The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Child Discipline - Myanmar DHS 2015-2016 (Unit of analysis and files to be used )
Child Discipline - Myanmar DHS 2015-2016 [message #22333] Mon, 01 March 2021 02:31 Go to next message
Nyan Linn is currently offline  Nyan Linn
Messages: 8
Registered: December 2020
Member
Dear Sir/Madam,

I would like to find the association of the family and household characteristics with child discipline. So, I would like to ask some questions.
Which files should I use? HR file or PR file?

Could you explain me please about unit of analysis; will be household or household members or children?

If PR file is needed to use, how should I select the sampling units which were used for child discipline module (because the PR files have over 40000 samples).

I would like to include occupation & employment of parents. So, I would like to combine IR files and MR files to PR file. Could you instruct me please how to combine?

Thank you very much.
With Deep Respect
NYAN LINN.
Re: Child Discipline - Myanmar DHS 2015-2016 [message #22338 is a reply to message #22333] Mon, 01 March 2021 10:49 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3203
Registered: February 2013
Senior Member

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

I assume you have already looked at chapter 17, and the relevant part of the household questionnaire, in the main report. In this survey, of all the children age 2-14 in the household, one was randomly selected. The child discipline questions were asked with reference to that one child. However, in the PR household file, which has one record for everyone in the household, the responses for that child were attached to every person in the household, regardless of whether they are the reference child or are even age 2-14. The variables are sh148-sh161.

For most purposes, the units of analysis are the children. If you want to get an estimate of, say, what proportion of children age 2-14 are "Yes" on sh150, I would use the PR file, restrict to children age 2-14, and run "tab sh150 [iweight=hv005/1000000]". (You can also use svyset and svy.) The response for the reference child would be imputed to all the other children in the household. But if you wanted this proportion by characteristics of the children, such as their age, you would need to restrict to the reference children, with "if hvidx==sh148". I would recommend that when you select with hvidx=sh148, you also weight by k*hv005, where k is the number of children age 2-14 in the household. The sampling probability within the household is 1/k, and you should compensate for that.

For the selected child (the one with hvidx=sh148) you have up to 3 possible relevant adults in the household: the mother, the father, and the caregiver. The line numbers for these adults are hv112, hv114, and sh149, respectively. Usually the caregiver will be the mother (that is, sh149==sh112), but not always. The mother and father are not always alive and in the household. I see that some children are missing a code for the caregiver.

If you look on the forum you can find ways to attach characteristics of the mother (in the IR file) to the child in the PR file, IF the child and the mother are in the same household. The same approach can be used to link the characteristics of the father (in the MR file) to the child. You may be able to link the occupation (for example) of the caregiver to the child, if the caregiver is in the IR or PR file. The caregiver may be a grandparent and outside the age range of the IR or MR file. If the caregiver is the mother or father, and in the same household as the child, you will be ok.

Re: Child Discipline - Myanmar DHS 2015-2016 [message #22340 is a reply to message #22338] Mon, 01 March 2021 11:43 Go to previous messageGo to next message
Nyan Linn is currently offline  Nyan Linn
Messages: 8
Registered: December 2020
Member
Thank you very much, Sir.
Re: Child Discipline - Myanmar DHS 2015-2016 [message #22590 is a reply to message #22338] Wed, 07 April 2021 15:39 Go to previous messageGo to next message
Nyan Linn is currently offline  Nyan Linn
Messages: 8
Registered: December 2020
Member
Dear Sir/Madam,
I am using Stata for the analysis of DHS data. I have some difficulties in data management. I would like to ask some questions.
1. To investigate factors associated with child disciplinary practices, i would like to add some variables in the PR file . "age of primary caregiver", "Sex of primary caregiver", "education level of primary caregiver". I tried to change to wide form but failed. Could you instruct me and give me your expert opinion, please?

2. "occupation of primary caregiver"/ "occupation of the mother" and "occupation of the father", "Employment of primary caregiver"/ "Employment of the mother" and "Employment of the father". Could you give me your expert opinion to focus on primary caregiver or (mother and father)? Which is more appropriate to focus according to structure of the data sets? I think i need to combine IR file and MR file with PR file (main file). Could you give me your expert opinion and instructions to combine data sets?

3. I would like to add the variable " presence of domestic violence towards any women in the family of the child" / " presence of domestic violence towards mother of the child". Which one is more appropriate to use? Could you give me your expert opinion, please? To add this variable, i think i need to combine IR file to PR file. Could you instruct me please how to combine?

