| Contraceptive Prevalence Rate [message #18119] |
Mon, 23 September 2019 04:29  |
Srini
Messages: 1
Registered: September 2019
|
Member |
|
|
I am trying to calculate contraceptive prevalence rate (CPR) among currently married women (CMW) in India based on the DHS 2015-16 dataset. In the numerator I am including all CMW who are currently using any method of contraception.
1. To get the count of CMW in the denominator, I am using the variable V502 and selecting factor level = "1". This gives me D = 499,627 records.
2. In the subset of data obtained above, to get the count of number of CMW who are currently using any method of contraceptive, I am using the variable V361 with factor level = "10" (currently using). This gives me N = 252,603 records.
3. Thus CPR among CMW = N/D*100 = 50.6%.
But, the summary reports of DHS 2015-16 (or NFHS 2015-16) as well as figures from the STATCompiler show CPR among CMW as 53.5%. Am I missing something?
As per convention in many publications, I have removed all currently pregnant women (V213 ≠ 1) from the denominator and the corresponding CPR figure is 54.04%.
Am I on the right track? What am I missing here? Why am I not getting the figure of 53.5%?
|
|
|
|
| Re: Contraceptive Prevalence Rate [message #18241 is a reply to message #18119] |
Fri, 18 October 2019 13:35  |
Bridgette-DHS
Messages: 3038
Registered: February 2013
|
Senior Member |
|
|
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:
We apologize for the delay in answering your post. DHS bases the CPR on v313, rather than v361. The CPR may be restricted to modern methods, i.e. cases with v313=3. This is called the mCPR. The denominator should be women who have v502=1 or v528<=30 (women who are currently married or living as married OR had sex in the past 30 days). The cases must be weighted with v005/1000000. Did you weight the cases?
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
