Reproducible Research with R

Preparation for the course

Each participants should prior to the course:

• Complete DataCamp courses Introduction to R and Intermediate R. If you have not yet received an invitation please contact us!
• Install in the newest(!) version:
  • R
  • RStudio
  • Git (this is for version control)
• Get accounts for:
  • Github You can have the academic version with Charité email

Important resources to use if you do not know how to proceed/need help:

• Cheat sheets on different packages like tidyverse, ggplot, etc.
• Stackoverflow: A community of nerds at your disposal. All beginners questions have been already posted so a simple search is enough!
• help function in R (put cursor on function name and press F1 or type ?function_name)
• at the end of each help text there is a so called vignette that will give an executable example so that you can experiment with the function and its outputs
• Contact us. We will explain rules for our (online) classes in the first session.

Course Program

We will meet on Tuesdays, 10am - 2pm, in presence in the conference room Domblick at the BIH (5th floor, Anna-Lousia-Karsch-Straße 2, 10178 Berlin). Each course part will consist of two short lectures on issues in reproducibility in research, an introduction to a specific topic in programming with R, and an exercise.

0. Pre course work: Course work on DataCamp (Introduction to R + Intermediate R)

1. Session 09.05.

Part 1 Preregistration, Functions in R
Exercise: Functions

Part 2 Data Management
Exercise: Vectorisation

Homework:
I Introduction to the Tidyverse on DataCamp
II Read Tutorial for Git/Github

2. Session 23.05.

Part 1 Git and Github
Exercise: Git and Github

Part 2 Tidyverse
Exercise: tidyverse
Read: https://doi.org/10.1371/journal.pone.0185195

Homework:
Datacamp course ggplot2

3. Session 06.06.

Part 1 Advanced Plotting
Exercise: ggplot

Part 2 Introduction Statistics with R Exercise: linear_models

Homework:
2 courses of your choice on DataCamp

4. Session 20.06.

Part 1 Linear Mixed Models
Exercise: linear_mixed_models

Part 2 Reproducible Research from Planning to Publication
Exercise: Open Peer Review

Technical challenges

We plan to conduct all sessions in presence. Please bring a laptop with Chrome browser installed. You can also install everything locally, but it is not certain whether we will have time to troubleshoot every computer.

In case this course will be taught online some challenges apply. In the first session, we will try to resolve as many as possible. We will use the Charité instance of MS teams as our platform for video meetings. For questions during the exercises, we have separate rooms. You can raise your hand and we will then invite you in one of the rooms.

All code will be provided on Github where also this document here resides. Additionally, an OSF project will contain all presentations and a link to the Github repository.

Each course part will begin with a 20-minute presentation on issues of reproducibility (see Course Program). After this, there will be a short introduction to R functionalities needed in the upcoming exercise. The exercise itself will be done on your computer, which is why it is essential for you to make sure you have installed and tested R and RStudio on your computer beforehand. If you experience persistent difficulties with this, please let us know. Please bear in mind that you need to install all libraries prior to using them.

install.packages("tidyverse")
library(tidyverse)

The installation has to be done only once. The library has to be loaded each time R is restarted. It is good practice to load the libraries at the start of the script where they are used.