Many courses running on the edx platform have their student learning data processed using the code edx2bigquery, and stored in Google Bigquery. This repository is a demonstration of how the data stored in BigQuery could be presented on an interactive dashboard that could allow instructors and researchers to visualize and analyze student learning data.
A secondary goal of this repo is to generate discussions about what learning-analytics quantities are easily available on the edX platform, and how instructors/researchers could use such quantities to improve/study student learning.
If you just want to play with the demo, go to http://198.199.87.145/demo.
If you are cloning/downloading the repository to build on top of it, we strongly recommend using a virutalenv. Follow the usual procedure to install the requirements, and then run the code by issuing the command "python run.py". Once the program is running, open a tab in your browser and type http://localhost:8050/ to access the dashboard.
-
This code is not suitable for touchscreen devices, according to our tests.
-
The code is NOT production-grade, and not intended to be used directly with the BigQuery data to interpret student learning. We are working on the code. Email anindyar@mit.edu if you'd like to be notified about the "new and improved" versions.
-
The sample dataset included here is representative but fake.
-
An earlier version of this demo (along with other useful jupyter notebooks) is available from Vita Lampietti's Github account.
-
Best way to find out what different quantities mean is to look through the edx2BigQuery repository. More information on edX tracking log definitions are available here.
Please email your comments/suggestions to anindyar@mit.edu