Changing data from persons to teams

[carla Oth]

Hi all,

I have to do something that I have never done before and I didn't find any info online, so I am trying luck here. Basically, I have a dataset composed of individuals. I will paste an example, as I believe it is easier to explain:

Person	Team	Year	Score
1	A	2000	3
2	B	2001	2
1	B	2000	3
3	A	2000	4
1	A	2001	2

In essence, I have a (very disorganized, sorry) list of people who belong to different teams (they can belong to more than 1 team simultaneously, as person 1 in 2000), and in each year these persons have a score.
Each team is composed of a number of individuals that is not fixed (for instance, some teams will have 3 people, others 5, others 10...).

My goal is to end up with team-level data, so that each team will have 1 observation per year, and the variable "score" will measure the average score of all team members (tricky, because as I said before, the number of individuals in each team is random, and I do not have a variable that indicates the number of individuals in each team).

The end result should be something like this (1 observation per team and year, and I do not care about the individuals anymore):

Team	Year	Average_score
A	2000	3.5
A	2001	3
A	2002	5
B	2000	2
B	2001	3
B	2002	4

How should I proceed to do this?

Thank you very much, any help is greatly appreciated.

Best,

Carla

[William Lisowski]

* Example generated by -dataex-. For more info, type help dataex
clear
input byte person str1 team int year byte score
1 "A" 2000 3
2 "B" 2001 2
1 "B" 2000 3
3 "A" 2000 4
1 "A" 2001 2
end
collapse (mean) score, by(team year)
list, clean

. list, clean

       team   year   score  
  1.      A   2000     3.5  
  2.      A   2001       2  
  3.      B   2000       3  
  4.      B   2001       2

[carla Oth]

Thank you so much!