The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Domestic Violence » Peru domestic violence data
Peru domestic violence data [message #9135] Mon, 15 February 2016 10:24 Go to next message
bendaudv
Messages: 11
Registered: February 2016
Member
Hello,

I am trying to calculate prevalence of recent IPV from the Peru 2012 DHS dataset. The domestic violence weight variable seems to be empty (Stata returns 0 observations for the command "summarize d005"). Is there perhaps another weight variable that should be used for domestic violence data for this dataset? Thank you very much in advance for your guidance.
Re: Peru domestic violence data [message #9140 is a reply to message #9135] Mon, 15 February 2016 13:27 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
A response from Dr. Shea Rutstein:
Quote:

There is not much difference between the household, individual and domestic violence weights, since there is almost a one to one correspondence of respondents to households (all women 15-49 for individual questions and one woman per household for domestic violence questions). The difference lies in the response rates, which given that they are more than 90% in Peru and do not differ much by area, do not affect the results. I would just use the individual woman's weight, v005 for the domestic violence questions.
Shea

Re: Peru domestic violence data [message #9146 is a reply to message #9140] Tue, 16 February 2016 04:07 Go to previous messageGo to next message
bendaudv
Messages: 11
Registered: February 2016
Member
Thank you very much for this response and guidance.

I have tried using the individual women's weight, but am getting different results for prevalence of recent IPV than what is available from the API extract for this indicator (10.3% with a denominator of 16,049 compared to 12.9% with a weighted denominator of 12,842 in the API extract). Would it be possible to replicate this result from the dataset? Thank you very much.
Re: Peru domestic violence data [message #9249 is a reply to message #9146] Fri, 26 February 2016 16:52 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
A response from Dr. Shea Rutstein:
Quote:

In spite of not knowing what API and IPV ( can you please clarify) are, the number of cases for the numerator are the same (about 1655 or so), given the percentages below. Therefore the difference is in the denominator only and is not the result of differential weighting. Could it be the difference between currently married and ever married women?

[Updated on: Fri, 26 February 2016 16:58]

Report message to a moderator

Re: Peru domestic violence data [message #9268 is a reply to message #9249] Tue, 01 March 2016 06:07 Go to previous messageGo to next message
bendaudv
Messages: 11
Registered: February 2016
Member
Thank you very much for this response and suggestion, and apologies for the use of acronyms. By recent IPV (intimate partner violence) I was referring to the percentage of ever-partnered women 15-49 who reported experiencing physical or sexual violence by an intimate partner in the last 12 months. The API referred to the DHS data query site http://api.dhsprogram.com.

I have checked whether the difference could be between currently married and ever married women, but the weighted number of currently married women is greater than the denominator in the API extract (13,624 compared to 12,842). Also in the country report the denominator (12,842) refers to ever-partnered women.

Thank you very much for your help.
Re: Peru domestic violence data [message #9292 is a reply to message #9268] Fri, 04 March 2016 13:18 Go to previous message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
Here is more from Dr. Shea Rutstein:
Quote:

The Domestic Violence questionnaire applies only to one woman per household. It was answered by 17299 women 15-49 out of 23888 women respondents 15-49, 72.4%. Obviously, there should be fewer currently married women than ever married women. I looked up the API results, they match the report for ever-married women (12,842 weighted). Using the same percentage currently married among ever married respondents (84.8% from table 5.1 of the report), and applying that to the ever-married with DV responses, I get .848*12842= 10893 currently married women. So the IPV in the last 12 months should be about 12.9/.848=15.2%.

Best regards,
Shea

Previous Topic: Liberia 2007 DHS Past 12 month IPV
Next Topic: Nepal 2011 Men's Attitudes about when Beating Wife is Justified
Goto Forum:
  


Current Time: Sun Nov 24 06:55:02 Coordinated Universal Time 2024