The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » mege databases of multiple countries over several years (merging database of multiple countries over several years)
mege databases of multiple countries over several years [message #20483] Tue, 10 November 2020 10:12 Go to next message
Tedy is currently offline  Tedy
Messages: 4
Registered: July 2020
Member
Good evening dear DHS database experts. I am once again asking for your help in merging the databases. I am working on the effect of parental substance use on the health of their children under 5 years old. I am not familiar with stata, and i would like to merge the databases of multiple countries over several years for children under 5 years old containing information from their parents (father and/or mother), their household and the family structure (single or two-parent).
thank you in advance
sincerely
Re: mege databases of multiple countries over several years [message #20485 is a reply to message #20483] Tue, 10 November 2020 16:56 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2105
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

What variables do you intend to use to describe "parental substance use" and child health? Which surveys and countries include those variables? I suggest that before you think about combining multiple surveys and countries you look at one survey in one country. See how you would identify/describe a relationship in that survey. It may not be as simple as you think.


Re: mege databases of multiple countries over several years [message #20486 is a reply to message #20485] Wed, 11 November 2020 10:07 Go to previous messageGo to next message
Tedy is currently offline  Tedy
Messages: 4
Registered: July 2020
Member
Thank you for your response.
The variables we intend to use to describe "parental substance use" are: tobacco use and alcohol use, and we measure "child health" by the nutritional status of children under 5 years of age in terms of weight/age, height/age and weight/height.
In terms of country restriction, we will first use data from Cameroon.
We want to use the 2018, 2011, 2004 surveys from Cameroon.
Re: mege databases of multiple countries over several years [message #20487 is a reply to message #20486] Wed, 11 November 2020 10:53 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2105
Registered: February 2013
Senior Member

Following is another response from DHS Research & Data Analysis Director, Tom Pullum:

I suggest that you first select just one of these Cameroon surveys and try to identify a relationship between the predictors of interest (tobacco use, etc.) and the outcomes (stunting, underweight, wasted), including controls for other known predictors of the outcomes. You can do most of this with just the KR file, for children who are living with the mother (b9=0). Linking the child with the father in the MR file is not trivial. Stata code for stunting, etc. is available on the website (see the GitHub link). Good luck....


Previous Topic: Sibling Data Set
Next Topic: Child health and literacy
Goto Forum:
  


Current Time: Sat Jan 16 16:14:56 Coordinated Universal Time 2021