Accelerometer Processing Workshop

Workshop Background

In this workshop, we’ll cover the how and why of processing accelerometer data using open-source tools. Specifically, we’ll focus on the popular GGIR package.

Agenda

We will cover:

  • The advantages (and some disadvantages) of using open-source processing methods
  • The basics of using R
  • Installing and running GGIR
  • Understanding the settings and options
  • Interpreting the output
  • Common issues and troubleshooting steps
  • Advanced options (e.g., day segment analysis)

Getting Setup

To particpate in the workshop, you’ll need three things:

  1. R installed on your machine
  2. RStudio (or another IDE like VS Code, if you prefer)
  3. The example files and scripts

Install R

We’ll be using Version 4.2.x for this workshop, but it should work for versions greater than 4. If you’re running an older version of R, now is a great time to upgrade.

To install R, navigate head to CRAN and follow the instructions to install the version that matches your operating system.

Install RStudio

RStudio is an Integrated Development Environment (IDE) - a way to interact with programming languages. RStudio is specifically designed with R in mind, so it’s a great place to start if you have never used R before.

You can install RStudio from here. Again, just follow the install steps they provide to get started.

Download Workshop Files

I’ve provided some example scripts to use during the workshop. To download these:

  1. Head to the GitHub Repository
  2. Click Code then Download ZIP
    Screenshot of GitHub Repo download button
  3. Unzip the files. You will only need the Workshop Files folder.

Run the Setup Script

You’ll also need to install a few packages, and download some example files. Fortunately, there’s a script to help:

  1. In your unzipped Workshop Files folder, double click the workshop_files.Rproj file. RStudio should open.
  2. On the right hand panel, click the 0_setup.R file.
  3. In the large left hand panel, click the Source button to run the script. It will take a few minutes.

If you have any issues with the above, just reach out.

Resources