The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Other countries » Zambia DHS 2018
Zambia DHS 2018 [message #21754] Tue, 15 December 2020 03:25 Go to next message
Zoe is currently offline  Zoe
Messages: 1
Registered: December 2020
Member
I downloaded the 2018 ZDHS data set. The data for some variables are missing. As a result I can not proceed with the analysis for risky sexual behavior. Among these are indicators for risky sexual activity such as condom use and multiple partners. Looking at the questionnaire they were covered during the survey.
Can anyone kindly assist in this regard.
Re: Zambia DHS 2018 [message #21759 is a reply to message #21754] Tue, 15 December 2020 08:11 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member
A good resource for researching a question/standard variable is the Standard Recode Manual:
https://www.dhsprogram.com/pubs/pdf/DHSG4/Recode7_DHS_10Sep2 018_DHSG4.pdf

You can also visit the IPUMS-DHS site, which allows users to do a web-based search on a string or variable. Variable names in IPUMS are links that provide extensive information, including variable description, cautionary notes, the survey text used to generate each variable, including its codes and frequencies, variable meaning and universe (who was asked the question), and precise question wording.. The variables are also harmonized across samples, making this an excellent tool for comparative analysis. Easily log in with your DHS user name email and password.

You should also check the questionnaires to see if the information you are looking for was collected in the survey. Lastly, you can search the codebook (the file with the .map extension) for any string in the question you are looking for.
Re: Zambia DHS 2018 [message #24136 is a reply to message #21759] Thu, 03 March 2022 08:18 Go to previous messageGo to next message
Svenja is currently offline  Svenja
Messages: 3
Registered: March 2022
Member
Hello!

I am facing a similar issue. I am conducting a study about the association between intimate partner violence and small child nutrition quality. However, several variables show no observations. Most importantly, there are no observations for the variables m35 and m36 or v407 and v408 which all refer to the number of times a child was breastfed. But there are other variables that show the same issue: d105g (ever physical violence by partner), d110e-h (experienced any of listed partner's actions), m55a-o,x,z (first 3 days given [food]).
I have tried all datasets within ZDHS2018 but could not find any observations. I have consulted the questionnaire and recode manual and both show that there should be data. Furthermore, the full report on the ZDHS2018 data reports about minimum acceptable diet which needs a 24 hour recall including information about breastfeeding behaviour.

I would appreciate information on how to access observations on these variables.

Thank you!
Re: Zambia DHS 2018 [message #24146 is a reply to message #24136] Thu, 03 March 2022 14:13 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

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

We do not have a place where we keep data that does not appear in the standard recode files. ALL the data we have is/are in those files. A blank or a dot for a variable indicates that the variable is NA or not applicable. The only way to find why a variable was NA for specific cases is to identify the questions in the questionnaire that correspond with that variable, and look for skips or filters prior to that question.
Re: Zambia DHS 2018 [message #24159 is a reply to message #24146] Mon, 07 March 2022 06:39 Go to previous messageGo to next message
Svenja is currently offline  Svenja
Messages: 3
Registered: March 2022
Member
Hello,

Thank you for the swift reply!
There is a filter to the variable but that filter suggests that the variable applies to over 3/4 of respondents. As I said, the ZDHS2018 report does report on the variable so clearly there must be data. I use Stata but have now also tried SPSS but there are no observations there either. Do you have any ideas as to what I can try/do to get observations?

Thank you!
Re: Zambia DHS 2018 [message #24161 is a reply to message #24159] Mon, 07 March 2022 08:50 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member
Following is another response from DHS Research & Data Analysis Director, Tom Pullum:

I will paste below what I get when I look for those variables in ZMIR71FL.dta. "na" in the variable label means that the variable was not included in the survey. I repeat that we do not have such data hidden away somewhere. The relevant questions were simply not asked. I know this can be frustrating when making international comparisons. Countries, not DHS staff, make the final decision about what to include in a survey.

/index.php?t=getfile&id=1806&private=0
  • Attachment: summarize.png
    (Size: 34.90KB, Downloaded 291 times)
Re: Zambia DHS 2018 [message #24162 is a reply to message #24161] Mon, 07 March 2022 09:27 Go to previous message
Svenja is currently offline  Svenja
Messages: 3
Registered: March 2022
Member
Oh I see, okay. I'm sorry for causing frustration. I was just confused because I think they reported it in their report but then I will find another solution.
Thank you very much for your assistance!
Previous Topic: Colombia municipalities
Next Topic: Ethiopia Injury Module
Goto Forum:
  


Current Time: Fri Mar 29 07:14:49 Coordinated Universal Time 2024