|Replicate Weighted/Unweighted Number of Children [message #14858]
||Wed, 09 May 2018 11:16
Registered: May 2018
Location: Washington D.C.
Dear DHS Experts,|
I am trying to replicate the numbers from Table 11.1 Nutritional Status of Children (Page 159), in Ethiopia 2011 report. Below is my code for getting the percentages for different age groups and the weighted and unweighted number of children:
gen stratid = v022
gen psu = v021
gen weight = v005/1000000
svyset psu [pw=weight], strata(stratid)
*living with mother (respondents)
keep if b9==0
*child's age in months
label var age "Age in Months"
*drop if not alive
keep if b5==1
recode age (0/5=1 "0-5")(6/8=2 "6-8")(9/23=3 "9-23")(24/59=4 "24-59")(else=.), gen(child_age)
*tab of all living children
tab child_age [iweight=weight]
*Stunted & Wasted
*The measures are presented with two implied decimal places.
*No decimal points are included in the data file.
tab hw70, missing
replace stunted=1 if hw70<-200
replace stunted=. if hw70==.
replace stunted=. if hw70>600
tab stunted, missing
tab stunted [iweight=weight]
tab child_age stunted, row
tab child_age stunted [iweight=weight], row
My age intervals are slightly different from the report. But for example, the first category, children under 6 months, the numbers I got are still different from the report.
Percentage below -2SD: 9.92% (10.0% in the report); Weighted number of children: 1,042 (1,078 in the report); Unweighted: 995 (1015 in the report).
It would be really great if anybody could help me with this. Thanks in advance!