Tuned Models of Peer Assessment in MOOCs

TL;DR

This paper introduces algorithms to improve peer grading accuracy in MOOCs by estimating and correcting grader biases and reliabilities, significantly enhancing grading quality on large-scale real data.

Contribution

It develops novel algorithms for bias and reliability correction in peer assessment, improving accuracy and enabling smarter grader assignment in massive online courses.

Findings

Significant accuracy improvement in peer grading with proposed algorithms

Large-scale analysis of 63,199 peer grades from Coursera

Relation of grader biases to student engagement and performance

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.