Summing data based on Column in R

Question

I have a data set that looks like this (actual data is 10K by 5K so I really need a shortcut):

Cluster	Item1	Item2	Item 3
1	1	2	2
1	3	1	1
1	1	3	0
2	3	2	0
2	0	0	2
2	4	2	2
3	0	1	1
3	1	1	2

I want to add the columns of each data set by cluster so it will look I this:

Cluster	Item1	Item2	Item 3
1	5	6	3
2	7	4	4
3	1	2	3

I want to sum them by a certain column.

Answer 1

You can use aggregate ( dat is the name of your data frame):

aggregate(dat[-1], dat["Cluster"], sum)

#   Cluster Item1 Item2 Item3
# 1       1     5     6     3
# 2       2     7     4     4
# 3       3     1     2     3

---

With data.table :

library(data.table)
setDT(dat)[ , lapply(.SD, sum), by = Cluster]
#    Cluster Item1 Item2 Item3
# 1:       1     5     6     3
# 2:       2     7     4     4
# 3:       3     1     2     3

---

With dplyr :

dat %>%
  group_by(Cluster) %>%
  summarise_each(funs(sum))
#   Cluster Item1 Item2 Item3
# 1       1     5     6     3
# 2       2     7     4     4
# 3       3     1     2     3

Answer 2

thanks for your answer, I also used this good and it worked perfectly:

 aggregate(. ~ Cluster, data=dat, FUN=sum)



#   Cluster Item1 Item2 Item3
# 1       1     5     6     3
# 2       2     7     4     4
# 3       3     1     2     3

Answer 3

Try:

> sapply(ddf[-1], function(x) tapply(x,ddf$Cluster,sum))
  Item1 Item2 Item3
1     5     6     3
2     7     4     4
3     1     2     3

Answer 4

If you want to sum all varibales except that of grouping, use across in dplyr

df <- read.table(text = "Cluster    Item1   Item2   Item3
1   1   2   2
1   3   1   1
1   1   3   0
2   3   2   0
2   0   0   2
2   4   2   2
3   0   1   1
3   1   1   2", header = T)

df %>% group_by(Cluster) %>% summarise(across(everything(), ~sum(.)))

# A tibble: 3 x 4
  Cluster Item1 Item2 Item3
    <int> <int> <int> <int>
1       1     5     6     3
2       2     7     4     4
3       3     1     2     3