keep common household information only in the first row of the household

[Nirosha Dissanayaka]

Hi all,

I have data set with household characteristics.

District	Village	Household_ID	Menber_ID	Monthly_NonFood_Exp
5	1	1	1	610
5	1	1	2	610
5	1	1	3	610
5	1	1	4	610
7	3	2	1	845
7	3	2	2	845
7	3	2	3	845
7	3	2	4	845
7	3	2	5	845

The monthly food expenditure of the household repeats for all the household members.
I want to keep the monthly food expenditure on the first row only.
Simply, I want the data set to look like the following table.
Can anyone please help me?

District	Village	Household_ID	Menber_ID	Monthly_NonFood_Exp
5	1	1	1	610
5	1	1	2	.
5	1	1	3	.
5	1	1	4	.
7	3	2	1	845
7	3	2	2	.
7	3	2	3	.
7	3	2	4	.
7	3	2	5	.

[Mike Lacy]

I'll explain how to do this. However, my best guess is that doing what you want is unnecessary, as changing those values to missing will not save any space, and will not enable you to do anything that you couldn't do by leaving the data as is and using "if" for certain commands. If the latter puzzles you, you might want to explain why you wanted to do this so that (if necessary) your understanding might be corrected.

Presuming that the "first" person in each household is always the one with Menber_ID == 1, all you would need is

replace Monthly_NonFood_Exp = . if Menber_ID != 1

If "first" is defined in some other way, you'll need to tell us what that is.
(Note also "Menber" for "Member")

[Nick Cox]

I agree with Mike Lacy.

One likely answer to Why do you want to do this? is So I can work household data at that level. Fine, and egen, tag() makes this much easier.

* Example generated by -dataex-. For more info, type help dataex
clear
input byte(district village household_id menber_id) int monthly_nonfood_exp
5 1 1 1 610
5 1 1 2 610
5 1 1 3 610
5 1 1 4 610
7 3 2 1 845
7 3 2 2 845
7 3 2 3 845
7 3 2 4 845
7 3 2 5 845
end

. egen tag = tag(district village household_id)

. su monthly if tag

    Variable |        Obs        Mean    Std. dev.       Min        Max
-------------+---------------------------------------------------------
monthly_no~p |          2       727.5    166.1701        610        845

[Nirosha Dissanayaka]

Dear Mr. Nick and Mr. Mike,
Thank you very much for your kind reply and help.
I actually wanted to do this, to run the balance test.
When I try to get the balance test for treatment and control groups, I found these repeat numbers do not provide the correct balance test.
Since I am a beginner to stata, I may wrong.
I will try these commands that you showed me.
Thank you once again for both of you.