use column from function input for group_by variable when using dtplyr

user_8772229 0
Viewed 108

When trying to summarise columns by group using dtplyr, grouping seems to not be working. Since the group variable is an input of my function, I tried using group_by_ only to receive an error message.

Data:

df <- data.frame(
  id=c(1,1,1,2,2,2,3,3,3),
  year=c(2014, 2015, 2016, 2015, 2015, 2016, NA, NA, 2016),
  code=c(1,2,2, 1,2,3, 3,4,5),
  dv1=1:9,
  dv2=2:10
) %>% as.data.table()

cols <- c("dv1", "dv2")

> df
   id year code dv1 d2
1:  1 2014    1   1  2
2:  1 2015    2   2  3
3:  1 2016    2   3  4
4:  2 2015    1   4  5
5:  2 2015    2   5  6
6:  2 2016    3   6  7
7:  3   NA    3   7  8
8:  3   NA    4   8  9
9:  3 2016    5   9 10

Function:

for(i in seq_along(columns)) {
 sub1 <- df %>% 
   select("id", columns[i], group) %>%
   group_by(group) %>%
   summarise(mean=mean(.data[[columns[i]]], na.rm=T), sd=sd(.data[[columns[i]]], na.rm=T)) %>%
   ungroup() %>%
   as_tibble() 
 print(sub1)
}
}

test(data=df, columns=cols, group="year")

# A tibble: 1 x 3
  group  mean    sd
  <chr> <dbl> <dbl>
1 year      5  2.74
# A tibble: 1 x 3
  group  mean    sd
  <chr> <dbl> <dbl>
1 year      6  2.74
dtplyr dplyr r
2 Answers

Here's a reprex that's similar to what you're looking for I think. It's not the all time sexiest solution, but it'll work:

library(tidyverse)
f <- function(grouping_var) {
    iris %>%
        group_by(!!sym(grouping_var)) %>%
        summarize(N = n())
}

f('Species')
#> # A tibble: 3 x 2
#>   Species        N
#> * <fct>      <int>
#> 1 setosa        50
#> 2 versicolor    50
#> 3 virginica     50

Created on 2021-03-17 by the reprex package (v1.0.0)

user_154030000

We can use .data

f<- function(dat, grouping_var) {
     dat %>%
         group_by(.data[[grouping_var]]) %>%
         summarise(N = n())
  }

-testing

f(iris, "Species")
# A tibble: 3 x 2
#  Species        N
#* <fct>      <int>
#1 setosa        50
#2 versicolor    50
#3 virginica     50

using the OP's example data

library(purrr)
map(cols, dat = df, .f = f)
user_37322710
Related