Loading data

library(tidyverse)

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──

## ✔ ggplot2 3.3.6     ✔ purrr   0.3.4
## ✔ tibble  3.1.8     ✔ dplyr   1.0.9
## ✔ tidyr   1.2.0     ✔ stringr 1.4.0
## ✔ readr   2.1.2     ✔ forcats 0.5.1

## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()

data <- iris %>%
  janitor::clean_names() 

head(data)

##   sepal_length sepal_width petal_length petal_width species
## 1          5.1         3.5          1.4         0.2  setosa
## 2          4.9         3.0          1.4         0.2  setosa
## 3          4.7         3.2          1.3         0.2  setosa
## 4          4.6         3.1          1.5         0.2  setosa
## 5          5.0         3.6          1.4         0.2  setosa
## 6          5.4         3.9          1.7         0.4  setosa

If you’re unfamiliar with the iris data, there are four species of iris included, with four attribute variables describing the plants. Which attribute contains the most variation across the different species? One reasonable way to answer this question would be to 1) find the means of each attribute across the species, and then 2) calculate the standard deviation of the means across each attribute. Let’s make a table containing all of this information.

Creating table

First I need to calculate the means for each variable.

data %>%
  group_by(species) %>%
  summarize(
    across(everything(), ~ mean(.x))
  )

## # A tibble: 3 × 5
##   species    sepal_length sepal_width petal_length petal_width
##   <fct>             <dbl>       <dbl>        <dbl>       <dbl>
## 1 setosa             5.01        3.43         1.46       0.246
## 2 versicolor         5.94        2.77         4.26       1.33 
## 3 virginica          6.59        2.97         5.55       2.03

Easy enough. Now, the tricky part: How can we include information about the standard deviations of these columns? The first thought would be to calculate these values and somehow append them as a separate row. This would work fine, but it would be a bit difficult to format successfully. An alternative approach is to include the standard deviations within the column labels themselves. Normally this would involve a few clunky steps, but with the ability to create local variables within a pipe, this turns out to be a pretty easy and concise task!

data %>%
  group_by(species) %>%
  summarize(
    across(everything(), ~ mean(.x))
  ) %>%
  {
  sds <- summarize(.,across(-1, ~ sd(.x))) %>% # calculating SDs of each column which are stored in a local variable, "sds".
    mutate(across(everything(), ~ round(.x,2))) # rounding SDs

  rename_with(., .cols = -1, ~ paste0(.x," (SD = ",sds[1,],")")) #renaming columns to include SD values
  } %>%
  gt::gt()

Pretty cool, huh? But what if you wanted to store the sds object for something later on? Easy: assign it to a global variable!

data %>%
  group_by(species) %>%
  summarize(
    across(everything(), ~ mean(.x))
  ) %>%
  {
  sds <- summarize(., across(-1, ~ sd(.x))) %>% # calculating SDs of each column which are stored in a local variable, "sds".
    mutate(across(everything(), ~ round(.x,2))) # rounding SDs

  SDs <<- sds # Creating global variable which will be saved in environment
  
  rename_with(., .cols = -1, ~ paste0(.x," (SD = ",sds[1,],")")) #renaming columns to include SD values
  } %>%
  gt::gt()

And here is the stored SDs object:

SDs

## # A tibble: 1 × 4
##   sepal_length sepal_width petal_length petal_width
##          <dbl>       <dbl>        <dbl>       <dbl>
## 1          0.8        0.34         2.09         0.9

There are plenty more useful applications beyond what I’ve shown here, but hopefully this example provides a glimpse into the potential power of using local variables within pipes.