Datathons are a type of competition where teams work together to solve a common data-oriented problem in a limited time period over a day or weekend. The final product of a Datathon is often a report alongside of analysis code that will be judged by a panel of experts. The main goal of the Psychometric Society’s Datathon is to bring together a diverse group of students and practitioners to explore an education-related data set.
The first annual Datathon is brought to you by Math Garden in partnership with Professor Han van der Maas (University of Amsterdam), Abe Hofman (University of Amsterdam), and Joost Kruis (Prowise Learn). Math Garden is currently maintained by Prowise Learn and is a popular adaptive learning platform for primary school children. It consists of multiple math games, collecting collecting about 1.5 million daily responses. You can make a demo account here or read more about the psychometrics of the system here.
Over the 10 day Datathon period, teams will work together to answer a question of their own design that targets the problem statement provided by Math Garden. There are often multiple angles to pursue the problem statement, and taking a unique approach can often benefit in judging. In the end stages, a strong report will focus on how you solve the problem in an interpretable manner and the direct impacts your work has on real world applications.