R join same row and calculate mean value

Question

I have a data frame that looks like this:

data<-data.frame(y=c(1,1,2,2,3,4,5,5),x=c(5,5,10,10,5,10,5,5))

  y  x
1 1  5
2 1  5
3 2 10
4 2 30
5 3  5
6 4 10
7 5  4
8 5  8

How can a merge those rows with same value in y column and modify the x column value to the mean of them.

I would like something like this:

I'm trying:

unique(data)

But it removes the values instead of doing the mean of same rows.

The actual dupe should be [this](http://stackoverflow.com/questions/21982987/mean-per-group-in-a-data-frame) I suppose. But who cares, right? No one uses Google anymore anyway. — David Arenburg, Jan 04 '16 at 14:42
@DavidArenburg Why google when the helpful people on SO give you a tailor-made answer in no time anyway? — Henrik, Jan 04 '16 at 15:57

Marta · Answer 1 · 2016-01-04T13:04:05.183

3

It is easy with dplyr. Like here:

library("dplyr")

data %>% 
    group_by(y) %>% 
    summarise(x=mean(x))

edited Jan 04 '16 at 13:04

answered Jan 04 '16 at 12:54

Marta

3,032
3
17
34

Add `y` to summarize for a complete output. i.e. `summarize(y, x=mean(x))` – Tal J. Levy Jan 04 '16 at 13:26
@Tal J. Levy, I don't think it's necessary. `y`'s already there. – Marta Jan 04 '16 at 14:06
It won't be after the summarize. – Tal J. Levy Jan 04 '16 at 14:09
I've checked and it is there. It's because `y` is a grouping variable. – Marta Jan 04 '16 at 14:10
Of course, sorry I forgot about the grouping. You are correct. – Tal J. Levy Jan 04 '16 at 14:21

score 2 · Answer 2 · answered Jan 04 '16 at 12:56

2

We can use aggregate

aggregate(x~y, data, mean)

answered Jan 04 '16 at 12:56

akrun

874,273
37
540
662

score 1 · Answer 3 · answered Jan 04 '16 at 12:55

1

User plyr.

# Create dummy data.
nel = 30
df <- data.frame(x = round(5*runif(nel)), y= round(10*runif(nel)))

# Summarise means
require(plyr)
df$x <- as.factor(df$x)
res <- ddply(df, .(x), summarise, mu=mean(y))

answered Jan 04 '16 at 12:55

Joe

1,455
2
19
36

R join same row and calculate mean value

3 Answers3