The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Creating members sleeping per room (Issues in recreating table)
Creating members sleeping per room [message #24637] Mon, 13 June 2022 01:55 Go to next message
Preksha Mishra is currently offline  Preksha Mishra
Messages: 5
Registered: June 2022

I am trying to recreate table Table 2.3 of NFHS-4(2015-16) India for number of members sleeping per room. I tried the code provided with the wealth index construction for the year (I ran a code in R based on the following:

if (hhusual=0) hhusual=hhslept.
if (QH45>0) memsleep=trunc(hhusual/QH45).
if (QH45=0) memsleep=hhusual.
if (missing(QH45) or QH45>=99 or memsleep>=98) memsleep=99.
variable labels memsleep 'Number of members per sleeping room'.
value labels memsleep 0 'Less than 1 per room'.
formats memsleep (f2.0).
missing values memsleep (99).

In this case my figures don't match the ones in Table 2.3. Most importantly, I did not get any missing values while the table suggests 0.1% missing values.
Next I tried, considering all household with 0 usual members, while keeping rest the same. Here, although the missing value matched, none of the other figures matched, though they were close.

Is it the case that methodology for recreating the table 2.3 is different from recreating the memsleep variable for PCA. If yes, could you please tell me how to do these matches.

Any help would be much appreciated
Re: Creating members sleeping per room [message #24722 is a reply to message #24637] Tue, 28 June 2022 10:18 Go to previous message
Janet-DHS is currently offline  Janet-DHS
Messages: 770
Registered: April 2022
Senior Member
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

Your "code" uses question numbers ("QH45") rather than variables in the PR or HR file. It's not clear what you are doing. The standard code on GitHub only looks at the number of rooms for sleeping and not persons per room, with a recode of hv216 into categories. To answer your question we would have to search the CSPro code for this specific survey and we don't have the staff time to do that.
Previous Topic: Extracting data from HH
Next Topic: Trend analysis with different number of 'PSU' and 'Strata'
Goto Forum:

Current Time: Sun Jul 14 10:28:46 Coordinated Universal Time 2024