Reputation: 783
I have a table that looks like this:
ID Total
3 3
3 3
3 3
4 11
4 11
4 11
4 11
4 11
4 11
6 9
6 9
7 13
7 13
7 13
7 13
7 13
7 13
7 13
7 13
7 13
7 13
7 13
7 13
7 13
I would like to calculate the median of column B (Total), excluding duplicate combinations of columns A and B. This could be achieved by constructing a table as below, and calculating the median from that table.
ID Total
3 3
4 11
6 9
7 13
Is there any way of obtaining the median without having to go through this process of manually deleting duplicates?
Upvotes: 0
Views: 925
Reputation: 7742
=MEDIAN(IF(FREQUENCY(MATCH(A2:A25&"|"&B2:B25,A2:A25&"|"&B2:B25,0),ROW(A2:A25)-MIN(ROW(A2:A25))+1),B2:B25))
Upvotes: 1
Reputation: 201
There is a way with two additional columns. The first column is concatenation of ID and Total, the second counts occurences of each individual combination. Then you just do the median on those rows where the combination occurs for the first time.
Upvotes: 0