Another Life google archive photo.

I’ve been grading for the past several days.  I’ve thought about switching to the “staircase” method (recommended by O&M a few years ago).

Or alternatively, I’ve thought about using  a “first impressions” method of grading.  So, Nalini Ambady’s research on “first impressions” shows that students can accurately judge a teacher’s abilities by watching a short 10-second clip —  I wonder about the reverse, and also about whether I can grade a paper by the “first impression” I get within 10-seconds from the opening paragraphs.  I’m guessing that grades based on reading just the opening paragraphs versus the full paper would be very highly correlated.  Alas, I’m trying to also teach something and provide feedback, and not just give a grade.


Written by teppo

December 22, 2009 at 7:08 pm

Posted in uncategorized

6 Responses

Subscribe to comments with RSS.

  1. Someone who’s now a professor at the University of Chicago taught me the “first impressions” method in the context of grading math (!) exams.



    December 22, 2009 at 8:24 pm

  2. Ruchira — perhaps related to your point, one can probably grade final projects/exams with great accuracy by just reading the name (past performance tends to be a strong predictor of future performance).



    December 22, 2009 at 8:41 pm

  3. I can see an A paper in the first two paragraphs– and a D paper, too. But the B- / C+ / C range takes me a long time to distinguish. A paper with things going wrong requires some time to figure out how many things have gone wrong, and why, and how. T


    Jacob T. Levy

    December 22, 2009 at 11:43 pm

  4. Jacob: You’re right, the stuff in the middle can indeed be harder to distinguish.



    December 23, 2009 at 3:43 am

  5. I’m increasingly thinking I could pretty accurately predict student grades by administering a directions test ( on the very first day of the semester.



    December 23, 2009 at 3:57 pm

  6. Is it true that the NSF performed a test by evaluating a sample of projects on reading only the first page, and realised that the first page evaluation give the same results as the full reading?



    January 13, 2010 at 2:59 pm

Comments are closed.

%d bloggers like this: