Report-Sensitive Spot-Checking in Peer-Grading Systems
Peer grading systems make large courses more scalable, provide students with faster and more detailed feedback, and help students to learn by thinking critically about the work of others. A key obstacle to the broader adoption of peer grading is motivating students to provide accurate grades. To incentivize accurate grading, previous work has considered mechanisms that spot-check each submission (i.e., randomly compare it to a TA grade) with a fixed probability. In this work, we introduce a mechanism that spot checks students in a way that depends on the grades they report, providing the same incentives at lower costs than the fixed-probability mechanism. We also show, surprisingly, that TA workload is reduced by choosing not to spot check some students even when TA assessments are available.
READ FULL TEXT