The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Unmet Need » Response code as "In-consistence" be valuable to include in analysis ? (How to manage the responses coded as _"In-consistence" during the analysis of association between FP unmet need and its predictor variables)
Response code as "In-consistence" be valuable to include in analysis ? [message #18019] Sun, 18 August 2019 05:11 Go to next message
drseinhlaing is currently offline  drseinhlaing
Messages: 7
Registered: August 2019
Member
Dear DHS specialist

I am analyzing FP unmet need and it predictors in Myanmar DHS 2015-16. In Myanmar DHS data, for example : v531 : Age at first Sex (Imputed) , there is responses coded as "In-consistence". I would like to ask , what should be the suitable practice of managing this category of responses - "In-consistence" while analyzing association between v531 ( predictor variable) and FP-Unmet need ( out come variable) ?
Similarly, another example of V729, Husband / partner Education attainment , there is responses coded as "Don't Know". what should be the suitable practice of managing this category of responses - "Don't know" while analyzing association between predictor variable and FP-Unmet need ( out come variable) ?

Could I re-coded those responses "In-consistence", "Don't Know" as missing data and treated as missing data? or any other recommended practices to manage those responses recoded as "In-consistence" / "Don't Know".

Thanks for your kind guidance
With regards
Sein Hlaing

[Updated on: Sun, 18 August 2019 05:17]

Report message to a moderator

Re: Response code as "In-consistence" be valuable to include in analysis ? [message #18021 is a reply to message #18019] Sun, 18 August 2019 22:56 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 671
Registered: January 2013
Senior Member
Yes, it is generally best to treat the "inconsistent" and "Don't know" responses as missing data.
Re: Response code as "In-consistence" be valuable to include in analysis ? [message #18022 is a reply to message #18021] Mon, 19 August 2019 13:14 Go to previous message
drseinhlaing is currently offline  drseinhlaing
Messages: 7
Registered: August 2019
Member
Dear Trevor
Thanks so much for your guidance
Regards
Sein
Previous Topic: Selecting variable and categorization of unmet need
Next Topic: matching unmet need in surveys to stata code
Goto Forum:
  


Current Time: Wed Nov 20 20:51:57 Eastern Standard Time 2019