The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » India » Construction of level weights from individual weights
Construction of level weights from individual weights [message #22872] Thu, 27 May 2021 06:28 Go to next message
Indiana is currently offline  Indiana
Messages: 4
Registered: May 2021
Member
Hello DHS team,

I am trying to approximate cluster-level weights from individual weights for 2015-16 Indian DHS (NFHS-4). While replicating the code in "Multilevel Modelling Using DHS Surveys...:Methodological Report 27" for India, I could not find the distribution of EAs and their average size in number of households by strata in the NFHS-4 final report.

I would be very grateful if anyone could help me in this regard.

Regards,

Report message to a moderator

Re: Construction of level weights from individual weights [message #22875 is a reply to message #22872] Thu, 27 May 2021 10:15 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member

Following is a response from DHS Senior Sampling Specialist, Mahmoud Elkasabi:

It is true that the final report of NFHS-4 only includes a table for number of HHs by states but does not include the table for number of clusters in the census frame. However, you can obtain that information from the census website. This page include a list of blocks with their sizes for each state: https://censusindia.gov.in/pca/cdb_pca_census/cd_block.html .

Report message to a moderator

Re: Construction of level weights from individual weights [message #22879 is a reply to message #22875] Thu, 27 May 2021 15:58 Go to previous messageGo to next message
Indiana is currently offline  Indiana
Messages: 4
Registered: May 2021
Member
Thank you for your response, sir.

Following your suggestion, I have to compile the required information from the excel file of 640 districts comprising data of 597608 villages and 7933 towns in India. Wouldn't it be too tedious for a single person?
I request you to please consider my problem and enlighten me with some alternative.

Regards,

Report message to a moderator

Re: Construction of level weights from individual weights [message #22909 is a reply to message #22879] Wed, 02 June 2021 11:09 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member

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

We have determined that the Excel files available from that censusindia.gov.in link are not sufficient for estimating the cluster weights in the NFHS-4. They don't have sufficient detail about the urban areas. We are trying to get more information about the sampling frame from the 2011 census and will let users know if we are successful.

Report message to a moderator

Re: Construction of level weights from individual weights [message #22916 is a reply to message #22909] Thu, 03 June 2021 05:04 Go to previous messageGo to next message
Indiana is currently offline  Indiana
Messages: 4
Registered: May 2021
Member
Thank you for your consideration, sir. Meanwhile could you please suggest if it's possible to conduct a multilevel analysis without calculating cluster level weights?

Report message to a moderator

Re: Construction of level weights from individual weights [message #22926 is a reply to message #22916] Thu, 03 June 2021 14:31 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member

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

We expect to have the necessary census data by the end of next week. We will then make it available to you and other users.

Report message to a moderator

Re: Construction of level weights from individual weights [message #23205 is a reply to message #22926] Fri, 30 July 2021 03:00 Go to previous messageGo to next message
Indiana is currently offline  Indiana
Messages: 4
Registered: May 2021
Member
Sir please make the required data available, or suggest an alternative way to conduct multilevel analysis.

Report message to a moderator

Re: Construction of level weights from individual weights [message #23962 is a reply to message #23205] Fri, 21 January 2022 11:54 Go to previous messageGo to next message
lmcdougal
Messages: 17
Registered: March 2013
Member
Hello - wondering if there is any update available on the necessary information to create multilevel weights for the 2015-16 India DHS using 2011 census data? Thanks in advance!

Report message to a moderator

Re: Construction of level weights from individual weights [message #23990 is a reply to message #23962] Tue, 25 January 2022 14:29 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member
We have just uploaded a folder here: https://github.com/DHSProgram/DHS-Analysis-Code/tree/main/Mu ltilevel_Weights/India.

Report message to a moderator

Re: Construction of level weights from individual weights [message #24518 is a reply to message #23990] Wed, 25 May 2022 12:04 Go to previous messageGo to next message
lmcdougal
Messages: 17
Registered: March 2013
Member
Thank you, Bridget! This was very helpful for the 2015-16 India DHS. The recently released 2019-21 India DHS has more districts included - are you able to provide guidance on the calculation of multilevel weights for this most recent India data?

Report message to a moderator

Re: Construction of level weights from individual weights [message #24528 is a reply to message #24518] Thu, 26 May 2022 16:42 Go to previous messageGo to next message
Janet-DHS is currently offline  Janet-DHS
Messages: 888
Registered: April 2022
Senior Member
Following is a response from DHS Senior Sampling Specialist, Mahmoud Elkasabi:

You can use the same materials/steps we provided for NFHS-4. Note that the link we provided earlier does not work anymore; here is the current link https:// github.com/DHSProgram/DHS-Analysis-Code/tree/main/Multilevel _Weights/India

Report message to a moderator

Re: Construction of level weights from individual weights [message #24938 is a reply to message #24528] Thu, 04 August 2022 12:41 Go to previous messageGo to next message
lmcdougal
Messages: 17
Registered: March 2013
Member
Unfortunately, the NFHS-4 guidance does not directly correspond with NFHS-5; NFHS-5 was collected using 707 districts, as compared to the 640 in NFHS-4, thus including more districts than were used in the 2011 India Census. The multilevel weight calculations require the total number of census clusters in a given strata, the average number of households per cluster by sampling strata, and the total number of households according to the census frame. The NFHS-5 (India 2019-21) final report on the DHS website right now doesn't seem to contain appendices, so I can't tell if any of this information is provided therein.

