| Re: Early Childhood mortality in Myanmar [message #14766 is a reply to message #14239] |
Fri, 04 May 2018 04:45  |
Sai San Moon Lu
Messages: 5
Registered: January 2018
Location: Umea, Sweden
|
Member |
|
|
Dear User,
Regarding your question, I like to share what i have done so far.
When I tested log regression to investigate association between mother's education and under five mortality (uni-variate analysis), I used the following commands for instance;
** SAMPLE WEIGHTING **
generate wgt = v005/1000000
svyset [pw=wgt], psu(v021)strata(v023)
label variable wgt " sample weight"
*** RECODING INDEPENDENT VARIABLES ***
* Recode Mother Education *
recode v106 (3=0 "higher") (2=1 "secondary") (1=2 "primary") (0=3 "no education") (.=99 "missing"), generate (M_E)
label variable M_E "Mother Education Recode"
*** RECODING DEPENDENT VARIABLES**
* Under 5 mortality (0-4yrs)*
recode b7 (0/59=1 "U5 death") (else=0 "alive >5yrs"), generate (u5)
label variable u5 "Under 5 mortality (0-4yrs)"
** Logistic regression for mother's education and u5 deaths (Uni-variate) ***
drop if M_E==99
svy: logistic u5 i.M_E
If you want to calculate the under-five mortality rate additionally, pls refer the following link;
https:// userforum.dhsprogram.com/index.php?t=msg&th=6371&got o=13114&#msg_13114
Hope it will be helpful..
Regards,
Moon
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
