« The Real Housewives Get Real | Main | Faculty Lounge on Financial Market Reform »

Tuesday, September 13, 2011

The One Final Exam Mentality and Blind Grading: You Can't Have One Without the Other?

I've written several posts on this blog about the "one exam to rule them all" mentality that exists in law schools across the country (see, e.g., here and here). And, I've written several posts about the process of "legally blind grading," which is also encoded in the DNA of the modern law school (see, e.g., here, here, here, here, and here). I get the sense that students recgnize that there are pros and cons to blind grading in law school and that having the practice ablosished isn't high on their priority list. But I think that many a student laments the fact that one final exam can make or break her. So, I think that many students would welcome multiple points of grading in law school, such as, at a minimum, a graded midterm. But here's the thing: Are the "one exam to rule them all" mentality and the process of blind grading like love and marriage, you can't have one without the other?

Try to imagine a world in which a law professor gives a graded midterm at roughly the midway point of the semester. Under a blind grading system, students will turn in exams with exam codes and not names. And then...radio silence. As far as I know, most schools don't allow students to talk to professors about the final exam after they take the exam and before the professor turns in her grades, and I imagine that the same would apply to graded midterms. I imagine that it would take the average professor at least 3 weeks to finish grading midterms, especially if the professor is teaching 2 classes and giving midterms in both classes.

If I'm giving a fair midterm, I assume that it covers most of the material from the first half of class. And, I'm assuming that, for many students, the midterm is a gut check, after which they will have many questions for me about what they've learned in class and how it applied to the fact patterns they just answered. Indeed, I give an ungraded midterm in my classes, and I think that some of the best learning in my classes takes place in the week or 2 after the midterm when I meet with individual students to discuss how they performed on the midterm and how they can improve on the final. With a blind graded midterm, I could do this as well...but it would be 3 or more weeks after the midterm...after the dust has settled...after we've become entrenched in material from the back nine of class.

Pushing exam review back in this way would also cause me to question whether final exam grading is truly "blind." As noted, after my ungraded midterm, I meet with students at about the midpoint of the semester to discuss their performance on the midterm. A small, but significant minority of students avail themselves of this opportunity, and, for the most part, these reviews are out of sight, out of mind when I grade final exams. The exam I grade is by 24601, not Jean Valjean. That said, every once in a while, thoughts creep into my head based on meeting with students after the ungraded midterm. Isn't this the way that I told Student A to address this type of issue? Doesn't Student B often confuse these 2 rules? I usually dispel these thoughts, though, and I doubt that they have any influence on my grading.

But if I gave a graded midterm... My meetings with students would be later in the semester. I imagine that (many) more students would want to meet with me. I imagine that my meetings with these students would be more in depth. I would look at what they wrote more closely. These are all good things. But when I got to grading final exams, would there still be the veil of ignorance required for blind grading? I don't think so, which is part of the reason that I haven't (yet) pulled the trigger on a graded midterm.

So, are these concerns legitimate? Can you have a graded midterm and still engage in blind grading?  

-Colin Miller

Posted by Evidence ProfBlogger on September 13, 2011 at 09:05 AM in Teaching Law | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference The One Final Exam Mentality and Blind Grading: You Can't Have One Without the Other?:


"But I think that many a student laments the fact that one final exam can make or break her. So, I think that many students would welcome multiple points of grading in law school, such as, at a minimum, a graded midterm."

I understand the use of ungraded midterms, or midterms generally. But I've never understood some of my fellow students' wishes that grading be diffused over multiple grade points. The reason I don't understand is that grading already is diffused over multiple grade points - the multiple questions on a final, each of which contain a number of sub-parts. A good criminal law exam, a good evidence exam, a good criminal procedure exam, will tend to touch on the majority of what one learned in class. It would be problematic if one's grade came down to a single topic, or two topics, that one just didn't get, amidst a sea of untested material that one did. But most well-written finals cover a broad random sample of covered material, such that a midterm seems pretty duplicative. The people who do well on one will do well on the other; little will change. Such has been my experience in my midterm classes, at least. Now, I have nothing against ungraded midterms, or midterms, as an exam practice device, but I don't think they lead to a more "accurate" grade.

Posted by: Asher | Sep 14, 2011 1:44:50 AM

There is nothing inherent in blind grading that requires reliance on a single end-of-semester exam. CUNY strongly encourages at least two grading opportunities for students in every class. When I teach required classes I always give a mid-term that is worth somewhere between 15% and 30% of the final grade. The grading is blind but I am always willing to meet with students to discuss their performance (an interesting sidenote, very few students actually avail themselves of the opportunity.) I also devote some classtime to reviewing the midterm--focusing on whatever the students seemed to struggle with the most.

