Tuned Models of Peer Assessment in MOOCs

TLDR

Peer grading is essential for scaling complex assignments in MOOCs, but it often fails to match expert accuracy. The paper develops algorithms to estimate and correct grader biases and reliabilities, aiming to improve peer grading accuracy. The authors relate grader biases and reliabilities to student engagement, performance, and commenting style. The model significantly improves accuracy on 63,199 Coursera HCI grades and enables smarter grader assignment.

Abstract

In massive open online courses (MOOCs), peer grading serves as a critical tool for scaling the grading of complex, open-ended assignments to courses with tens or hundreds of thousands of students. But despite promising initial trials, it does not always deliver accurate results compared to human experts. In this paper, we develop algorithms for estimating and correcting for grader biases and reliabilities, showing significant improvement in peer grading accuracy on real data with 63,199 peer grades from Coursera's HCI course offerings --- the largest peer grading networks analysed to date. We relate grader biases and reliabilities to other student factors such as student engagement, performance as well as commenting style. We also show that our model can lead to more intelligent assignment of graders to gradees.