Content adapted from Candace Savonen.
Content adapted from Candace Savonen.
Regularly cleaning your environment and trying your code again, can help ensure that your code is running as expected.
Occasionally we might forget to save a step of our code in our R Markdown file that we ran only in the console. This will help us figure that out.
Regularly checking if your file knits will help you spot a missing step or error earlier when you have less code to try to identify where your code might have gone wrong.
Image by Allison Horst.
Provide sufficient detail so that you can understand what you did and why.
# Taking a random sample of 100 individuals from the population # WITHOUT replacement samp_pop <- sample(100, replace = FALSE) # Then split them into two groups of 50 # a[x:xx] is the syntax for indexing a vector samp_pop1 <- samp_pop[1:50] samp_pop2 <- samp_pop[51:100]
Use set.seed()
: sets the starting state for the random number generator.
set.seed(123) sample(10)
[1] 3 10 2 8 6 9 1 7 5 4
set.seed(123) sample(10)
[1] 3 10 2 8 6 9 1 7 5 4
set.seed(456) sample(10)
[1] 5 3 6 10 4 9 1 2 8 7
Note that these are only pseudo random and the values are created doing calculations based on the given seed. Thus the same “random” values will be reproduced by everyone using the same seed with set.seed
.
Before:
After knit:
Go to the RStudio toolbar: Help > Cheat Sheets > R Markdown Cheat Sheet (which will download it)
Or Help > Cheat Sheets > R Markdown Reference Guide
Or check out the 🏠 Class Website! The Resources page has links to additional helpful cheat sheets.
Why is reproducibility so important?
A. It helps to ensure that your code is working consistently and it helps others understand what you did
B. It ensures that your code is correct
What is NOT a practice to improve the reproducibility of our work?
A. Using R Markdown files to describe what your code is doing
B. Using scripts instead of R Markdown files
C. Testing your code with R Markdown files or the run previous button
D. Regularly cleaning the environment
These are just some quick tips, for more information:
To help make your work more reproducible:
💻 Lab
Image by Gerd Altmann from Pixabay