With Deep Respect
Nyan Linn.
Re: Child Discipline - Myanmar DHS 2015-2016 [message #22594 is a reply to message #22590] Thu, 08 April 2021 08:51 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3203
Registered: February 2013
Senior Member
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

For children under 18, hv112 is the line number of the mother and hv114 is the line number 5. For children under 5, hc51 is the line number of the caregiver. Usually, for children under 5, hc51=hv112; the mother is the caregiver. You could start by looking at how often hc51 equals hv112 or hv114. If the mother is not in the household, then hv112 equals 0 or NA, and similarly for the father and hv114. Below is Stata code that should be helpful.

For your final question, I would look at the DV variables in the IR file and construct a reduced household-level file for which the ID codes are v001 v002. You then do a merge that matches v001 and v002 with hv001 and hv002 in the child file constructed below. Every child in the same household would get the same household-level code. Some children in the PR file are in a household for which there will not be any domestic violence data. It's a little more complex to merge the DV data from the mother or from the caregiver (if the caregiver is not the mother, but is another woman in the household and received the DV module). I would recommend household-level DV variables. Good luck--an interesting research topic.

* need a scratch folder
cd e:\DHS\DHS_data\scratch

use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\KHPR73FL.DTA" , clear
keep hv001 hv002 hvidx hv104 hv105 hv106 hc51 hv112 hv114
save  KHPRtemp.dta, replace

* Construct a file of children with hc51
keep if hc51<.
rename hc51 line_caregiver
rename hv104 hv104_child
rename hv105 hv105_child
rename hv106 hv106_child
gen caregiver=.
replace caregiver=0 if line_caregiver==0
replace caregiver=1 if line_caregiver==hv112
replace caregiver=2 if line_caregiver==hv114
replace caregiver=3 if caregiver==. & line_caregiver>0
label define cg 0 "Not in hh" 1 "Mother" 2 "Father" 3 "Other hh member"
label values caregiver cg
sort hv001 hv002 line_caregiver
save KHchildtemp.dta, replace

use  KHPRtemp.dta, clear
rename hvidx line_caregiver
rename hv104 hv104_caregiver
rename hv105 hv105_caregiver
rename hv106 hv106_caregiver
sort  hv001 hv002 line_caregiver
merge hv001 hv002 line_caregiver using KHchildtemp.dta
tab _merge

* _merge is 2 for children whose caregiver is not in the household
* _merge is 3 for children who are matched with caregiver
keep if _merge==2 | _merge==3
drop _merge

tab hv105_child caregiver 
Re: Child Discipline - Myanmar DHS 2015-2016 [message #22641 is a reply to message #22594] Thu, 15 April 2021 03:55 Go to previous messageGo to next message
Nyan Linn is currently offline  Nyan Linn
Messages: 8
Registered: December 2020
Member
Dear Sir,
I would like to add the variables "caregiver's age", "caregiver's sex" and "education level of caregiver" for the child asked for child disciplined module. But, i have difficulties to drop and keep the observations after adding those variables. I am not clear about sh11a, hv102 and sh148==hvidx.

*****create a file of children including caregiver's information
****1st step - construct KHRtemp.dta file
keep hhid hv009 hvidx hv024 hv000 hv025 idxh4 hv102 hv001 hv201 hv002 hv204 sh11a hv235 hv003 hv205 hv225 hv238 hv004 hv206 hv005 hv213 hv010 hv214 hv011 hv215 hv012 hv217 sh149 hv013 hv218 hv020 hv219 hv021 hv220 hv101 hv022 hv226 hv023 hv241 hv027 hv242 hv028 hv270 hv044 sh148 hv104 hv105 sh148f sh150 sh151 sh154 sh153 sh157 sh152 sh155 sh156 sh159 sh158 sh160 sh161 hv106 hv109 hv111 hv112 hv113 hv114
save KHRtemp.dta, replace

****2nd step - Construct a file of children with sh149 - rename child's information and create caregiver variable (kchildtemp.dta)
keep if sh11a==1
keep if hv102==1
keep if sh148==hvidx
tab sh148
rename sh149 line_caregiver
rename hv104 hv104_child
rename hv105 hv105_child
rename hv106 hv106_child
gen caregiver=.
replace caregiver=0 if line_caregiver==0
replace caregiver=1 if line_caregiver==hv112
replace caregiver=2 if line_caregiver==hv114
replace caregiver=3 if caregiver==. & line_caregiver>0
label define cg 0 "Not in hh" 1 "Mother" 2 "Father" 3 "Other hh member"
label values caregiver cg
sort hv001 hv002 line_caregiver
save Kchildtemp.dta, replace

