CODEWITHSUNDEEP
r
Danielle
Danielle

Reputation: 795

Paste two levels of different factors given conditional statement

I have a df like so:

df<- data.frame(result=c("p","p","p","d","d","p"),lep=c("A", "C", "E", "F","G", ""), psit= c("B", "D", "F", "G","H", ""))

I would like to make a new column paste that pastes two levels of different factors together, if it meets the following requirements: 1)result must = p 2)The lep and psit columns are not blank

And the output should appear as follows: 

output<- data.frame(result=c("p","p","p","d","d", "p"),lep=c("A","C", 
"E", "F","G", ""), psit= c("B", "D", "F", "G","H",""), paste= 
c("A_B","C_D", "E_F", "", "",""))

I have tried the following without success. I believe it is because I am not using the paste function properly:

df<-df %>%
 group_by(result) %>% 
  mutate(interact2=ifelse(psit==""| lep==""|result!="p", psit, 
                   paste0(lep, psit, sep= "_")))

Upvotes: 2

Views: 50

Answers (2)

akrun
akrun

Reputation: 887108

Here is an option with data.table to assign in place

library(data.table)
setDT(df)[result == "p" & lep !="" & psit != "", paste := paste(lep, psit, sep="_")][]

Upvotes: 0

Ronak Shah
Ronak Shah

Reputation: 388982

As @thelatemail suggested the condition seems to be mixed up. Also I don't think you need a group_by here.

library(dplyr)
df %>%
   mutate(paste = ifelse(result == "p" & (lep != "" & psit != ""), 
                  paste(lep, psit, sep = "_"), ""))

#  result lep  psit paste
#1      p   A    B   A_B
#2      p   C    D   C_D
#3      p   E    F   E_F
#4      d   F    G      
#5      d   G    H      
#6      p

Upvotes: 1

Related Questions