The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Other countries » Namibia 2013 domestic violence module for males
Namibia 2013 domestic violence module for males [message #22990] Sun, 20 June 2021 15:08 Go to next message
Anne is currently offline  Anne
Messages: 7
Registered: June 2021
Member
Hi,

Am conducting analysis on domestic violence among males in African countries.
However, I get different totals for Namibia depending on the weight used, which is not the case with other countries.

I have also used svy: tab which gives the percent but just double checking on the totals.

For example, using the variable dm111 for experienced any of the listed wife/partner's actions
The totals are 1105 for the unweighted and manual weighted totals using mv005, and 1002 for the manual weight using dm005.

What is causing this for Namibia? Based on previous posts, I suppose I need to use the dm005 as the weight. Is it something I should be worried about?

tab dm111

experienced |
any of |
listed |
wife/partne |
r's actions |
(dm110x |
series)*** | Freq. Percent Cum.
------------+-----------------------------------
no | 1,077 97.47 97.47
yes | 28 2.53 100.00
------------+-----------------------------------
Total | 1,105 100.00

. tab dm111 [aweight=dm005]

experienced |
any of |
listed |
wife/partne |
r's actions |
(dm110x |
series)*** | Freq. Percent Cum.
------------+-----------------------------------
no | 976.107237 97.42 97.42
yes | 25.892763 2.58 100.00
------------+-----------------------------------
Total | 1,002 100.00

. tab dm111 [aweight=mv005]

experienced |
any of |
listed |
wife/partne |
r's actions |
(dm110x |
series)*** | Freq. Percent Cum.
------------+-----------------------------------
no | 1,075.0611 97.29 97.29
yes | 29.9389073 2.71 100.00
------------+-----------------------------------
Total | 1,105 100.00


Re: Namibia 2013 domestic violence module for males [message #22996 is a reply to message #22990] Mon, 21 June 2021 09:35 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 Senior DHS Specialist, Kerry MacQuarrie:

dm005 is the weight variable you want to use.
Re: Namibia 2013 domestic violence module for males [message #22997 is a reply to message #22996] Mon, 21 June 2021 09:44 Go to previous messageGo to next message
Anne is currently offline  Anne
Messages: 7
Registered: June 2021
Member
Thank you. My question was why am I getting different totals only for Namibia. I get different totals for any variable I use for Namibia once I use the dm005 as the weight variable which is not the case with the rest of the countries. Is there anything am doing wrong?
Re: Namibia 2013 domestic violence module for males [message #22998 is a reply to message #22997] Mon, 21 June 2021 10:56 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 will look into the construction of dm005 in the Namibia 2013 survey and will let you know if we find that it was different from other surveys.

[Updated on: Fri, 30 July 2021 11:14]

Report message to a moderator

Re: Namibia 2013 domestic violence module for males [message #23220 is a reply to message #22998] Mon, 02 August 2021 08:23 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:

Very sorry for the delay with this response. We have not found anything unusual about the way dm005 was prepared for this survey. If you are still having a problem with it, please be more specific about the inconsistency, for example, referring to a specific table in the survey report.
Re: Namibia 2013 domestic violence module for males [message #23275 is a reply to message #23220] Mon, 16 August 2021 05:58 Go to previous messageGo to next message
Anne is currently offline  Anne
Messages: 7
Registered: June 2021
Member
Thank you for the reply.

For example, using the variable dm104 for experienced any emotional violence in the case of Namibia. The tabulation for the total is 1100. When we use the weight for men mv005, the total remains the same 1100. But when we use the domestic violence weight, the total reduces to 997.

If I use another country, for example Uganda. The total remains the same regardless of whether the weight is applied or not, unlike in Namibia where the total changes when the domestic violence weight is applied. Using the same dm104 variable in the case of Uganda, when tabulated, the total is 2858, when either the weight for men or the domestic violence is used, the total remains the same at 2858. I have experimented this on other countries, like Burundi, Cameroon, Kenya, Mozambique, Rwanda and Sierra Leone, and the total remains the same regardless of whether no weight, the men's weight or the domestic violence weight is used. But for Namibia, the total changes when a tabulation is done and the domestic violence weight is used. This has only been done for the males. Thus, I was wondering what was causing the inconsistency for Namibia or if I missed something.
The Stata output tables are shown below, for dm104 first for Namibia then Uganda

Tabulation for Namibia
tab dm104

experienced |
any |
emotional |
violence |
(dm103a-c |
series) | Freq. Percent Cum.
------------+-----------------------------------
no | 968 88.00 88.00
yes | 132 12.00 100.00
------------+-----------------------------------
Total | 1,100 100.00

. tab dm104 [aweight = mv005]

experienced |
any |
emotional |
violence |
(dm103a-c |
series) | Freq. Percent Cum.
------------+-----------------------------------
no | 957.920123 87.08 87.08
yes | 142.079877 12.92 100.00
------------+-----------------------------------
Total | 1,100 100.00

. tab dm104 [aweight = dm005]

experienced |
any |
emotional |
violence |
(dm103a-c |
series) | Freq. Percent Cum.
------------+-----------------------------------
no | 876.568036 87.92 87.92
yes | 120.431964 12.08 100.00
------------+-----------------------------------
Total | 997 100.00


Tabulation for Uganda
tab dm104

experienced |
any |
emotional |
violence |
(d103x |
series)*** | Freq. Percent Cum.
------------+-----------------------------------
no | 1,805 63.16 63.16
yes | 1,053 36.84 100.00
------------+-----------------------------------
Total | 2,858 100.00

.
. tab dm104 [aweight = mv005]

experienced |
any |
emotional |
violence |
(d103x |
series)*** | Freq. Percent Cum.
------------+-----------------------------------
no | 1,836.6087 64.26 64.26
yes | 1,021.3913 35.74 100.00
------------+-----------------------------------
Total | 2,858 100.00

.
. tab dm104 [aweight = dm005]

experienced |
any |
emotional |
violence |
(d103x |
series)*** | Freq. Percent Cum.
------------+-----------------------------------
no | 1,831.5457 64.08 64.08
yes | 1,026.4543 35.92 100.00
------------+-----------------------------------
Total | 2,858 100.00

.
Re: Namibia 2013 domestic violence module for males [message #23282 is a reply to message #23275] Mon, 16 August 2021 09:58 Go to previous messageGo to next message
Shireen-DHS is currently offline  Shireen-DHS
Messages: 140
Registered: August 2020
Location: USA
Senior Member
Hello,

Different surveys would have different sample selection for the domestic violence module. Please check the introduction chapter for the survey for the survey implementation section. For analyzing this data you use the domestic violence weight divided by 1000000 and use iweight for tabulations and pweight for svy and regressions, etc.

For instance:

gen dwt= dm005/1000000
tab dm104 [iw=dwt]

Best,
Shireen Assaf
The DHS Program
Re: Namibia 2013 domestic violence module for males [message #23285 is a reply to message #23282] Mon, 16 August 2021 11:32 Go to previous message
Anne is currently offline  Anne
Messages: 7
Registered: June 2021
Member
Okay, thank you for the clarification.
Previous Topic: Philippines Multilevel analysis
Next Topic: Indonesia breastfeeding variables (m4 m5) 2012
Goto Forum:
  


Current Time: Thu Nov 28 17:37:04 Coordinated Universal Time 2024