The 2nd CSEDM Data Challenge, in cooperation with SoLAR is happening now!

We invite you to participate in the 2nd CSEDM Data Challenge, an event held in cooperation with SoLAR! Participants will compete to develop new modeling techniques to predict students' outcomes in CS/programming classrooms, based on their code submissions to prior programming problems.

Overview & Dataset: The challenge is designed to bring together researchers from across the disciplines of Computing Education, Educational Data Mining, AI in Education and Learning Analytics to solve a challenge on a common dataset. Specifically, researchers will use log data of students' programming behavior to predict their future performance on programming tasks (Track 1) and a final exam (Track 2). Our dataset includes 130,000+ code submissions across 50 coding problems, from 800+ students.

Prizes: Entrants will compete for a $1,500 prize pool (split among the 2 Tracks), thanks to support from the Prof. Ram Kumar Memorial Foundation through IEDMS. Winners of the competition will also be invited to submit their approaches and results to a pending special journal issue on CSEDM (to be announced soon).

To get started:

* Go to our website for details, datasets and submission:

* Join our Piazza forum to ask questions, find teammates and get updates/announcements: (access code csedmdc2021).

* Use our starter code to dive right into the competition.

New to Student Modeling or CS Education? No problem - one of our goals is to bring together researchers with different types of expertise. Use the Piazza forum (access code csedmdc2021) to meet other researchers and find a team that can benefit from your research strengths..

About the Challenge: The CSEDM data challenge was launched as part of the 5th CSEDM Workshop. It is held in-cooperation with the International Educational Data Mining Society (IEDMS), the Society for Learning Analytics Research (SoLAR), and other societies are currently in the process of adding their support.


Tracks & Timeline: The Data Challenge has two main phases, as outlined here, which will start on October 1st, and a practice phase currently open for submission:

* Practice (open now until Sep 30, 2021): Get to know the data and submission system.

* Cross-semester prediction (until Feb 06, 2022): Train on one semester; predict on another.

* Within-semester prediction (until May 01, 2022): Train and predict on the same semester.


For any questions about the 2nd CSEDM Data Challenge, please join our Piazza forum (access code csedmdc2021) or contact Thomas Price <

SoLAR In-Cooperation

Are you hosting an event and would like In-Cooperation from SoLAR? Please review our Events In-Cooperation information on our website or contact our Events Working Group Chair, Isabel Hilliger Carrasco at