CODEWITHSUNDEEP
rdplyr
Amer
Amer

Reputation: 2251

Summarise in dplyr package: How to go around `Error: expecting a single value`

I want to make a summary by ID, DRUG, FED summarising the sum of the CONC for DVID =1 and DVID==2

df<- 
ID  DRUG  FED  DVID  CONC
1    1     1    1     20
1    1     1    2     40
2    2     0    1     30
2    2     0    2     100

I tried using this:

df2 <- df %>% 
  group_by(ID,DRUG,FED) %>% 
  summarise(SumCOnc=CONC+lag(CONC))

However I am getting this error:

Error: expecting a single value

I don't get the error when I use mutate. Is there a way to go around it so I use summarise in the case described above?

The output should basically be this:

ID  DRUG  FED  SumConc
1    1     1    60
2    2     0    130

Upvotes: 3

Views: 7353

Answers (3)

akrun
akrun

Reputation: 886938

Or from data.table

library(data.table)
setDT(df)[, .(SumConc = sum(CONC)), .(ID, DRUG, FED)]
#   ID DRUG FED SumConc
#1:  1    1   1      60
#2:  2    2   0     130

Upvotes: 4

Ronak Shah
Ronak Shah

Reputation: 388797

A simple Base R approach would be using aggregate 

aggregate(CONC ~ ID + DRUG + FED, df, sum)

#  ID DRUG FED CONC
#1  2    2   0  130   
#2  1    1   1   60

Upvotes: 4

Ben Bolker
Ben Bolker

Reputation: 226057

This seems pretty straightforward: just use sum(), don't mess around with lag() ...

Get data:

df<- read.table(header=TRUE,
text="
ID  DRUG  FED  DVID  CONC
1    1     1    1     20
1    1     1    2     40
2    2     0    1     30
2    2     0    2     100
")

Process:

library(dplyr)
df %>% 
  group_by(ID,DRUG,FED) %>% 
  summarise(SumConc=sum(CONC))

##      ID  DRUG   FED SumConc
## 1     1     1     1      60
## 2     2     2     0     130

Upvotes: 6

Related Questions