Do data competitions improve learning: A study on student performance, engagement, and experience with Kaggle InClass data challenges

Abstract

In the November CAUSE/Journal of Statistics and Data Science Education webinar series, we have invited the authors of this recently published paper to share their experiences in running data competitions as part of classes on statistical learning. Kaggle is a data modeling competition service, where participants compete to build a model with lower predictive error than other participants. Several years ago Kaggle released a simplified service that is ideal for instructors to run competitions in a classroom setting. This webinar describes the results of an experiment to determine if participating in a predictive modeling competition enhances learning. The evidence suggests it does. In addition, students were surveyed to examine if the competition improved engagement and interest in the class. The authors will also discuss the main issues to consider when setting up a data competition in a class, including the technical aspects of using the Kaggle InClass platform.

Julia Polak is a lecturer in Statistics at the University of Melbourne. She has a broad range of research interests including nonparametric methods, forecasting and data visualisation. In addition, Julia has many years of experience in teaching statistics and data science for different audience.

Di Cook is a Professor in Econometrics and Business Statistics at Monash University in Melbourne. Her research is in the area of data visualisation, especially the visualisation of high-dimensional data using tours with low-dimensional projections, and projection pursuit. A current focus is on bridging the gap between exploratory graphics and statistical inference.

Syllabus: https://handbook.unimelb.edu.au/2017/subjects/mast90083