Home » Data » Merging data files » Merging child's malaria results from HH member recode dataset with other child data in Child recode
| Re: Merging child's malaria results from HH member recode dataset with other child data in Child recode [message #2900 is a reply to message #2854] |
Mon, 08 September 2014 15:06   |
kate2
Messages: 15
Registered: May 2014
|
Member |
|
|
Hello again Bridgette.
I hope you don't mind if I ask you another question as I move forward with trying to do this. I think the problem is I don't understand fully how the datasets are structured.
In the child recode as I understand it, each case is a child. So v001, v002 and v003 themselves don't identify each unique child - for example in cluster 1, hh number 32 ( v001 = 1, v002 = 32) there are three cases all with v003 = '3', so the same woman with respondent line number '3' has her information on three different children broken down into three different cases, 1 per child. So far so good.
In the household member recode as I understand it, each case is a household member. So here we can see one case for v001 = 1, v002 = 32 and HDIX = 3. Presumably, then, all the information this woman has given about her three children is included in this one case. I don't see though how this is done in the variables. As far as I can see there are no variables with $ that indicate variables are repeated for each child, for example.
Most of the data I need are in the child recode and I will need to analyse by child. But I need to somehow merge into the child recode, the information on child's parasitaemia, whether they slept under a create net the previous night, and whether their household was sprayed with IRS in the past 12 months. Those pieces of information are in the hh member recode. At the moment I am struggling to see how a merge on the variables v001, v002 and v003/HDIX will give me a dataset in the format I need, i.e. with each case being an individual child. I am not sure how the merge I am trying to do will add the variables that I need against each child case in the child recode.
In any case when I try to merge as I think I am supposed to based on the table your referenced (matching v001 with hv001; v002 with hv002 and v003 with HDIX, with the appropriate renaming and sorting done first) I get the warning message that these my merge by variables don't uniquely identify cases. Which I can see when I look at the data itself but can't work out an alternative.
Sorry for additional questions and I will be much grateful for your guideance.
Best wishes
Kate
|
|
|
|
Current Time: Fri Nov 8 12:36:52 Coordinated Universal Time 2024
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
