Who am I?

Workshop outline

What to expect during this workshop

What to expect during this workshop

I don’t always follow my own advice…

Workshop resources

Course website: nrennie.rbind.io/training-better-r-code

What’s this session about?

Who are you?

• You’re (reasonably) comfortable writing R code to do some analysis

But…

• Your scripts are getting longer and longer
• There are R scripts everywhere
• You don’t want other people to see the code you’ve written because it’s all a bit of a mess!

Why do we care?

• Writing code that is readable and understandable is something that future you will be grateful for.

• Writing code that is readable and understandable is something that other people will be grateful for.

  • Sharing code prevents duplicating work.
  • It makes work easier to replicate.
  • Some journals may require analysis code to be shared.

Problem: You store data in a different place

data <- read_csv("C:/Users/username/OneDrive/R code/LOS.csv")

setwd("C:/Users/username/OneDrive/R code")

Where do we put files?

• Let’s make our own lives easier by keeping our files organised!

• Organised files, also makes it easier for R to find our (e.g. data) files.

• An easy way to do this is using an R project

  • Use a new R project for each analysis project
  • Double click on the project file to open RStudio with your files in the right place.

R projects

R Projects are a special type of file with a .Rproj extension that makes it easier for you to keep all of the data, code, and images for a project in one place.

Open up RStudio, then click File –> New Project –> New Directory – New Project.

• Type in the name that you want to call your new folder e.g. R Workshop. Then use Browse to select where on your computer you want to make the folder. IMPORTANT: remember where this is!

• Finally, click Create Project. Your new folder will be created and opened in RStudio - sometimes it can take a couple of minutes.

Project folders

R
│   messy_example_script.R
│   R scripts for other project 1.R
│   R scripts for other project 2.R
│   R scripts for other project 3.R

project folder
│   messy_example_script.R
|   data.csv
|   project_name.Rproj

Problem: Objects don’t seem to exist

Where do you write your code?

• Console?

• .R Script?

• R Markdown file

Environments

Preventing bad habits

But it takes a long time to create an object…

• Create a script that creates the object.

• Save the object as an RDS file.

• In further analysis, load the RDS object.

Live Demo!

• Creating a project folder

• Downloading data

• Project settings

Exercise 1

1. Create an R project for today’s workshop - name it something sensible!

2. Download the messy_exercises_script.R script from nrennie.rbind.io/training-better-r-code/exercises.html and the hypoxia.csv data.

3. Add those files to your R project in a sensible way.

4. Edit the Global / Project options appropriately.

5. Edit the script to use relative file paths.

Adding comments

Adding comments

starwars |> 
  summarise(
    mean_height = mean(height, na.rm = TRUE), # calc height mean 
    sd_height = sd(height, na.rm = TRUE) # calc height sd
  )

# calculate summary statistics and 
# remove NA values as missing `height`
# values are also missing `mass`
starwars |> 
  summarise(
    mean_height = mean(height, na.rm = TRUE),
    sd_height = sd(height, na.rm = TRUE)
  )

Sections and subsections

# Load data ---------------------------------------

## Geospatial files -------------------------------

## Population files -------------------------------

Structuring scripts

• All library() calls at the start - only load the packages you actually need!

• Don’t add install.packages() to a script - run it in the console!

• Break it down in big steps - give sections useful names!

• Sections aren’t the only things that should be well-named - variables and functions too!

Live Demo!

• Restructuring scripts

• Adding sections

• Renaming variables

• Namespacing?

Exercise 2

1. Reorganise the messy R script by adding sections and subsections.

2. Edit the comments in the document to make them more useful.

3. Rename variables and functions if you think they need to be renamed.

Code style

This code runs without errors but…

starwars |> filter(height>100) |>select(eye_color, mass)|> group_by(eye_color) |>summarise(mean_mass =mean(mass, na.rm = T))

starwars |> 
  filter(height > 100) |> 
  select(eye_color, mass) |> 
  group_by(eye_color) |> 
  summarise(mean_mass = mean(mass, na.rm = TRUE))

Linting

Linting - analysing source code for:

• stylistic issues e.g. x<-3 vs x <- 3
• common errors e.g. mean(x, na.rm = T, na.rm = F)
• missing packages
• …

{lintr}

Run lintr::lint("file.R"):

Styling

{lintr} tells you what’s wrong, but doesn’t fix it.

Live Demo!

• Linting code

• Styling code

• Adding keyboard shortcuts

Exercise 3

1. Install and load the lintr and styler packages if you don’t already use them.

2. Run lint() on the messy R script. Do you understand all of the messages?

3. Run style_file(). What has changed in your script?

4. Re-run lint() on the script. Have all of the issues been fixed? If not, manually implement changes to the file.

5. Bonus: Add an RStudio keyboard shortcut for style_active_file().

Breaking up a single file

Imagine a directory structure like this:

project
│   Rscript.R

that contains all of the code for your analysis.

Breaking up a single file

Multiple files

project
│   data wrangling.R
│   load data.R
│   modelling.R
│   packages.R
│   plots.R
│   plots2.R

project
│   00_packages.R
│   01_load_data.R
│   02_data_wrangling.R
│   03_exploratory_plots.R
│   04_modelling.R
│   05_final_plots_tables.R

Multiple files

Multiple folders

Often, you don’t just have R code for a project…

project
│   00_packages.R
│   01_load_data.R
│   02_data_wrangling.R
│   03_exploratory_plots.R
│   04_modelling.R
│   05_final_plots_tables.R
│   data.csv
│   residuals.png
│   outcome_by_age.png
│   outcome_by_occupation.png

Multiple folders

… so don’t just organise your R code!

project
│   project.Rproj
│   README.md
└───data
│   │   data.csv
└───plots
│   │   residuals.png
│   │   outcome_by_age.png
│   │   outcome_by_occupation.png
└───R
│   │   00_packages.R
│   │   01_load_data.R
│   │   02_data_wrangling.R
│   │   03_exploratory_plots.R
│   │   04_modelling.R
│   │   05_final_plots_tables.R

R script dependencies

project
└───R
│   │   00_packages.R
│   │   01_load_data.R
│   │   02_data_wrangling.R
│   │   03_exploratory_plots.R
│   │   04_modelling.R
│   │   05_final_plots_tables.R

Live Demo!

• Creating multiple folders

• Creating multiple files

• Dependencies of files

Exercise 4

1. Re-organise your R project folder with multiple sub-directories.

2. Split your messy R script into multiple files, that are appropriately named.

3. What is the order and dependencies of each script?

Problem: your coauthor uses a different package version

Problem: your coauthor uses a different R version

Problem: your coauthor uses a different R version

Problem: copying and pasting values into a paper

Problem: forgetting which scripts to re-run

Problem: I need help with my code

reprex: reproducible example (small, rigorous, self-contained example)

• Makes the problem more specific

• Makes the problem reproducible

• Makes you think more clearly about programming

Problem: I need help with my code

…and I can’t share data

• Are there built-in data sets that you can use to reproduce the problem e.g. mtcars?

• Can you make a small, synthetic data set?

small_example <- tibble::tribble(
  ~A, ~B,
  1, 4.7,
  5, 2.2, 
  2, 9.8
)

Live Demo!

• How to make a reprex

Exercise 5

1. The following code doesn’t work. Imagine you are not allowed to share the hypoxia data with anyone else. Build a reprex that you could share with someone else.

library(ggplot2)
ggplot(hypoxia) +
  geom_col(aes(AHI, fill = Smoking))

Workshop resources

Course website: nrennie.rbind.io/training-better-r-code

Feedback

Feedback form: forms.gle/nCutoyKy1GbnEpKq5