**** 3rd step - rename caregiver's information on KHRtemp.dta file and merge with Kchildtemp.dta file
use KHRtemp.dta, clear
rename hvidx line_caregiver
rename hv104 hv104_caregiver
rename hv105 hv105_caregiver
rename hv106 hv106_caregiver
sort hv001 hv002 line_caregiver
merge hv001 hv002 line_caregiver using Kchildtemp.dta
tab _merge
* _merge is 2 for children whose caregiver is not in the household
* _merge is 3 for children who are matched with caregiver

**** 4th step - drop unnecessary observations.
**** here, I have difficulties to drop unnecessary observations. I couldn't get the observations correctly. Could you please check it for me? After adding the above variables, how to make the dataset to be appropriate for data analysis? I would like to find the factors affecting commitment of child discipline. Which people should be for my outcome variables?

keep if _merge==2 | _merge==3

tab sh11a
keep if sh11a==1
keep if hv102==1
keep if sh148==hvidx
tab sh11a

drop _merge
tab hv105_caregiver
tab sh148
save KHRtemp.dta, replace
Re: Child Discipline - Myanmar DHS 2015-2016 [message #22647 is a reply to message #22641] Fri, 16 April 2021 05:55 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3203
Registered: February 2013
Senior Member

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

When I run the program on your data, after "tab _merge" I get 70,099 cases with _merge=1. These are people in the PR file who are never named as caregivers. There are 21 cases with _merge=2. These are children for whom the caregiver is not in the household (there really should not be any such cases). There are 4267 cases with _merge=3. These are children who are successfully matched with their caregiver.

I suggested you keep the cases with _merge =2 or =3. You could just keep the cases with _merge=3, because the 21 cases with _merge=2 really don't contain any information about the caregiver. I suggested keeping them because then you have a complete file of children who have caregivers.

I don't know why you have so many variables in the "keep hhid...." line. But if you want to keep variables like that you will have to distinguish between whether they refer to the child or to the caregiver. You can just construct more "rename...child" and "rename....caregiver" commands, for those variables.

Hope this helps.
Re: Child Discipline - Myanmar DHS 2015-2016 [message #22656 is a reply to message #22647] Sat, 17 April 2021 13:35 Go to previous messageGo to next message
Nyan Linn is currently offline  Nyan Linn
Messages: 8
Registered: December 2020
Member
Thank you very much, Sir.
Dear Sir, I would like to reproduce the table 17.6 on page 296 of 2015-2016 Myanmar DHS final report.
But, I could not get the exact frequency and percentage for
- "Residence"
- "States/Regions"
- "Wealth quintile".
Could you please check my do file?
Which variable was used as "wealth quintile" in the report? When I tried with hv270 or hv270a, the frequency and percentages are differ from the tables. Could you please mention how to generate the variable "wealth quintile"?
With Deep Respect
Nyan Linn
use "C:\Users\ASUS\Desktop\child violent\Myanmar DHS\STATA\scratch\MMPR71FL.DTA"
ta sh11a
keep if sh11a==1
keep if hv102==1
keep if sh148==hvidx
gen wt=hv005/1000000
replace sh150=0 if sh150==.
replace sh151=0 if sh151==.
replace sh152=0 if sh152==.
replace sh153=0 if sh153==.
replace sh154=0 if sh154==.
replace sh155=0 if sh155==.
replace sh156=0 if sh156==.
replace sh157=0 if sh157==.
replace sh158=0 if sh158==.
replace sh159=0 if sh159==.
replace sh160=0 if sh160==.
replace sh161=0 if sh161==.
gen discipline = 0
replace discipline=1 if sh150==1|sh151==1|sh152==1|sh153==1|sh154==1|sh155==1|sh156= =1|sh157==1|sh158==1|sh159==1|sh160==1
label define dis 0"not use any discipline/missing" 1"number of children be used by discipline"
label values discipline dis
*******************************
gen only_non_violent=.

