| Colombia 2010 [message #12428] |
Mon, 15 May 2017 16:27  |
Michelle_A
Messages: 2
Registered: May 2017
|
Member |
|
|
Hi there!
I have a question about the Colombia data from 2010. I am noticing that there is quite a bit of data missing:
While data was mostly available, as 14,521 responses were collated for which there was urban/rural data available and who responded either yes or no to taking alcoholic beverages during pregnancy, data was missing for other confounding variables. Respondent answers were largely missing for if they had previously terminated a pregnancy (missing 50,018 responses) and if they consumed alcohol during their pregnancy (missing 39,000 responses).
I'm wondering if respondents were never asked these latter two questions or if respondents refused to answer, or another reason? Your assistance is appreciated.
Best,
Michelle
|
|
|
|
| Re: Colombia 2010 [message #12431 is a reply to message #12428] |
Tue, 16 May 2017 12:57  |
Bridgette-DHS
Messages: 3208
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
In DHS surveys, a blank or "." (in Stata) means "not applicable"--that is, the question was not asked. If it was asked and refused, there would be a numeric code such as 9. In many DHS surveys, questions about such topics as drinking, domestic abuse, insurance, etc., are only asked in a subset of households. That's probably the reason why there are so many "not applicable" cases.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
