Home » Topics » Nutrition and Anthropometry » incorrect stunting rates
| Re: incorrect stunting rates [message #14243 is a reply to message #14233] |
Sun, 11 March 2018 11:38   |
Nicholus Tint Zaw
Messages: 13
Registered: March 2018
Location: Myanmar
|
Member |
|
|
H, I am Nicholus from Myanmar and I am also experiencing same issued as margovg did before.
I am now working on child nutrition status data analysis and used KR at first and use following stat code to identify the mean HAZ score and stunting rate of children. But my result was quite different in term of denominator with the number from DHS Myanmar report. (4,213 in my analysis and 4,089 in DHS report)
[global
global kr_child D:\SPA SCI\Technical Knowledge\GOVERMENT INFO\DHS 2015-16\dataset\MM_2015-16_DHS_05022017_255_106315\mmkr71dt\ MMKR71FL.dta
global pr_person D:\SPA SCI\Technical Knowledge\GOVERMENT INFO\DHS 2015-16\dataset\MM_2015-16_DHS_05022017_255_106315\mmpr71dt\ MMPR71FL.dta
*----------------------------------------------------------- --------------------
use "${kr_child}", clear
** Calculate Stunting Rate **
** Use DHS HAZ Variable **
tab hw70, m
gen haz_dhs_revised = hw70
replace haz_dhs_revised = .n if hw70 >= 9996
replace haz_dhs_revised = hw70/100 if !mi(haz_dhs_revised)
tab haz_dhs_revised, m
gen stunt = (haz_dhs_revised < -2)
replace stunt = .n if mi(haz_dhs_revised )
tab stunt, m
** Construct WEIGHT var using Women Insidividual Sample Weight **
gen wgt = v005/1000000
tab wgt, m
lookfor sampling // search for samping unit var
** Generate Stunting Rate & HAZ Mean Score by Weighted Data **
svyset, clear
svyset [pw = wgt], psu(v021) strata (v022)
svy: mean haz_dhs_revised stunt
svyset, clear][/code]
Then, when I found this post in user forum, I revised my code and dataset usage with PR file (as follow). But, still got the different figure compare with DHS report and this time the denominator discrepancy became more larger as my analysis figure become 4,640.
[code][/clear
use "${pr_person}", clear
** Check with DHS Calculated HAZ Variable **
tab hc70, m
gen haz_dhs_revised = hc70
replace haz_dhs_revised = .n if hc70 >= 9996
replace haz_dhs_revised = hc70/100 if !mi(haz_dhs_revised)
replace haz_dhs_revised = .n if hv103 != 1
tab haz_dhs_revised, m
tab hv103, m // should be 1
** Construct WEIGHT var using Women Insidividual Sample Weight **
gen wgt = hv005/1000000
tab wgt, m
lookfor sampling // search for samping unit var
gen stunt = (haz_dhs_revised < -2 & hv103 == 1)
replace stunt = .n if hv103 != 1 | mi(haz_dhs_revised)
tab stunt, m
svyset, clear
svyset [pw = wgt], psu(hv021) strata (hv022)
svy: mean haz_dhs_revised stunt
svyset, clear
]
So, I was wondering there is anyone who can review my code and provide guidance to fix this issue. Thanks in advance.
Best regards,
Nicholus
nicholus
[Updated on: Sun, 11 March 2018 11:39] Report message to a moderator |
|
|
|
Current Time: Sun Dec 22 10:38:28 Coordinated Universal Time 2024
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
