Extracting contraceptive discontinuation data in Calendar [message #28547] |
Mon, 29 January 2024 07:10 |
hirwa94
Messages: 1 Registered: July 2023
|
Member |
|
|
Hi,
I am analyzing contraceptive discontinuation using Rwanda DHS 2019. I extracted information related to contraceptive discontinuation using calendar but the resulting data indicates a mismatch with variable 359.
First, I filtered out Calendars of women who used any contraceptive in the 5 years that proceeded the survey. Then I recorded episodes of discontinuation for each. But the number of women who didn't discontinue a method don't match with Variable V359.
I attached the Python Notebook of the process.
|
|
|
Re: Extracting contraceptive discontinuation data in Calendar [message #28569 is a reply to message #28547] |
Wed, 31 January 2024 09:42 |
Janet-DHS
Messages: 893 Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member, Tom Pullum:
This question is too specific for DHS staff to be able to justify spending time on it. I can only propose a possible strategy for dealing with inconsistencies like this. What I would do is to reduce the file to fewer and fewer cases in an attempt to locate at least one specific case for which there is a mismatch.
It is conceivable that there is a problem in the data. When digging very deeply we occasionally find duplicates, illegal codes, blanks instead of dots for NA, etc. Another option is that there was a programming error by DHS in some recode or constructed variable. It's also possible that you (or defaults in Python) are misinterpreting how something should have been coded. I really hope you can figure it out--but it's too specific, as I said, for us to get involved.
|
|
|