| Zimbabwe 2015 Couples Men mv836 and mv503 [message #16684] |
Fri, 15 February 2019 16:53  |
DaniD
Messages: 13
Registered: November 2014
|
Member |
|
|
Hello!
First, thank you for this forum--it has been a lifesaver!
Second, I am wondering if someone can tell me if there is a protocol in terms of which sexual behavior variables are more reliable. I have found that for approx. 15% of the men in my sample of Zimbabwe couples (merged with HIV results) they report only one lifetime partner (mv836) but more than one lifetime union (mv503). It seems unlikely to me that men are coupling without having sex within those couples. Based on what I know about DHS questionnaires and sexual behavior research, it seems to make the most sense to trust the number of reported unions over the number of reported partners. Is this what DHS advises?
Thanks!
Dani
|
|
|
|
| Re: Zimbabwe 2015 Couples Men mv836 and mv503 [message #16798 is a reply to message #16684] |
Tue, 05 March 2019 10:46  |
Bridgette-DHS
Messages: 3208
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Specialist, Kerry LD MacQuarrie:
Thank you for your question. An error has been detected in the data, and we plan to have a new release of the recode data files online in the near future.
Specifically, there was an error in the recoding of the questionnaire question Q410 to variable mv503. The codes were reversed, so Q410 "more than once" was coded to "Once" in mv503 and vice versa. The exception to this is if the man had other partners (QM405/= "Yes", mv505 > 1), in which case mv503 was correctly coded as "More than once".
Whenever we release a new version of a dataset, an alert goes out to all registered users like yourself who previously downloaded data from that survey.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
