The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Child discipline-South Africa Demographic health Survey 2016 (S934a and S934b (child discipline))
Child discipline-South Africa Demographic health Survey 2016 [message #22302] Wed, 24 February 2021 17:14 Go to next message
Shek is currently offline  Shek
Messages: 19
Registered: November 2014
Member
I am trying to replicate Table 19.11.1 Child discipline: Women and Table 19.11.2 Child discipline: Men of the South African Demographic Health Survey, 2016

Using the IR file i am getting a total of 4,612 women with one or more children less than age 18 living with them as compared to 4, 613 reported in the survey report.

Furthermore, how was the figures for child discipline derived for Men (Which file was used?)

I will also be glad if you could assist me with the process of merging the child discipline variables to the household file

Thank you very much
Re: Child discipline-South Africa Demographic health Survey 2016 [message #22311 is a reply to message #22302] Thu, 25 February 2021 09:36 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2265
Registered: February 2013
Senior Member
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

The following lines will lead to table 19.11.1, after opening the IR file:

gen     hita=.
gen     hitb=.
gen     hitab=.
replace hita=0  if s934a<. 
replace hitb=0  if s934b<.
replace hitab=0 if hita==0 | hitb==0

replace hita=1  if s934a==1
replace hitb=1  if s934b==1
replace hitab=1 if hita==1 | hitb==1

tab v013 hita [iweight=v005/1000000], row


This would be repeated for hitb and hitab and other covariates. There are more elegant procedures for constructing the table but this is the basic coding. Similarly for men, to get table 19.11.2, using the MR file and sm620a/b.

I also get 4612 weighted cases rather than 4613. Don't worry about that discrepancy. The coding is correct.

The forum posting includes examples of merging the IR and MR files with the PR file. Such merges will put the child discipline variables onto the mother's or father's lines in the PR file. It will not link them to specific children.
Re: Child discipline-South Africa Demographic health Survey 2016 [message #22318 is a reply to message #22311] Thu, 25 February 2021 16:36 Go to previous message
Shek is currently offline  Shek
Messages: 19
Registered: November 2014
Member
Thank you very much for your prompt response.
Previous Topic: Immunization Coverage
Next Topic: Child Health and WASH
Goto Forum:
  


Current Time: Thu May 6 11:21:58 Coordinated Universal Time 2021