Flag id's that have a tag==0

[Maria Elstad]

I have a maternity dataset, where each mother has a record per pregnancy. I want to do two things:
1. Find the mothers with multiple pregnancies.
I do this by sorting the id's and generating a tag.
sort id
egen tag=tag(id)

2. Now I want to browse all the records for all the mothers that have multiple pregnancies.
Currently I do:

browse if tag==0 //to find the id's
browse if id==1 | id==4

Is there a more efficient way to do this? Maybe flag the id's that have a tag==0?


Example data:

id	tag
1	1
1	0
2	1
3	1
4	1
4	0
5	1
6	1
7	1

Would be grateful for any help.

Thanks,
Maria

[Andrew Musau]

bys id: egen wanted= min(tag)

See
https://www.stata.com/support/faqs/d...ble-recording/

[Rich Goldstein]

how about:

egen count=count(id), by(id)
browse if count>1

untested code - if you had used -dataex- (see the FAQ), I could have tested this on your data

[Nick Cox]

As Rich explains you can count pregnancies directly, as with his code or with

bysort id: gen wanted = _N

from which you can look at mothers with multiple pregnancies by

browse if wanted > 1 & tag

That way you see each mother just once -- but the details for one pregnancy only.

[Maria Elstad]

Thank you all for your quick relies.

All three suggestions are doing exactly what I was looking for, Brilliant!