The DHS Program User Forum - RDF feed
https://userforum.dhsprogram.com/index.php
Re: IYCF - MAD indicator
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28847&th=13552#msg_28847
Any further advice on how to match the report numbers?
Thanks again for your time and assistance. ]]>geoK2024-03-18T20:00:44-00:00Re: IYCF - MAD indicator
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28846&th=13552#msg_28846
Following is a response from Senior DHS staff member, Tom Pullum:
I recommend that you try what you said--run the entire Main program even though you don't want all the indicators. It is quite possible that something is buried in the code for an earlier indicator that affects your results for MAD. Please let us know if you still have this discrepancy after trying that.
]]>Bridgette-DHS2024-03-18T19:42:45-00:00Re: Kenya 1998 Dataset Content Questions
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28845&th=13458#msg_28845
As I said, I do not use R, but in Stata, the sample design adjustments are specified with this statement (there are variations on it):
The problem you have is handled with the singleunit option. You just need to find what is the R equivalent of that Stata option.
]]>Bridgette-DHS2024-03-18T19:38:53-00:00IYCF - MAD indicator
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28829&th=13552#msg_28829
Specifically, I run the !NTmain.do code where I deactivate those commands for creating indicators I am not interested in (as i only want to get MAD). I therefore only use KEKR8BFL.dta file and the NT_IYCF.do file.
The discrepancy I observe seems to me to be due to the miscalculation of the denominator: I get almost double the cases (Number of all children age 6-23 months) and about 15% MAD.
Do I have to necessarely run all the do files included in the Chap11_NT/DHS8/ repository to get the right figures, or could the mismatch be due to something else?
Many thanks!]]>geoK2024-03-16T09:58:34-00:00Re: Kenya 1998 Dataset Content Questions
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28828&th=13458#msg_28828
Thank you for the response. I'd like to ask if you have any recommendations for this error message that appears in R when I try to run a chi square test between the 'excessive bleeding' variable and the 'urban', as I am trying to see if there is a significant association with excessive bleeding by urban-rural status.
The following error message is indicating that there's a problem with the survey design object 'mysurveydesign'. Specifically, it's indicating that one of the strata defined in the survey design has only one Primary Sampling Unit (PSU) at stage 1. However, my DHS dataset has multiple strata and PSUs.
> # chi square results for excessive bleeding (last birth)
> chisquareresult1 <- svychisq(~ excessbleed1 + urban, design = mysurveydesign)
Error in onestrat(`attr<-`(x[index, , drop = FALSE], "recentering", recentering), :
Stratum (2) has only one PSU at stage 1
]]>BevB2024-03-15T17:20:38-00:00Re: KDHS 2022: Table 2.17 Food security status
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28814&th=13339#msg_28814
The code is very helpful and thank you.
For this same table, I would be happy if you managed to get around how the coping strategy index was computed in Stata.
Thank you.
Damazo]]>dkadengye2024-03-13T09:54:28-00:00Re: KDHS 2022: Table 11.4 Breastfeeding status according to age
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28733&th=13341#msg_28733
Thanks for the informative answer. It all makes sense to me and I can see the challenge. Best wishes.]]>geoK2024-03-01T08:59:26-00:00Re: KDHS 2022: Table 11.4 Breastfeeding status according to age
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28731&th=13341#msg_28731
Following is a response from Senior DHS staff member, Tom Pullum:
It is a challenge to make comparisons when classifications of any kind have changed between surveys. For diet, it's not just that something like yogurt can move from one category to another. Sometimes a specific type of food is listed in one survey and not in another. You can add a footnote, saying that some small (we hope!) changes are due to changes in the classification.
DHS surveys have been conducted for almost 40 years, and there have been MANY changes in the questions and categories. On top of that, many countries have revised their geographic areas substantially. As I said, a challenge!
]]>Bridgette-DHS2024-02-29T21:45:51-00:00Re: KDHS 2022: Table 11.4 Breastfeeding status according to age
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28723&th=13341#msg_28723
I was able to match the respective figures for the two rounds using your revised program and the one on the GitHub for the first round. I guess my question is more about being able to compare the two time points despite small changes e.g. in the food list. In other words, whether despite the small changes in definition and program between the two rounds (e.g. that let's assume yogurt was considered liquid in round 1 but not in round 2), the % change between the two rounds would still be considered valid and meaningful.
I guess so, but I just wanted to make sure. I am looking at MAD right now, and It seems I am encountering a similar situation (but this is for a different post / topic). Thanks again.
]]>geoK2024-02-29T10:05:42-00:00Re: KDHS 2022: Table 11.4 Breastfeeding status according to age
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28718&th=13341#msg_28718
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
The definition of EBF and the construction of the code have been stable for a long time, but there have been some differences across surveys and changes over time in the specific questions. I believe you are looking at the program I revised to list the variables for types of liquids and foods in groups. You should look at the questionnaire and variable list to review what goes where, with the goal of matching the EBF percentage in the relevant table. Sometimes there are subtle distinctions, for example whether yogurt is a liquid. I hope that a careful reading of the relevant part of the final report will help too.