The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Merging data files » Merging datasets for different countries on SPSS
Merging datasets for different countries on SPSS [message #17446] Tue, 19 March 2019 20:40 Go to next message
nicolegray1 is currently offline  nicolegray1
Messages: 1
Registered: February 2019
Member
I'm analysing secondary sex ratio trends in 14 different African countries using DHS data and chi-square tests of association for 8 variables for each . However the results are very inconsistent and this may be because the sample size is too small for individual countries; it must supposedly be at least 1 million births. Therefore, as I've seen other studies do, I wanted to combine the data for the 14 countries and rerun my chi-square tests on the same variables, to see what associations hold when the data is aggregated. Does anyone know how I can merge these different datasets so that they are in one dataset- copy and pasting the data just crashes my laptop as the files are too large.

Thanks in advance.
Re: Merging datasets for different countries on SPSS [message #17479 is a reply to message #17446] Thu, 28 March 2019 11:29 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1671
Registered: February 2013
Senior Member

Following is a response from Senior DHS Stata Specialist, Tom Pullum:

You are talking about appending data files rather than merging. There have been many forum postings on this topic, mostly using Stata rather than SPSS.

Why do you think you need "at least 1 million births"? You could definitely analyze the sex ratio at birth (I think "at birth" is what you are talking about) using the BR files for specific countries. It is NOT necessary to pool or append the files. The sex of the child is given as b4, with 1=boy and 2=girl. You could, for example, recode this into a 0/1 variable and use logit regression. I suggest that you start with a search of what other researchers may have done with the sex ratio and DHS data.
Re: Merging datasets for different countries on SPSS [message #17581 is a reply to message #17446] Fri, 19 April 2019 16:55 Go to previous message
boyle014 is currently offline  boyle014
Messages: 70
Registered: December 2015
Location: Minneapolis
Senior Member
Dr. Pullman is correct that it may not be necessary to append all these files to accomplish your goals. However, if you do decide that you want to append many files, while retaining just a few variables, you should think about using IPUMS DHS. IPUMS DHS allows you to select the samples and variables that interest you, and download a single, fully harmonized and integrated data file. Here is the online user guide to help you get started. Since you are using SPSS, be sure to change the data format to SPSS when you are checking out.

Professor Elizabeth Boyle
Sociology & Law, University of Minnesota, USA
Principal Investigator, IPUMS-DHS
Previous Topic: Merging IR and PR?
Next Topic: Merging Malawi MIS 2017 Datasets
Goto Forum:
  


Current Time: Mon Oct 14 00:16:39 Eastern Daylight Time 2019