CODEWITHSUNDEEP
r
David Z
David Z

Reputation: 7061

How to generate a proportion table based on count table in R?

Suppose I have a count table such like:

library(tidyverse)
set.seed(123)
df<-data.frame(x=letters[1:3],
                y=sample(1:10, 3),
                z=sample(3:20, 3))%>%
    mutate(tot=rowSums(select(., -1)))
df
  x  y  z tot
1 a  3 20  23
2 b 10 13  23
3 c  2  7   9

How to quickly generate a proportion table like

  x     y     z tot
1 a 13.04 86.96 100
2 b 43.48 56.52 100
3 c 22.22 77.78 100

Upvotes: 1

Views: 126

Answers (2)

AntoniosK
AntoniosK

Reputation: 16121

You can use:

df %>% mutate_if(is.numeric, ~./tot*100)

#   x        y        z tot
# 1 a 13.04348 86.95652 100
# 2 b 43.47826 56.52174 100
# 3 c 22.22222 77.77778 100

which will update only your numeric variables, wherever they exist in your database.

Upvotes: 1

Henry Navarro
Henry Navarro

Reputation: 953

Is this what you want?

Your code

library(tidyverse)
set.seed(123)
df<-data.frame(x=letters[1:3],
               y=sample(1:10, 3),
               z=sample(3:20, 3))%>%
  mutate(tot=rowSums(select(., -1)))
df
  x  y  z tot
1 a  3 20  23
2 b 10 13  23
3 c  2  7   9

Proportion dataframe

proportion_df<-cbind(x=df$x,df[,-1] / df$tot * 100)
proportion_df
  x        y        z tot
1 a 13.04348 86.95652 100
2 b 43.47826 56.52174 100
3 c 22.22222 77.77778 100

Upvotes: 0

Related Questions