The DHS Program User Forum: Weighting data » Inconsistencies in v022, v024 and Appendix A Tables

Home » Data » Weighting data » Inconsistencies in v022, v024 and Appendix A Tables (v022 is different from v024)

Show: Today's Messages :: Show Polls :: Message Navigator

Inconsistencies in v022, v024 and Appendix A Tables [message #27273]

Tue, 11 July 2023 11:06

Yawo
Messages: 45
Registered: February 2019

Member

Hello, I am in the process of approximating level-2 weights for a number of sub-Saharan African countries, and I am noticing discrepancies between v022, v024, and those areas listed in Appendix A of the reports. Here are some examples:

-- Chad2014: the number of strata reported by v022 (about 50) is more than that in the report (about 25). The areas in the report correspond with those regions reported under v024.

-- Congo2013: v022 lists 66, but the report does not correspond to the listing - fewer regions are listed in the report, and even then in percentages, (and not actual figures). V024 also does not correspond to the listing in the report.

-- Ethiopia2016: V022/ lists 25 strata, but the report only lists 11 regions. V024 corresponds to those listed in the report

How should handle these inconsistencies, use v024 for generating A_h.?
How about when it comes to svyset, use v024 or still stick to v022?

Thanks, cY

[Updated on: Wed, 12 July 2023 15:40]

Report message to a moderator

Re: Inconsistencies in v022, v024 and Appendix A Tables [message #27309 is a reply to message #27273]

Mon, 17 July 2023 07:37

Bridgette-DHS
Messages: 3208
Registered: February 2013

Senior Member

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

I assume you are working with the 2014-15 survey. I looked at the report, and see at the bottom of page 7 that the strata (as is normal) were constructed as the urban and rural parts of the regions. So v022 would be the correct identifier for the strata.

Report message to a moderator

Re: Inconsistencies in v022, v024 and Appendix A Tables [message #27324 is a reply to message #27309]

Tue, 18 July 2023 07:55

Yawo
Messages: 45
Registered: February 2019

Member

Thanks so much - I was able to resolve the issue for Chad. But I have run into a different issue with Congo 2013.

The values from Tables A.2 & A.3 (needed to calculate A_h and M_h) are reported on the following "Departments". These "departments" do not match the geographic regions or "Provinces" reported in the data (v023 and v024) (note: v022 itself contains does not have value labels, but I noted that each stratum is located in a "Province" reported in v024.

Here are the "Departments" used in the official report:

"Département
Kouilou"
Niari
Lékoumou
Bouenza
Pool
Plateaux
Cuvette
Cuvette-Ouest
Sangha
Likouala
Brazzaville
Pointe-Noire

And here are the labels for v024

 label list V024:
           1 kinshasa
           2 bandundu
           3 bas-congo
           4 equateur
           5 kasai-occidental
           6 kasai-oriental
           7 katanga
           8 maniema
           9 nord-kivu
          10 orientale
          11 sud-kivu

Please, how do I match these two regional values together, so I can use the values from the report to help me calculate A_h and M_h?

Thanks again in advance for your assistance

- Yawo

Report message to a moderator

Re: Inconsistencies in v022, v024 and Appendix A Tables [message #27325 is a reply to message #27324]

Tue, 18 July 2023 08:00

schoumaker
Messages: 66
Registered: May 2013
Location: Belgium

Senior Member

Hello,
I think you are mixing the two Congos.
The report you mention seem to refer to the Republic of Congo (Congo-Brazzaville), and the label for v024 refers to the Democratic republic of Congo (Congo-Kinshasa).
Best.
Bruno Schoumaker

Bruno Schoumaker
Centre for Demographic Research
Université catholique de Louvain

Report message to a moderator

Re: Inconsistencies in v022, v024 and Appendix A Tables [message #27328 is a reply to message #27325]

Tue, 18 July 2023 23:25

Yawo
Messages: 45
Registered: February 2019

Member

Bruno et al:

Thanks so much: you are right, I did mix the two. I have now downloaded the correct Democratic Republic of Congo dataset. Now, I need some assistance in deciphering the table containing the values of A_h and M_h needed for the approximation of level 1 & level 2 weights.

With the help of google translate, I was able to figure out that Table B.3 (columns 4 & 5 gives me the number of census sampled (A_h). But I am unable to figure out the values needed for calculating M_h (average number of households per cluster) for urban and rural areas.

I would appreciate some further assistance from you. The relevant Appendix (B) is attached.

Best, Yawo

Attachment: CongoD_2013_Appendix B.pdf
(Size: 401.84KB, Downloaded 134 times)

Report message to a moderator

Re: Inconsistencies in v022, v024 and Appendix A Tables [message #27333 is a reply to message #27328]

Wed, 19 July 2023 14:03

Bridgette-DHS
Messages: 3208
Registered: February 2013

Senior Member

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

You need to distinguish between the Republic of the Congo, or Congo, whose capital is Brazzaville, and the Democratic Republic of the Congo (DRC), whose capital is Kinshasa. DHS has done surveys in both countries.

If you are having trouble with the French labels, please look at Appendix A in an English language report on another DHS survey. The table formats are standardized across reports. You should be able to translate the labels this way.

[Updated on: Wed, 19 July 2023 14:11]

Report message to a moderator

Re: Inconsistencies in v022, v024 and Appendix A Tables [message #27335 is a reply to message #27333]

Wed, 19 July 2023 14:10

Yawo
Messages: 45
Registered: February 2019

Member

Yes, thanks so much. this is exactly what I have been doing - comparing table info for English reports with the French. it helped most of the time. /Yawo

Report message to a moderator

Previous Topic:	Replicating Step 7 and Level-2 Weight from Report 27
Next Topic:	Sex ratio at 2nd birth order

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Dec 11 17:51:05 Coordinated Universal Time 2024