CODEWITHSUNDEEP
rdplyrtidyversejanitor
NM_
NM_

Reputation: 1999

Calculate and append column totals of select columns in a dataframe

I have the following code for calculating certain quantities of interest, specifically the sum of the two right-most columns.

library(dplyr)
library(janitor)

m = c(0, 0.8, 2.3, 4.1, 2.1)
l = c(0.3, 0.8, 0.9, 0.75, 0.25)

mytable = data.frame(l, m)
rownames(mytable) = paste("Group", 1:5)

# Initial population
n0 = c(1,1,1,1,1)

mytable = mytable %>%
  mutate(lm = l * m) %>%
  mutate(n = n0) %>%
  mutate(offspring = lm * n) %>%
  adorn_totals("row")

This gives the following output:

> mytable
     l   m    lm n offspring
   0.3 0.0 0.000 1     0.000
   0.8 0.8 0.640 1     0.640
   0.9 2.3 2.070 1     2.070
  0.75 4.1 3.075 1     3.075
  0.25 2.1 0.525 1     0.525
 Total 9.3 6.310 5     6.310

I have the following issues:

Upvotes: 5

Views: 2282

Answers (2)

Sam Firke
Sam Firke

Reputation: 23014

To your first and third points: you can control which columns are totaled by specifying column names to the ... argument of adorn_totals(). Using ... requires specifying values for the other arguments, even if they're empty, thus the ,,,, below to accept the default values for those arguments.

The first column is skipped by default, as this is usually a group ID (like your rownames), but you can specify that it should be totaled.

Here is how you'd total the columns l, n, and offspring:

mytable %>%
  mutate(lm = l * m) %>%
  mutate(n = n0) %>%
  mutate(offspring = lm * n) %>%
  adorn_totals("row",,,,l, n, offspring)

Returns:

    l   m    lm n offspring
 0.30   0     0 1     0.000
 0.80 0.8  0.64 1     0.640
 0.90 2.3  2.07 1     2.070
 0.75 4.1 3.075 1     3.075
 0.25 2.1 0.525 1     0.525
 3.00   -     - 5     6.310

Along with the warning:

Because the first column was specified to be totaled, it does not contain the label 'Total' (or user-specified name) in the totals row

Upvotes: 7

akrun
akrun

Reputation: 887118

An option is to convert the columns other than the required columns to character class and then change it later. Regarding the row names, tibble doesn't allow for row names. We may need to create a column first with rownames_to_column

library(dplyr)
library(tibble)
library(janitor)
out <- mytable %>% 
    rownames_to_column('rn') %>%
    mutate(lm = l *m, n = n0, offspring = lm * n) %>% 
    mutate(across(-c(n, offspring), as.character)) %>% 
    adorn_totals('row', fill = NA) %>%
    type.convert(as.is = TRUE)

-output

> out
      rn    l   m    lm n offspring
 Group 1 0.30 0.0 0.000 1     0.000
 Group 2 0.80 0.8 0.640 1     0.640
 Group 3 0.90 2.3 2.070 1     2.070
 Group 4 0.75 4.1 3.075 1     3.075
 Group 5 0.25 2.1 0.525 1     0.525
   Total   NA  NA    NA 5     6.310
> str(out)
Classes ‘tabyl’ and 'data.frame':   6 obs. of  6 variables:
 $ rn       : chr  "Group 1" "Group 2" "Group 3" "Group 4" ...
 $ l        : num  0.3 0.8 0.9 0.75 0.25 NA
 $ m        : num  0 0.8 2.3 4.1 2.1 NA
 $ lm       : num  0 0.64 2.07 3.075 0.525 ...
 $ n        : int  1 1 1 1 1 5
 $ offspring: num  0 0.64 2.07 3.075 0.525 ...
 - attr(*, "core")='data.frame':    5 obs. of  6 variables:
  ..$ rn       : chr [1:5] "Group 1" "Group 2" "Group 3" "Group 4" ...
  ..$ l        : chr [1:5] "0.3" "0.8" "0.9" "0.75" ...
  ..$ m        : chr [1:5] "0" "0.8" "2.3" "4.1" ...
  ..$ lm       : chr [1:5] "0" "0.64" "2.07" "3.075" ...
  ..$ n        : num [1:5] 1 1 1 1 1
  ..$ offspring: num [1:5] 0 0.64 2.07 3.075 0.525
 - attr(*, "tabyl_type")= chr "two_way"
 - attr(*, "totals")= chr "row"

Upvotes: 2

Related Questions