The DHS Program User Forum - RDF feed
https://userforum.dhsprogram.com/index.php
NFHS -5, Table 12.10
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=25075&th=11963#msg_25075
I need help in the identifying the data file and the variables for the Table 12.10 in the NFHS-5 (India).
Please help
Regards
Lincoln ]]>Lincoln2022-08-28T07:58:52-00:00Re: NFHS -5, Table 12.10
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=25090&th=11963#msg_25090
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:
Table 12.10 is based on a household-level question about whether there had been any deaths in the household in the past two years, and if so, the number of deaths. The variable is sh88 in the HR file (IAHR7BFL.dta). ]]>Bridgette-DHS2022-08-30T15:18:25-00:00Re: NFHS -5, Table 12.10
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=25092&th=11963#msg_25092
So does it mean that for this indicator the denominator = number of households+sh88 and the numerator= SH88
Regards
Lincoln ]]>Lincoln2022-08-30T15:25:24-00:00Re: NFHS -5, Table 12.10
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=25097&th=11963#msg_25097
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:
NFHS-5 Table 12.10 is not a standard table. There is a matching table in the NFHS-4 report--NFHS-4 and NFHS-5 had almost identical questionnaires and tabulation plans--but I have not seen this table in other reports.
The tables in these reports were produced in CSPro. Stata versions were not prepared. The CSPro code is hard to find and hard to work with. The analysis team at DHS had nothing whatsoever to do with the construction of indicators or the preparation of the report.
With forum questions such as this one, we just try to "reverse engineer" the construction of survey-specific indicators. The Guide to DHS Statistics is an important resource but it does not describe this indicator.
If P is the household population at the time of the survey, D is the number of deaths in the past two years, and B is the number of births in the past two years, I would use P-(B/2)+(D/2) as the denominator and D as the numerator. This estimate would just refer to "the two years before the survey", where the survey date varies from one household to another, rather to any calendar interval of time. It does not take into account that households are social constructs and may divide, combine, etc., during the two-year interval, particularly following events such as deaths (marriages, migration, etc.).
You can try to match the indicator. If you don't match it, you may actually have a better estimate than the one in the table. But who knows?
]]>Bridgette-DHS2022-08-31T11:49:18-00:00