Home dplyr function for measuring mean and proportion of a particular value in different columns

Questions

dplyr function for measuring mean and proportion of a particular value in different columns

March 13, 2023

I have a dataset, df, and I wanted to measure some column-wise indicators. I want to extract the proportion of instances of value 1 in my var4 column, and also measure the mean of var3, and finally return them as text. I need to do that by creating a function and I did try with my basic understanding of R and dplyr, but I got errors. I would appreciate if anyone could let me know a better way to formulate my function.

library(dplyr)

set.seed(1)
ch   <- sample(LETTERS[1:5], size = 20, replace = TRUE)
cls  <- sample(c("CLASS ONE", "CLASS TWO"), size = 20, replace = TRUE)
var1 <- sample(1:20, size = 20, replace = TRUE)
var2 <- sample(20:40, size = 20, replace = TRUE)
var3 <- sample(40:60, size = 20, replace = TRUE)
var4 <- sample(c(0,1), size = 20, replace = TRUE)

df <- data.frame(ch, cls, var1, var2, var3, var4, stringsAsFactors = TRUE)

funcA <- function(col1, col2) {
    # for measuring percentage of col1== 1 in length of col1 
    select1   =  df %>% select(col1) %>% filter(col1 == 1)
    prop1     =  (length(selected$col1)/length(df$col1))*100
    # for measuring median of col2
    select2   = df %>% select(col2) 
    avg       = mean(select2)
    # pasting them in texts
    paste("The percentage of value 1 in", "col1", "is: ", prop1)
    paste("The mean value of", "col2", "is: ", avg)
}


funcA(var4, var3)

>Solution :

Here’s an option that also uses glue:

library(dplyr)
library(glue)

set.seed(1)
ch   <- sample(LETTERS[1:5], size = 20, replace = TRUE)
cls  <- sample(c("CLASS ONE", "CLASS TWO"), size = 20, replace = TRUE)
var1 <- sample(1:20, size = 20, replace = TRUE)
var2 <- sample(20:40, size = 20, replace = TRUE)
var3 <- sample(40:60, size = 20, replace = TRUE)
var4 <- sample(c(0,1), size = 20, replace = TRUE)

df <- data.frame(ch, cls, var1, var2, var3, var4, stringsAsFactors = TRUE)

funcA <- function(col1, col2) {
  
  select1 <- df %>% 
    reframe(
      prop = prop.table(table({{col1}})),
      avg = mean({{col2}})
    )
  
  out <- list(
    prop = paste0("The percentage of value 1 in ", as.list(match.call())[[2]], " is: ", unique(select1$prop)),
    percent = paste0("The mean value of ", as.list(match.call())[[3]], " is: ", unique(select1$avg))
  )
  
  glue::glue_collapse(out, sep = "\n\n")
  
}

funcA(var4, var3)
#> The percentage of value 1 in var4 is: 0.5
#> 
#> The mean value of var3 is: 51.1