In a previous post I discussed the need for peer assessment in a course the size of Coursera (83,000 students). Daphne Koller, co-founder of Coursera, had presented some research on how the evaluation of peers closely matched the evaluation of teachers in her TED talk this summer.
Here's my contribution to this research based on a very small sample size - namely the evaluation of my latest Coursera exercise by my peers and by me. It's a simple scoring system: three categories of scoring, for Argument, Evidence, and Exposition, on each of which you can receive a score between 0 and 3. In each of the three categories, my own self-evaluation and that of my peers matched identically. I got back worthwhile feedback, that was completely appropriate, clearly stated, and right on target. Sweet.
I thoroughly expect that massive amounts of data are being collected on everything about this course. I'll be watching for papers assessing how this kind of evaluation works.
Evaluators are invited to provide their initials and where they are from. Two of the students who evaluated me self-identified as being from Chile and Germany. It's a truly international student cohort.