Birth order and pregnancy outcome [message #15486] |
Tue, 31 July 2018 02:01 ![Go to next message Go to next message](/theme/default/images/down.png) |
kclakhara
Messages: 30 Registered: June 2016 Location: New Delhi
|
Member |
|
|
Dear Sir/Madam,
I am analysing NFHS-4 data for the birth order (Table 4.6 in India NFHS-4 report) using variable "BORD" and selecting live births during three years preceding the survey, but figures are not matching between raw data and report. I have used data file "IABR74FL".
Kindly suggest which variable used for the computation of birth order.
Regards,
Kailash
|
|
|
Re: Birth order and pregnancy outcome [message #15763 is a reply to message #15486] |
Tue, 11 September 2018 11:08 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Bridgette-DHS
Messages: 3143 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Specialist, Tom Pullum:
We apologize for the delay in responding to your question about "birth order and pregnancy outcome". I have tried to match table 4.6 in the NFHS-4 report. I can only match the total number of births (147,598) by defining "three years preceding the survey" with v008-b3<=35. However, when I match the total I do NOT match the percentages at 1, 2, 3, 4+ exactly. Close, but not exact. I get 38.64%, 32.60%, 15.15%, and 13.62%. The report has 39.0%, 32.6%, 15.0%, and 13.5%. I will paste the Stata code below.
Normally the past three years is understood to be months 1-36, not including the month of interview. For example, that's how the interval for the fertility rates in the past three years would be defined. I believe table 4.6 is not a standard table. At any rate, what I have described is as close as I can get.
gen interval=v008-b3
gen bordr=bord
replace bordr=4 if bord>4 & bord<.
tab bordr if interval<36 [iweight=v005/1000000]
|
|
|