Do you have any recommendations for how to get this information and calculate these weights for NFHS-5?

Report message to a moderator

Re: Construction of level weights from individual weights [message #24939 is a reply to message #24938] Thu, 04 August 2022 13:00 Go to previous messageGo to next message
fred.arnold@icf.com is currently offline  fred.arnold@icf.com
Messages: 84
Registered: May 2021
Senior Member
The appendices are included in Volume 2 of the final report. Volume 2 has been drafted and it should be publicly available in the next couple of months. Volume 2 includes an appendix on the sample design and another appendix on estimates of sampling errors. NFHS-4 covered all of the 640 districts in the country at the time of the 2011 Census. NFHS-5 covered all of the 707 districts in the country as of March 31, 2017.

Report message to a moderator

Re: Construction of level weights from individual weights [message #26424 is a reply to message #24518] Mon, 20 March 2023 08:52 Go to previous messageGo to next message
Mukesh is currently offline  Mukesh
Messages: 13
Registered: March 2019
Member
Daer Imcdougal,

I tried to follow the codes given for "rural area" but unable to get the outcomes. Can you please help me.

Report message to a moderator

Re: Construction of level weights from individual weights [message #26453 is a reply to message #24528] Wed, 22 March 2023 08:52 Go to previous messageGo to next message
Mukesh is currently offline  Mukesh
Messages: 13
Registered: March 2019
Member
Dear DHS team,

I tried to replicate the results given for "URBAN AREA EB" in the x.xlsx file on the DHS GitHub page.
That do match with "Total Number of Household" and how to find EA for urban wards?

Thanks

Report message to a moderator

Re: Construction of level weights from individual weights [message #26456 is a reply to message #26453] Wed, 22 March 2023 10:07 Go to previous messageGo to next message
Mukesh is currently offline  Mukesh
Messages: 13
Registered: March 2019
Member
I am referring to NFHS-4 (2015-16) (India & 2015-16).

I am trying to replicate the urban area file given here in Excel format: https://github.com/DHSProgram/DHS-Analysis-Code/tree/main/Mu ltilevel_Weights/India
I found discrepancies in the "total number of households" in the "urban area" given in the file on "dhs_github" and the original file from Census of India-- link is given below.

I accessed the census_pca file here-- https://censusindia.gov.in/census.website/data/census-tables # than searched "pca block wise"
then I filtered this file for the urban area for many districts, but "the total number of household" do not match with dhs_excel file given on first link


My query is how to:
1. How to replicate the "EA average size_Urban.xlsx" given on github page of DHS.
2. How to find EA in urban area
3. Finally, how to find the size of the rural clusters

Hope this clarifies.

Report message to a moderator

Re: Construction of level weights from individual weights [message #26480 is a reply to message #26456] Fri, 24 March 2023 07:44 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member
We are looking into this problem and will try to provide what you need within the next few days.

Report message to a moderator

Re: Construction of level weights from individual weights [message #26609 is a reply to message #26480] Sat, 08 April 2023 06:08 Go to previous messageGo to next message
Mukesh is currently offline  Mukesh
Messages: 13
Registered: March 2019
Member
Dear DHS team any update in this regard?

Report message to a moderator

Re: Construction of level weights from individual weights [message #29425 is a reply to message #26609] Mon, 17 June 2024 07:26 Go to previous messageGo to next message
HannahG is currently offline  HannahG
Messages: 1
Registered: June 2024
Member
Hello,

I am wondering if there is an update re: calculating cluster level weights for NFHS-5? The 'EA average size_Urban' excel file uploaded to the Github repository for the NFHS-4 cannot be used for NFHS-5 as the number of districts are different. Volume 2 of the NFHS-5 final report has tables on sample design and error but doesn't contain the information for individual strata as required by the MR27 method and usually included in Final Reports. Can you advise on a method to compute the total number of census clusters in a given strata, the average number of households per cluster by sampling strata, and the total number of households according to the census frame?

Best,
Hannah

Report message to a moderator

Re: Construction of level weights from individual weights [message #29430 is a reply to message #29425] Mon, 17 June 2024 08:52 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member

Following is a response from Senior DHS staff member, Tom Pullum:

There have been other recent posts on calculating multilevel weights for the NFHS-5. DHS does not have the information required to calculate such weights, we do not know where to find the information (if it still exists), and we have no resources to put into this.

As I have said before, if anyone out there has been able to calculate the multilevel weights for the NFHS-5, please let us know and we will help make them available to other users.


Report message to a moderator

Previous Topic: How to extract the survey data
Next Topic: Regarding skip pattern in occupation questions (IR) file of DHS 5 India
Goto Forum:
  

[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Sun Nov 24 03:15:49 Coordinated Universal Time 2024
.:: Contact :: Home :: Privacy Policy ::.

Powered by: FUDforum 3.0.9.
Copyright ©2001-2018 FUDforum Bulletin Board Software

The information provided on this Web site is not official U.S. Government information and
does not represent the views or positions of the U.S. Agency for International Development or the U.S. Government...