merge women nutrition status to child file


I am using children file as the base file and I realize that women nutrition variables are not included in the children file. How can I merge this or access this? Baby step answers will be great. I am a novice but eager to learn.



In July, IPUMS-DHS did a data release that added the biometric variables for women from the IR (individual recode) files for women of childbearing age. We have linked the mother’s characteristics to the child (KR file) data, so you should now be able to access the mother’s (women’s) nutritional data (in terms of measures of height, weight, and anemia) if you make a data extract using children as the unit of analysis.

Here are the list of variables under

Female anthropometry variables (for mother) when children are the unit of analysis:…

and blood testing for anemia, for the child’s mother…

With these variables now available in IPUMS-DHS, you don’t have to do any merging, just choose children as your unit of analysis and select variables from the Biometrics (for mother) group in the drop down menu.

If you meant something other than this when you referred to the “women nutrition variables,” please repost with more information, such as some examples of what you want.


Thank you for your response. I was looking more specifically for the nutritional questions (1) ate fruits in the last 7 days and (2) ate veg in the last 7 days. Found them only available in Ghana 2014 children files. I am looking at Ghana, uganda, Malawi, Tanzania, Rwanda, and Zimbabwe all in Phase VII surveys. I need the variable to establish any association of nutrition intake and outcome anaemia.

Many thanks.



Thank you for drawing my attention to these variables not yet in IPUMS-DHS, Eunice. I can make sure they are included in our next data release, which I expect will happen around October.

If you need the material sooner than that, you could make an IPUMS-DHS data extract and link to the original DHS files for samples of interest to pick up the additional variables. You might even drop all the variables from the original DHS files except the variables you need for linking and the nutritional intake variables you need before doing the linkage, if you find the large original DHS files awkward to work with.

Here are guidelines on linking between your IPUMS-DHS file and the original DHS file(s):…

As we continue to add variables as well as samples to IPUMS-DHS, it will become less necessary to use the original DHS files as a supplement. Thank you for your patience in the interim.


Thank you very much I will do as suggested and merge only for the needed variables.

Thanks for your time and prompt response.



Hi, Eunice,

I did a quick search through the PDFs of women’s questionnaire forms for 11 Phase VII surveys, including all that you mentioned. Unfortunately, the only surveys that seemed to include your questions of interest, about number of days the woman ate fruit or vegetables in the past 7 days, seem to be those for Ghana 2014, Namibia 2013, and a somewhat differently worded but related question for India 2015. Both the Ghana and Namibia surveys gave these variables country-specific names (specifically, for the vegetable consumption question, S1036 for Ghana 2014 and S1010f for Namibia 2013). The variables are also in Ghana 2008 (from Phase 6, variable name S1021 for vegetable consumption).

Have you looked through the women’s questionnaires for the questions that interest you? I recommend doing so before you try doing any merging between your IPUMS-DHS file and the original DHS file(s).

Please let me know if I am overlooking something. Frequently a few countries try out a question before it becomes part of the standard questionnaire, and that may be the case for the questions/variables that interest you.

You can review (and search) questionnaire text in PDF format (or from transciptions into Word documents) from the following part of the IPUMS-DHS website:…


Thank you very much for doing that leg work for me. Definitely saves me time, and you are missing anything. I will go back to my drawing table and revisit my variables again.

Many thanks for your time.