Search the Lounge


« Pretty Soon You’re Talking Real Money | Main | “This Place is Hell” »

May 10, 2010


Feed You can follow this conversation by subscribing to the comment feed for this post.


I use the "gestalt method." In other words, I don't allocate points. I read the answer and assign it a grade based on my sense of how good it is. No key, no points. It's not precise in a point-like way, but I think it rewards what are in fact better exams. And it takes account of the fact that writing counts, big time. To me, points give a false sense of precision.

Michael Risch

I think the gestalt method wouldn't work for me because sometimes there are really well written exams that just miss a bunch of key issues. Sometimes I'll grade an answer and, after adding up the total points, think, "That was a great answer, why is the score so low?" Invariably, the student will have spent a lot of time writing about a couple things, but skipped other very important things.

I think the points keep me honest - ensuring that I don't give too high a grade to an answer that was incomplete. I instead reward really great writing with added points in Creativity, Organization, and "Other." That helps differentiate the scattershot answers from the well thought out ones.

Jacqui Lipton

Actually, your post also raises something that I was thinking of posting about myself which is how best to give feedback to students after the final exam is taken. Obviously some students (usually disgruntled students) come to see you to complain about their grade/go over their paper. But because so much of the learning is inherent in the "feedback" part of the equation, I'm interested in the format in which you are going to make your memo available to students. Post it on Blackboard? Email it to them? Send hard copy to last known address? And I guess in the "after the fact" mode, even if you do this, there's no guarantee that students will bother to look at it.

Jessie Hill

Interesting about the gestalt method. I don't completely trust myself to do that, either, but more fundamentally, I wonder how I could make sure I conformed to the curve if I did that. Our first-year curve is relatively rigid.

Jacqui, I was planning to post the memo on the course Blackboard site. I gather students retain access to those sites even after the course is over. I am partly hoping that it will provide an opportunity for students to see where they have gone wrong (if they are so inclined). I can certainly tell them that they are required to read it before coming to see me about their exam grade. But I am also hoping that it will be a good learning tool for next year's students. I'll let you know how it works out....

Jacqui Lipton

So did you tell the students in advance that you would be posting some feedback on Blackboard after the exam? Or how will they know/remember to look there if they don't contact you about going over the exam?

Michael Risch

I post my memos (which also give general exam tips) and the top three exams to my own website. This will work out well, as students at my new school can see my exams, as can students from year to year.

Henry  Noyes

I create a grading key that is standardized for the exam. I allocate some points to each issue for spotting the issue and stating the rule(s) for that issue corectly. I allocate some additional points to identifying the subissues and subrules. The above are purely objective (IMO). I then allocate some points to analysis for each issue. This is much more subjective. Before I finalize the key, I grade ten exams with it. I then modify the key and finalize it. In this way, I can (almost always) determine whether I have over-weighted or under-weighted some issues and some allocations of points. I then throw away the ten practice keys. I start the real grading with a different exam, say Exam #11 (not one among the ten I practiced on). When I am done with all of the exams (including the ten practice exams), I regrade Exams #11-20 to make sure that I am scoring them the same way at the end of the process that I did when I started.
This has worked well for me because I get a sense of molding/modifying the key to reflect the reality of the students' performance. I then can provide each student with a scoring sheet that reveals their performance on every part of the exam. A review of the key usually tells the students where they have consistent problems.
For my two-semester civil procedure course, I purposely weight the key a little heavier towards points allocated to issue spotting and rule statements. In the second semester I expect more and better analysis and the key is more heavily weighted in that direction. I tell the students about this expectation.
It is a lot of work, but the students really appreciate it and I believe they learn a lot from review of their exam between the first and second semester.
I did not like the "gestalt" method when I was a student and I do not like it now. I have serious concerns about whether that method is reliable and can be replicated (I think not) and it does not give the students much to go on if they want to learn what they did right/wrong and how to improve their performance.
Henry Noyes

I.G. Cohen

I just want to sound a note of caution about posting an exam memo in an easily obtainable form online (especially in take-home exams). What you make available online to one set of students will inevitably become available to later students.
Here is why I have concerns: Try as we might, from year to year there will be commonalities in our exams. Given enough years of memos and a finite amount of likely exam configurations, I think that can become problematic. It is entirely possible that if you make your memo available online, and it is a very good memo, your students during a take-home will essentially cut and paste your memo back into their exams answers with appropriate alterations. When I was a law student (class of '03) this was not an uncommon way of taking take-home exams. Moreover, students will start to approach the issue-spotting parts of your exams like wine tasting: "Oh this exam has a trace of the '05, with a hint of the '09, and a little more acidity."
Now perhaps you think this form of memo-aided issue spotting and answering produces a work product that more accurately measures student performance. I think that is an interesting thing to argue about, but unless you've resolved for yourself that it does I'd be cautious about making your memos available online.
One possible work-around is to write-up the memo, provide it to your assistant in hard copy, and allow students to examine it at your assistant's desk but not take it with them or photocopy, etc. Of course, that may suggest to students that you are overly concerned about the dissemination of your memo. Given that I am overly concerned, it wouldn't bother me for them to know that, but for you...


One solution to Glen's comments about the problem of exam memos is... to stop giving issue spotter exams. I typically give questions with two issues, sometimes just one. Often times, I'll identify the issues for the students. My focus is on analysis. Sometimes, I'll write an opinion (or find a circuit court opinion) and ask them to write a dissent. I do this for two reasons: (1) I hated issue spotting in law school and found it insulting after a theoretical class that barely touched on black letter law; indeed, to be provocative, I might even say I cannot fathom what the purpose of issue spotting really is; (2) I teach a theoretical class, and the skill set I emphasize focuses on reading cases closely and synthesizing them, and putting them in a larger context. Given that approach, issue spotting doesn't seem like the right way to go. And because I don't do issue spotting, I can use the gestalt grading method, because cogency and argumentative persuasiveness is the crux of it.

The comments to this entry are closed.


  • StatCounter
Blog powered by Typepad