Mentoring Teaching Assistants

This article is commenting oh how TA do not receive any "real" training before teaching a class or grading papers. Professor Samuel Gorovitz, professor of philosophy at Syracuse University, found this gap in philosophy and explains his 4 step method for fixing it.  This is the type of stuff I like to hear as someone applying to programs. Having professors who care this much seems important.





"Before allowing a TA to grade a student’s paper, I want to be sure the TA is sufficiently trained. Evaluating the written work of another is a demanding intellectual undertaking, especially when both substantive content and expository quality are under scrutiny. Few graduate students have had significant experience of this kind within their disciplines.

Here’s what I typically do (with small variations from course to course):

(1) Each TA must write the first essay that will later be assigned to the undergraduates, and submit that to me. I mark it up rigorously and return it. (Sometimes I get a different writing sample from them.)

(2) When that first essay is collected from the class, I gather a random set of perhaps 4-6 papers. I photocopy one, mark it in detail, and give each TA a copy of the marked paper to study. I also give each TA a copy of the remaining unmarked papers from the selected set. The TAs must mark these papers, guided by the model I provided.

(3) We meet and array the results. That is, for each paper, we put on the board the grade given by each TA. We then go through each paper, page by page, with a critique of the grading. So, for example, if TA Jones marks "non sequitur" in the margin of paragraph 4 on page 3 of paper B, but TA Lin has not written that, we discuss whether there is or is not a non sequitur here, and whether that comment belongs on the page. If it does, why did TA Lin miss it? If it does not, why did TA Jones write it? We do this for all the selected papers until we agree about what should be written on each paper and what the grade should be. This meeting can take 4-5 hours. The goal is to create such a solid commonality of understanding and approach that the grade a student gets will not depend on which TA grades the paper, and further that the rigor of the feedback will maximally benefit the student.

(4) Each TA then selects a paper from his or her section, grades it, and submits that to me for approval. If I am convinced that the TA now meets the required standard of quality, the TA goes on to complete the grading of the rest of the papers."

http://www.insidehighered.com/advice/2011/01/07/gorovitz_on_mentoring_teaching_assistants