Perusall Exchange 2022 highlights

The Perusall Exchange.....

......is an asynchronous conference held entirely within the Perusall platform, highlighting recent research in Social Learning and Annotation. 

Affordances of asynchronous participation:

• attend the talks you want at your own time
• speed up the pace of the talk
• annotate at will to interact with speaker (who is very likely to reply back)

My personal favorites from that conference, were talks by:

Raoul Mulder, University of Melbourne. Inter-Rater Agreement between Perusall’s Automated Grading Algorithm and Expert Instructors in an Upper-Level Undergraduate University Course

Why is it important?

Students may hold some skepticism regarding their output on an annotation platform is being graded by algorithm. However, they may not appreciate the variability that can occur between human graders (inter-rater reliability), or the variability that can occur even with a solitary grader (intra-rater reliability). Are the human graders more or less harsh than the Perusall algorithm? Do humans have more agreement with each other, or with the Perusall algorithm? Watch on......   

You can also see Raoul's RPub on: Analysis of Perusall scoring agreement

Kristin Peace, MacEwan University. Social Annotation to Speed Experimenting: Using Perusall to Scaffold Experimental Skills 

Why is it important?

As more teachers get involved with collaborative/social annotation approaches, they are contemplating the important question of how to get better annotation output from their students. Via scaffolding, how can a teacher draw out more critical and less surface treatment of a topic to be annotated by their students? What is your grading criteria? Have you experimented with annotation group sizes -- dyads, triads? This talk also promotes the idea of annotation as a springboard to better experimental ideas.