R tip: Counting percentual function with curly-curly

The count function from dplyr package is one of my favorites for data wrangling.

I like to include a percent column to the function results.

penguins %>% 
  count(species, sort = T) %>% 
  mutate(percent = scales::percent(n/sum(n), scale = 100))

# A tibble: 3 × 3
  species       n percent
  <fct>     <int> <chr>  
1 Adelie      152 44.2%  
2 Gentoo      124 36.0%  
3 Chinstrap    68 19.8%

Whenever more than one column will be counted it is nice to write a small function for that.

count_percent <- function(df, a_column) {
  df %>% 
    count({{a_column}}) %>% 
    mutate(percent = scales::percent(n/sum(n), scale = 100))
}

penguins %>% 
  count_percent(species)

# A tibble: 3 × 3
  species       n percent
  <fct>     <int> <chr>  
1 Adelie      152 44.2%  
2 Chinstrap    68 19.8%  
3 Gentoo      124 36.0%

penguins %>% 
  count_percent(island)

# A tibble: 3 × 3
  island        n percent
  <fct>     <int> <chr>  
1 Biscoe      168 49%    
2 Dream       124 36%    
3 Torgersen    52 15%

Para esta função funcionar é preciso utilizar o operador curly-curly.

Citation

BibTeX citation:

@misc{abreu2020,
  author = {Abreu, Marcos},
  title = {R tip: Counting percentual function with curly-curly},
  date = {2020-10-10},
  url = {https://abreums.github.io/posts/2022-10-10-r-tip-counting-percentual-function-with-curly-curly/},
  langid = {pt-br}
}

For attribution, please cite this work as:

Abreu, Marcos. 2020. “R tip: Counting percentual function with curly-curly.” October 10, 2020. https://abreums.github.io/posts/2022-10-10-r-tip-counting-percentual-function-with-curly-curly/.