Question

I have a dataset that looks like:
order year value
1 1990 2
1 1990 1
1 1990 1
2 1990 0
3 1990 4
1 1991 3
2 1991 2
2 1991 0

I have been trying to get the count of the unique orders in particular year. I tried table(data$year) which gives me:
1990 1991
5 3

However, I would like to get:

1990 1991
3 2

Answer 1

This will do the trick. It uses tapply() to separately count the number of unique values of order for each level of year.

df <- read.table(text="order year value
1 1990 2
1 1990 1
1 1990 1
2 1990 0
3 1990 4
1 1991 3
2 1991 2
2 1991 0", header=T)

with(df, tapply(order, year, function(X) length(unique(X))))
# 1990 1991 
#    3    2