Home » Countries » Nigeria » Child age variable
| Child age variable [message #26176] |
Tue, 14 February 2023 18:01  |
woojae1995
Messages: 6
Registered: January 2023
|
Member |
|
|
I found that there are several variables that calculate children's age. I am currently working with the hc1 variable in the PR data and the b19 variable in the KR data.
I was wondering which variable is more accurate for analysis.
From this website ( https://dhsprogram.com/data/calculating-the-age-of-children. cfm) I think it says that b19 is the age calculated when the interview was held and hc1 is the age calculated when biomarkers were measured.
I am trying to do analysis regarding malaria in children; would that mean hc1 is more accurate to use?
I am asking because there are quite a lot disconcordance between hc1 & b19; from my crude coding I did below after merging the PR data with KR data
PRtemp =subset(PRdata, select=c(v001, v002, v003, hc1), 'NA'= TRUE)
KRdata <- merge(KRdata, PRtemp, by=c("v001", "v002", "v003"))
KRdata <- KRdata%>%
mutate(concordance = case_when(
b19 == hc1 ~ 1,
!(b19 == hc1) ~ 0))
KRsvy <- svydesign(id = KRdata$v021, strata=KRdata$v022, weights = KRdata$v005/1000000, data=KRdata)
svymean(~concordance, KRsvy, na.rm=TRUE)
svytable(~concordance, KRsvy)
prop.table(svytable(~concordance,KRsvy),)
I get 3138.414 for 1, 2636.717 for 0 so nearly 45% of the data have disconcordance in their age data.
Please help me in identifying which variable I should use regarding children age. Thank you!
|
|
|
|
Current Time: Sun Nov 24 12:49:31 Coordinated Universal Time 2024
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
