Remove group based on count of another column (R)

Question

I have a dataframe and I want to only keep the groups that have at least 2 cases with cars AND another output to keep groups will at least 2 cases of No car:

Group = c('a','a','a','b','b','b','c','c','c','c')
Car = c(1,1,0,0,0,0,1,0,0,0) # 1 = Have car, 0 = No car
df = data.frame(Group,Car)
df$Group = factor(df$Group)
df$Car = factor(df$Car)

   Group Car
1      a   1
2      a   1
3      a   0
4      b   0
5      b   0
6      b   0
7      c   1
8      c   0
9      c   0
10     c   0

Output should be:

Group Car
a      1
a      1
a      0

2nd output:

Group Car
b      0
b      0
b      0
c      1
c      0
c      0
c      0

I have a very huge dataset. Please help. Thanks!

Answer 1

For 1st case : groups that have at least 2 cases with cars

library(dplyr)

df %>%
  group_by(Group) %>%
  filter(sum(Car) > 1)

#   Group   Car
#   <fct> <dbl>
#1  a         1
#2  a         1
#3  a         0

Or base R ave

subset(df, ave(Car, Group, FUN = sum) > 1)

and with data.table

library(data.table)
setDT(df)[, if (sum(.SD) > 1) .SD, by = Group]

---

For 2nd case : groups with at least 2 cases of No car

df %>%
  group_by(Group) %>%
  filter(sum(Car == 0) > 1)


#  Group   Car
#  <fct>  <dbl>
#1  b       0
#2  b       0
#3  b       0
#4  c       1
#5  c       0
#6  c       0
#7  c       0

and with base R ave

subset(df, ave(Car == 0, Group, FUN = sum) > 1)

with data.table

setDT(df)[, if (sum(.SD == 0) > 1) .SD, by = Group]

data

Group = c('a','a','a','b','b','b','c','c','c','c')
Car = c(1,1,0,0,0,0,1,0,0,0) 
df = data.frame(Group,Car)

Answer 2

We can get both the datasets in a list with one step using split

lst1 <- split(df, df$Group %in% names(which(rowsum(df$Car, df$Group)[,1] >= 2)))
lst1
#$`FALSE`
#   Group Car
#4      b   0
#5      b   0
#6      b   0
#7      c   1
#8      c   0
#9      c   0
#10     c   0

#$`TRUE`
#  Group Car
#1     a   1
#2     a   1
#3     a   0

If we need to extract the list elements, use [[

lst1[[1]]
lst1[[2]]

data

Group <- c('a','a','a','b','b','b','c','c','c','c')
Car <- c(1,1,0,0,0,0,1,0,0,0) 
df <- data.frame(Group,Car)