Multilevel data management

[jose antonio]

Hello everyone,

Im working with a dataset that includes 2 levels of data: individual level data and schools level data. I created school level variables using individual level observations. For example, i have something like this:

School_ID	Ind_ID	Math_score	mean_school_math_score
1	1	100	93.6
1	2	90	93.6
1	3	91	93.6
2	4	90	90.6
2	5	95	90.6
2	6	87	90.6
3	7	100	92
3	8	86	92
3	9	90	92

Now, i just need the school information variables, because i need to merge 5 datasets with information of different years, i created all the variables using egen. I think collapse can be helpful, but that means i have to create all the variables again(it is a lot of work). theres something i can do to get a table like this?

School_ID	Mean_school_math_score
1	93.6
2	90.6
3	92

Thank you very much, im sorry if my english is not very good.

[Clyde Schechter]

drop Ind_ID Math_score
by School_ID, sort: keep if _n == 1

[jose antonio]

Thank you very much. It worked

[Nick Cox]

See also collapse.