990-91, 2006-07 and 2012-2013. Please guide me, how can I do that?
MianRashid
|
|
|
|
| Re: Pooling 3 rounds of DHS dataset [message #13524 is a reply to message #13513] |
Tue, 14 November 2017 08:30  |
Bridgette-DHS
Messages: 3199
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
Because these surveys sampled completely different individuals, you cannot link individuals in successive surveys. Therefore you cannot do a merge, as such. Instead you construct a very long file that just adds the files together. In Stata, this is done with an "append" command.
Combining the three surveys into a single file can simplify some data processing if you are looking at trends, for example. However, I strongly recommend against, say, calculating the mean number of children ever born for the combination of the three surveys. Such a mean has no reference point in time and is affected by the arbitrary differences in the sample sizes and is not interpretable.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
