R Select rows based on condition in another column

Question

How can I select all the rows of those IDs, that have at least one B as Item?

df<-cbind(c(1,1,1,1,1,2,2,2,2,3,3),c("A","A","B","A","A","B","A","A","B","A","A"))
colnames(df)<-cbind("ID","ITEM")
df

`library(dplyr);df %>% group_by(ID) %>% filter(any(ITEM == "B")) ` — Ronak Shah, Jul 27 '18 at 12:25

score 1 · Answer 1 · answered Jul 27 '18 at 12:17

1

Not sure if this is what you mean, but:

df[df[, 2]=="B", ]

answered Jul 27 '18 at 12:17

NewUser

thx, but please read my specification:I select all the rows of those IDs, that have at least one B as Item, for this example it means: give me all rows for ID 1 and ID 2 (but no rows for ID 3, because ID 3 has no B in any Item). your query just results in all rows with a B. – flozygy Jul 27 '18 at 12:41

score 0 · Answer 2 · answered Jul 27 '18 at 12:20

0

Its easy if you construct a data frame rather than working with a list.

library(dplyr)
df<-data.frame(ID = c(1,1,1,1,1,2,2,2,2,3,3),ITEM=c("A","A","B","A","A","B","A","A","B","A","A"));

df %>% filter(ITEM == "B");

answered Jul 27 '18 at 12:20

MSW Data

thx, but please read my specification: select all the rows of those IDs, that have at least one B as Item, for this example it means: give me all rows for ID 1 and ID 2 (but no rows for ID 3, because ID 3 has no B in any Item). your query just results in all rows with a B. – flozygy Jul 27 '18 at 12:55
right answer is given by Ronak Shah – flozygy Jul 27 '18 at 12:55

2 Answers2