Auto-Peer: A Computational Tool Designed to Provide Automated Feedback
​
Team leader:
Philip McCarthy (American University of Sharjah)
​
Lead Programmer:
Ayah Al-Harthy
​
Collaborators:
Nicholas D. Duran (Arizona State University)
Anuja Thomas (Arizona State University)
Elena Shpit (Tomsk State University of Control Systems and Radioelectronics)
​
Auto-Peer
​
Research has established the value of student peer reviewing; however, students appear to seldom have the time or belief to make best use of the practice. An automated peer review system is therefore useful and necessary, as it provides students with limitless assessment and assistance.
​
Auto-Peer is a free, 24/7-available computational peer review tool. Auto-Peer uses a combination of basic and sophisticated approaches to 1) guide students towards improved writing skills through analysis and feedback; and 2) allow instructors insight into student misconceptions.
Auto-Peer identifies a wide range of student-writing issues, requiring students to either 1) modify their text and/or 2) justify their writing choices. Following the Auto-Peer checks, instructors receive all student-generated responses to the automated review. Auto-Peer is an important development because it provides students with on-demand peer reviewing that aims to improve writing excellence; it allows instructors to reallocate assessment time for providing interventions on more nuanced aspects of the paper; and it presents researchers with opportunities for conducting a wide array of studies on peer reviewing methods, approaches, and techniques.
​
Auto-Peer is currently in a testing phase at the American University of Sharjah. General release to students is scheduled for fall, 2020. Broader release is scheduled for fall, 2021.
Note: The website is currently under development (14/02/2020).