The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » General Data Questions » Flag variables
Flag variables [message #19075] Wed, 15 April 2020 05:56 Go to next message
nataliet is currently offline  nataliet
Messages: 1
Registered: April 2020
Member
Hi,

I am working with the Couple Recode from Angola 2015-2016 and I am not sure how should I work if there is some inconsistency among data. To be specific, what to do with flag variables for example - V227 Flag for last period (V226), even though I am not using variable V226, I would like to ask if I should keep the data if they have some flag or if I should work without them? I am afraid that the data won't me reliable if for example Respondent reported her last period was before her last birth, but she had never
given birth.

Could you help me with that please?

Natalie
Re: Flag variables [message #19223 is a reply to message #19075] Mon, 11 May 2020 22:32 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
If you are not using variable V226, you can ignore V227. There are some inconsistencies in the data, but that is the nature of survey data - respondents do not always give consistent answers - however in the aggregate the responses are generally consistent. Flag variables are provided just to aid the user if they are using the variables of interest, but should not be taken as measures of data quality.
Previous Topic: Problems with Peru continuous survey files (contain same information)
Next Topic: Women's Questionnaire Recode Map India
Goto Forum:
  


Current Time: Thu Mar 28 21:00:57 Coordinated Universal Time 2024