How to use “summarise” from dplyr with dynamic column names?

3
votes

I am summarizing group means from a table using the summarize function from the dplyr package in R. I would like to do this dynamically, using a column name string stored in another variable.

The following is the "normal" way and it works, of course:

myTibble <- group_by( iris, Species)
summarise( myTibble, avg = mean( Sepal.Length))

# A tibble: 3 x 2
  Species     avg
  <fct>      <dbl>
1 setosa      5.01
2 versicolor  5.94
3 virginica   6.59

However, I would like to do something like this instead:

myTibble <- group_by( iris, Species)
colOfInterest <- "Sepal.Length"
summarise( myTibble, avg = mean( colOfInterest))

I've read the Programming with dplyr page, and I've tried a bunch of combinations of quo, enquo, !!, .dots=(...), etc., but I haven't figured out the right way to do it yet.

I'm also aware of this answer, but, 1) when I use the standard-evaluation function standardise_, R tells me that it's depreciated, and 2) that answer doesn't seem elegant at all. So, is there a good, easy way to do this?

Thank you!

2
rdplyrsummarize

2 Answers

6
votes

1) Use !!sym(...) like this:

colOfInterest <- "Sepal.Length"
iris %>% 
  group_by(Species) %>%
  summarize(avg = mean(!!sym(colOfInterest))) %>%
  ungroup

giving:

# A tibble: 3 x 2
  Species      avg
  <fct>      <dbl>
1 setosa      5.01
2 versicolor  5.94
3 virginica   6.59

2) A second approach is:

colOfInterest <- "Sepal.Length"
iris %>% 
  group_by(Species) %>%
  summarize(avg = mean(.data[[colOfInterest]])) %>%
  ungroup

Of course this is straight forward in base R:

aggregate(list(avg = iris[[colOfInterest]]), iris["Species"], mean)
2
votes

Another solution:

iris %>% 
  group_by(Species) %>% 
  summarise_at(vars("Sepal.Length"), mean) %>%
  ungroup()

# A tibble: 3 x 2
  Species    Sepal.Length
  <fct>             <dbl>
1 setosa             5.01
2 versicolor         5.94
3 virginica          6.59