Peer Review Made Simple: a Professor’s Experience with Peerceptiv
Peerceptiv is thrilled to share a thoughtful blog post from Jon Baarsch, an instructor in the Department of Computer Science & Engineering at the University of Central Arkansas, about his experiences using peer review in Peerceptiv this fall. Read the original post from Jon below or on his blog:
I recently discovered a tool that has completely transformed the way I teach.
Peerceptiv is an online tool for leveraging and managing peer-learning in the classroom (online or face to face). This fall term, I have been using it to get students to review and assess each other’s work, and I could not imagine going back to my old methods.
As an instructor of writing and programming, I have always hated the traditional grading process because it emphasizes and reinforces the power dynamic between teacher and student, with the teacher being the knowing, active, thinking source of truth, and the students (or their work) as the passive subjects being cut up and dissected. This dynamic is antithetical to my objective: to empower my students to analyze, think critically, and evaluate themselves.
However, peer-review has hitherto been difficult to implement effectively. The logistics are clumsy. Even with shared documents and “Track Changes,” it is a challenge to monitor the feedback students give to each other to ensure that it is helpful.
With Peerceptiv, I can manage how they review each other—providing the criteria and critical questions they use for their evaluations. The reviewing process is anonymous, so students don’t know who they review or by whom they are reviewed (reducing social pressures). The peer-reviews are themselves evaluated by the reviewed students in terms of helpfulness and accuracy. The accuracy measure comes jointly from similarity scores to other reviews and to the instructor’s reviews which are considered by the system to be completely accurate.
Say I evaluate 10 assignments of 40. My evaluations become a baseline, and the students whose evaluations are like mine will have a higher accuracy measure than those that differ. Consequently, they are weighted more heavily when calculating the submission’s grade. Thus, I don’t need to evaluate all the assignments to influence all of the grades.
The application manages the logistics: Peerceptiv assigns the reviewers, transmits the submissions and reviews, and connects directly with the Blackboard (or other LMS) gradebook. The process is completely transparent to the instructor. I can see exactly what the students are saying to each other and when, and all communication related to any student—the submission, the peer-reviews, the feedback given back on the reviews, and an optional reflection—is centralized in a single very easy-to-use site.
Peerceptiv emphasizes the critical process of production. Though all weights are editable, I like the defaults: 40% of a student’s grade is based on the submission itself, as graded by peers and instructor. 40% is based on the reviews the students give each other, split between helpfulness, which is determined by the people reviewed, and accuracy which is determined by similarity scores. The final 20% is based on participation in the process, turning in the assignments, performing the reviews, and providing feedback on the reviews received.
Rather than submitting themselves passively to the vagaries of my red pen, the students sharpen their own wits and learn to regulate themselves.
– Jon Baarsch, Department of Computer Science & Engineering
A huge thank you to Jon for sharing his experience with us and for his dedication to fostering self-regulated learning and critical thinking in his students. Peerceptiv is excited to support educators like Jon in fostering self-directed learning, creating transparent evaluation processes, and building a collaborative classroom culture. To find out more information about Perceptiv or get started, contact info@peerceptiv.com.