Summing values in R table by 2 factors

Question

I have a single big text file which looks as follows:

tag            colony   diff
1035            03      498
1035            03     -44365
1035            03     -66652
1035            04      234234
1035            04     -343
1035            04     -23423
1037            10      234234
1037            10     -343
1037            10     -23423

Most 'tags' only have a single colony, such as 1037 in the above example. However, some have 2, such as 1036 having both 03 and 04. What I would like to do is sum the diff column for each tag, but separately for each colony, so the output would be something like this.

tag    colony    total
1035   03        -110 519
1035   04        210 648
1037   10        210 648

So far (I've been working in R), I have been using aggregate:

x2 = aggregate(x$diff, by=list(tag=x$tag), FUN=sum)

But this would count all tags together, irrespective of colony. Is there a way of 'adding another level', so to speak, into the aggregate function, so that it counts the colonies seperately?

Thanks

Answer 1

We can use dplyr

library(dplyr)
df1 %>%
   group_by(tag, colony) %>%
   summarise(total = sum(diff))

---

Or data.table

library(data.table)
setDT(df1)[, .(total = sum(diff)), .(tag, colony)]

Answer 2

x2 <- aggregate(x$diff, by=list(x$tag,x$colony), FUN=sum)

或等效地作为公式x2 <- aggregate(diff~tag+colony,data=x,FUN=sum)