| Merging child's malaria results from HH member recode dataset with other child data in Child recode [message #2846] |
Mon, 01 September 2014 06:56  |
kate2
Messages: 15
Registered: May 2014
|
Member |
|
|
Good afternoon,
I have looked through the other discussions on merging but am still not confident enough to go ahead without checking that I have merged my files correctly.
I am using SPSS and working on the Tanzania 2011 AIS/MIS files. I am using point and click in SPSS as I normally use STATA so don't have experience of using syntax in SPSS. I would prefer to stick with the point and click for this if possible.
I have opened the HH member recode file and the Child recode files. In the HH member recode I have renamed the variables HV001, HV002 and HV003 by dropping the h on each. Now both files have v001, v002 and v003 variables.
I have sorted each dataset by these three variables (ascending sort) and saved two new files called 'hh member sort' and 'child sort'.
I have then merged these files using the child sort as the active file and choosing the hh member sort as the one to be merged to the active one.
My lack of confidence in what I have done is down to not being sure whether the v001, v002 and v003 will sufficiently identify the correct child. I have done the merge with these three as I am following the table in your guidance notes on merging datasets which suggest merging child and hh member files with v001 and v002 (I have added v003 to be sure that I am merging by the same respondent). However, in the child and hh member recodes v001, v002 and v003 relate to the respondent. If the respondent has more than one child then won't there be duplicates of these which actually relate to different children? Do I in fact need to merge by a variable which is a unique identifier for the child? I can't see an obvious variable to use for this. B16 in the child recode is described as the 'child's line number in the household' but b16 is not replicated in the hh member recode as far as I can see.
Do you have any advice?
With thanks!
Kate
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
