How to calculate prevalence rate (twinning rate) using DHS child dataset

[Robel Alemu]

Hi all,

I am working on DHS child dataset and wanted to calculate the twinning rate (number of twins per 1000 births in a given country and survey year). I have a twin dummy variable called count (1=single birth, 2=twin birth, 3=triplet birth), country code (country_recode), id(hh_id), child date of birth (ch_dob). I would really appreciate it if you could let me know what code I should use to calculate the the number of twins in each 1000 births by country and survey year.

Thanks

[Robel Alemu]

I haven't heard from anyone on my question. Here is how I tried to create the twinning rate:

by sasia2 country_recode end_year, sort : tab1 ch_id

capture drop twin
gen twin=count==2

capture drop twin_birth
egen twin_birth = (1000*mean(twin))/total(twin), by(sasia2 country_code end_year)
bys sasia2 country_code end_year: eststo: estpost sum twin_birth if yrssincebirth!=.
esttab using twinbirth_sum.csv, cells("mean") title(Table A1:Summary of Demographic and Health Surveys conducted to date) ///
mtitles nodepvars nonumber compress replace

Hope this will be helpful for someone who wants to create the same kind of variable as I wanted.

[Mike Lacy]

In the future, you should post questions like this in the General forum. This forum ("Sandbox") is for people wanting to experiment with using StataList's web software, not for asking questions about how to use Stata. Consequently, people don't ordinarily look here to see if anyone has a question about using Stata.