Are variables comparable between years in the DHS? [message #28729] |
Thu, 29 February 2024 13:34 |
a.ortiz
Messages: 1 Registered: February 2024
|
Member |
|
|
Dear DHS community,
I am working with data from the DHS for Colombia for 2005 and 2010. I want to know how valid it is to assume that if a variable has the same name in the 2005 DHS and the 2010 DHS then it is the same variable. By being the same variable I mean that it has the same values and the values mean the same thing.
Does anyone know the answer to this?
Thanks for your help!
|
|
|
Re: Are variables comparable between years in the DHS? [message #28737 is a reply to message #28729] |
Fri, 01 March 2024 08:00 |
Bridgette-DHS
Messages: 3203 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
Most of the variables in two successive surveys in the same country are comparable. The questionnaire is given in an a appendix to the final report. You can check the wording and response categories for the questions/variables you are most interested in. You can tabulate the variable to see whether the category labels have changed. Some variables that are likely to change (not referring specifically to Colombia) are those dealing with occupation, diet, source of water, and sanitation. There can be changes in types of facilities, symptoms of illness, treatments, etc. It's good to be careful about potential changes, because when two surveys are pooled, the labels for the last survey will over-write all the others and will obscure any changes in definitions.
Survey-specific variables, which begin with "s" or "sh", can change completely.
|
|
|