How do I get a proportion of each row?-CodePudding

I have a large dataset over 10 000 records. The data looks like this

Id  Ap   Ak    Al   Aj
602 1    0     1    1
603 0    1     1    1
603 1    1     1    0

Some of the Ids appear more than once. How do I to get the proportion for each row? I have tried : prop.table(table(rep$id), margin = 1)* 100, this just returns 1s under each id.

CodePudding user response：

If we want the 'proportions' for 'Id', extend the proportions to match the names with the column value of 'Id' after converting to character

df1$proportion <- 100 * proportions(table(df1$Id))[as.character(df1$Id)]

-output

> df1
   Id Ap Ak Al Aj proportion
1 602  1  0  1  1   33.33333
2 603  0  1  1  1   66.66667
3 603  1  1  1  0   66.66667

data

df1 <- structure(list(Id = c(602L, 603L, 603L), Ap = c(1L, 0L, 1L), 
    Ak = c(0L, 1L, 1L), Al = c(1L, 1L, 1L), Aj = c(1L, 1L, 0L
    )), class = "data.frame", row.names = c(NA, -3L))

CodePudding user response：

Using dplyr

library(dplyr)

df1 %>% 
  add_count(Id) %>% 
  mutate(prop = n / nrow(.), 
         .keep = "unused")

Output

   Id Ap Ak Al Aj      prop
1 602  1  0  1  1 0.3333333
2 603  0  1  1  1 0.6666667
3 603  1  1  1  0 0.6666667