replace only_non_violent=0 if sh152==1 | sh153==1 | sh155==1 | sh156==1 | sh157==1 | sh158==1 | sh159==1 | sh160==1

replace only_non_violent=1 if sh152==0 & sh153==0 & sh155==0 & sh156==0 & sh157==0 & sh158==0 & sh159==0 & sh160==0

replace only_non_violent=. if sh150==. & sh151==. & sh154==.
tab only_non_violent
*********************************
gen any_non_violent=.
replace any_non_violent=0 if sh150==0 & sh151==0 & sh154==0
replace any_non_violent=1 if sh150==1 | sh151==1 | sh154==1
replace any_non_violent=. if sh150==. & sh151==. & sh154==.
tab any_non_violent
***********************************
tab any_non_violent only_non_violent
replace only_non_violent=0 if any_non_violent==0
tab any_non_violent only_non_violent

label define aaa 0"No" 1"Yes"
label values any_non_violent aaa
label values only_non_violent aaa
**************************************
generate any_psycho=.
replace any_psycho=1 if sh153==1 | sh157==1
replace any_psycho=0 if sh153==0 & sh157==0
replace any_psycho=. if sh153==. & sh157==.
tab any_psycho
label values any_psycho aaa
*******************************
gen any_mild_mod_physical=.
replace any_mild_mod_physical=1 if sh152==1 | sh155==1 | sh156==1 | sh159==1
replace any_mild_mod_physical=0 if sh152==0 & sh155==0 & sh156==0 & sh159==0
replace any_mild_mod_physical=. if sh152==. & sh155==. & sh156==. & sh159==.
label values any_mild_mod_physical aaa
tab any_mild_mod_physical
*********************************
generate any_severe=.
replace any_severe=1 if sh158==1 | sh160==1
replace any_severe=0 if sh158==0 & sh160==0
replace any_severe=. if sh158==. & sh160==.
label values any_severe aaa
tab any_severe
***************************
gen any_physical=.
replace any_physical=1 if sh152==1 | sh155==1 | sh156==1 | sh159==1 | sh158==1 | sh160==1
replace any_physical=0 if sh152==0 & sh155==0 & sh156==0 & sh159==0 & sh158==0 & sh160==0
replace any_physical=. if sh152==. & sh155==. & sh156==. & sh159==. & sh158==. & sh160==.
label values any_physical aaa
tab any_physical

***********************************8
gen any_violent=.
replace any_violent=1 if sh152==1|sh153==1|sh155==1|sh156==1|sh157==1|sh158==1|sh159= =1|sh160==1
replace any_violent=0 if sh152==0 & sh153==0 & sh155==0 & sh156==0 & sh157==0 & sh158==0 & sh159==0 & sh160==0
replace any_violent=. if sh152==. & sh153==. & sh155==. & sh156==. & sh157==. & sh158==. & sh159==. & sh160==.
label values any_violent aaa
tab any_violent
**********

. label variable discipline "Child Discipline Module"

. label variable only_non_violent "use only nonviolent methods"

. label variable any_non_violent "Use ANY nonviolent methods"

. label variable any_psycho "Use ANY psychological aggression"

. label variable any_mild_mod_physical "Use ANY mild to moderate physical punishment"

. label variable any_severe "Use ANY severe physical punishment"

. label variable any_violent "Use ANY Violent methods"

. label variable any_physical "Use ANY physical violent"

tab sh150 [iw=wt]
tab sh151 [iw=wt]
tab sh152 [iw=wt]
tab sh153 [iw=wt]
tab sh154 [iw=wt]
tab sh155 [iw=wt]
tab sh156 [iw=wt]
tab sh157 [iw=wt]
tab sh158 [iw=wt]
tab sh159 [iw=wt]
tab sh160 [iw=wt]
tab sh161 [iw=wt]

tab hv024 only_non_violent [iw=wt], row
tab hv024 any_psycho [iw=wt], row
tab hv024 any_physical [iw=wt], row
tab hv024 any_severe [iw=wt], row
tab hv024 any_violent [iw=wt], row

tab hv025 only_non_violent [iw=wt], row
tab hv025 any_psycho [iw=wt], row
tab hv025 any_physical [iw=wt], row
tab hv025 any_severe [iw=wt], row
tab hv025 any_violent [iw=wt], row

Re: Child Discipline - Myanmar DHS 2015-2016 [message #22669 is a reply to message #22656] Mon, 19 April 2021 14:32 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3203
Registered: February 2013
Senior Member

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

I have not done a complete check, but I am concerned with the initial lines like this: "replace sh150=0 if sh150==.". These variables (sh150-sh161) are only applicable to the children selected for the module. The dot or period indicates "Not applicable", which is not equivalent to "No". Can you try again without those lines? Also "discipline" should be coded NA if the sh variables are NA. It should not be initialized with code 0.

Previous Topic: ANC
Next Topic: Mothers Education on children's health
Goto Forum:
  


Current Time: Fri Dec 6 21:18:40 Coordinated Universal Time 2024