问题
I have data something similar to this:
User id Ranking Country
1 1 USA
2 3 AUS
3 1 USA
4 1 AUS
5 2 USA
and I would like to have the following results:
USA Ranking 1 = 2
USA Ranking 2 = 1
USA Ranking 3 = 0
AUS Ranking 1 = 1
AUS Ranking 2 = 0
AUS Ranking 3 = 1
How may I do this in R please?
回答1:
reshape2::melt(table(d$Country, factor(d$Ranking, 1:3)))
# Var1 Var2 value
#1 AUS 1 1
#2 USA 1 2
#3 AUS 2 0
#4 USA 2 0
#5 AUS 3 1
#6 USA 3 0
DATA
structure(list(User.id = 1:4, Ranking = c(1L, 3L, 1L, 1L), Country = c("USA",
"AUS", "USA", "AUS")), class = "data.frame", row.names = c(NA,
-4L))
回答2:
A dplyr approach. Pretty easy to alter if wanted.
library(dplyr)
df = data.frame(
UserID = seq(1:5),
Ranking = c(1, 3, 1, 1, 2),
Country = c("USA", "AUS", "USA", "AUS", "USA")
)
df_2 <- df %>%
group_by(Country, Ranking) %>%
summarise(Counts = n())
df_2
Country Ranking Counts
1 AUS 1 1
2 AUS 3 1
3 USA 1 2
4 USA 2 1
来源：https://stackoverflow.com/questions/58066131/count-how-many-fall-in-each-group-in-r