The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Bangladesh » Children from same household
Children from same household [message #14225] Tue, 06 March 2018 22:46 Go to next message
phyumar is currently offline  phyumar
Messages: 17
Registered: February 2018
Location: Singapore
Member
Hello Sir,

I am looking at Bangladesh DHS data (BDKR72FL) file for estimating factors influencing common childhood illnesses.
It would be great that if you could enlighten me the following questions:

- When I tried to compare my weighted results with Bangladesh DHS report page 142, Table 10.4 Diarrhoea and background characteristics particularly by sex.
I got the same estimated proportion and absolute number of children who got diahorea in last 2 weeks. However, I got different results in total number of children with or without diarhoea dis-aggregated by sex.
The different is 2 absolute number for male and 4 for female. These are all weighted. Could you kindly let me know why I don't get the same number of children?
Here is my result table,
svy: tab b4 h11,count
(running tabulate on estimation sample)

Number of strata = 21 Number of obs = 7560
Number of PSUs = 600 Population size = 7754.4242
Design df = 579

--------------------------------------------------
sex of | had diarrhea recently
child | no yes, las don't kn Total
----------+---------------------------------------
male | 3803 232.2 13.37 4049
female | 3490 208.2 7.439 3706
|
Total | 7293 440.4 20.81 7754
--------------------------------------------------
Key: weighted counts

The report described 4051 male and 3710 female.


- I also want to know the children from the same household. Which variable represent this information? I see that the variable "caseid" has the same numbers and hidx has different numbers then can we assume that these children are from same household!

- If I want to know birth order of the child, can I use the variable "bord"?

Thanks you so much in advance for clarifying all those queries.

Best regards,
Phyumar
Re: Children from same household [message #14297 is a reply to message #14225] Tue, 20 March 2018 08:41 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1504
Registered: February 2013
Senior Member
Following is a response from Senior DHS Stata Specialist, Tom Pullum:


Sorry for the delay with this reply. First, I am looking into the extra six cases that seem to be in table 10.4. Will let you know when I know more about that. Second, caseid is the id code for the mother. It is a string that includes v001, v002 v003 from the IR file, which are the same as or hv001, hv002, hvidx in the PR file. Children with the same value of caseid will have the same mother, and be in the same household, but there can be more than one eligible woman, or mother, in the same household. Children in the same household will have the same values of v001 and v002 (or hv001 and hv002). The line number in the household file (hvidx) of the child is given in the KR and BR files as b16. Third, the birth order of the children of a specific mother is indeed given by bord, which is numbered from the first birth to the most recent birth. By contrast, bidx is numbered backwards, from the most recent birth to the earliest birth.
Re: Children from same household [message #14308 is a reply to message #14297] Tue, 20 March 2018 22:16 Go to previous messageGo to next message
phyumar is currently offline  phyumar
Messages: 17
Registered: February 2018
Location: Singapore
Member
Hello Sir

Thanks for your great reply! Yes, I will just use v001 and v002 in order to get the children from same household. One more query is that why b16 in KR file( child's line number in household) got some missing values or not list although the children were already in datasets. I mean this is not very important. Just checking why it is.

Re: Children from same household [message #14319 is a reply to message #14308] Thu, 22 March 2018 11:55 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1504
Registered: February 2013
Senior Member

Following is another response from Senior DHS Stata Specialist, Tom Pullum:

If a child has died, indicated by b5=0, then b16 will be NA and coded ".". If a child has survived but is not living in the same household as the mother, the line number is "0".

Previous Topic: Regarding categories s823
Next Topic: Interpretation of stunting
Goto Forum:
  


Current Time: Tue Dec 11 02:52:03 Eastern Standard Time 2018