I believe, though have never confirmed that students use the same exam number for midterms and finals. Therefore, I ask them to block out their exam number on their copy of the answer sheet before comign to talk with me. That way, I can't inadvertantly remember their number (an unrealistic fear anyway given that the exam numbers are random sequences of somewhere between 5-7 digits.)

Posted by: rebecca bratspies | Sep 13, 2011 8:52:51 PM

I give a kind of modified mid-term (it is a take-home exercise and is not blindly graded) in my Evidence course (required at my law school, although for upper-level students). I don't mind the grading time (although that time is substantial), and I think the exercise is good for my students.

The one concern I have is that I am giving my students a thing they have to do in the middle of the term, which is unusual, which means they may neglect their other classes while they are doing it. In my 1L class, for that reason, historically I have had a midterm, but have made it voluntary and ungraded (most students do it, and I give them about 3/4ths of a page of written feedback). I have considered formally grading it and requiring it.

Posted by: Deb Ahrens | Sep 13, 2011 7:07:15 PM

Brad, GU, Orin, and Heidi: Thanks so much for the responses. I guess, then, that the answer to my question is that you can have blind grading AND a graded midterm...as long as you are a bit creative with it.

Posted by: Colin Miller | Sep 13, 2011 6:11:33 PM

In response to Colin...(1) I spend approximately 8 hours of my weekend to grade the essays and complete a quick checklist for each one (about 5-10 min per answer); and (2) At the end of the next class after the midterm, I give the students my model answer, even if I have not finished grading them all. Of course, I also forbid them from talking to me about it until I've finished grading. Then, after I've finished grading, I give each student his actual grading checklist so he can see his score as well as a list of the things that were done well versus those that need improvement. In that same class period, I discuss the model essay answer and common mistakes, as well as the multiple choice questions and answers. I also offer each student an opportunity to meet with me later to discuss his essay answer provided that he first color codes his essay answer (long story) and jumps through a few other hoops that I think will help him understand what he did right/wrong.

Overall, this approach seems to work for the students and for me. Is it more work for me? Yes, but not THAT much more. (And, I've convinced myself that the time spent on this midterm essay walk-through leads to better final exam essay answers which ultimately saves me some time down the road.) I'm always surprised (dismayed?) by how few students take the time to review their answers with me in person later. Perhaps that's because I do a decent job of showing them what they did right/wrong in class and via the checklist. Or, perhaps it's because many of my students simply don't want to take the time to meet in person. Or, maybe I'm less approachable than I think I am. Regardless, the students who do take the time to carefully review their answers versus my model answer seem to appreciate it and learn from it. And I sleep better knowing that I gave every one of them the chance to do so at least once before the ultimate "one exam to rule them all."

Posted by: Heidi Anderson | Sep 13, 2011 5:57:28 PM

I don't think this is a problem, as I don't think we professors have any real way to ID students based on these sorts of things. (In a class of 100, do you really know who confuses a particular set of rules?) But if you really think this is an error, why not just make sure that you don't give any one student advice that you're not giving to everyone else?

Posted by: Orin Kerr | Sep 13, 2011 4:26:33 PM

Back in law school, one of my duties as an editor of the law review was to decide which of my colleague's student notes would be published. I read all of them "blind," but knew most of the authors personally. Before choosing I only knew a few people's topics—probably 85% of the notes were anonymous for me (there were about 35 notes total).

Echoing Katie above, my guesses as to the identity of the author were very often wrong. And these were my peers (who I probably knew better than the average law prof knows her students), and they were notes, which are far more "personalized" than quickly-typed exam answers.

This is purely anecdotal to be sure, but I think prawfs might be surprised at the unpredictability student answers.

Posted by: GU | Sep 13, 2011 2:13:00 PM

Is the pedagogical value of a graded midterm exam worth at least one week of class time? If you were relieved of teaching duties for the week after the midterm would that be sufficient time to grade the exams?

If the answer to both those questions is yes, a good solution should present itself (though you'd probably need to get administrative buy-in).

Posted by: Brad | Sep 13, 2011 2:09:55 PM

Heidi: Thanks for the comment. That sounds like an interesting approach. I have two questions for you if you don't mind answering them:

(1) How long does it take you to grade all of the midterms?

(2) Are you allowed to discuss the material covered on the midterm with students before you finish grading the exams? And if not, do you find that limits your ability to discuss material from the first half of class in the aftermath of the midterm?


Posted by: Colin Miller | Sep 13, 2011 10:06:04 AM

Doug B.: Thanks, those are useful comments. In an earlier post, I discussed how I might want to incorporate "answer justification" multiple choice questions into my final exams (http://prawfsblawg.blogs.com/prawfsblawg/2011/09/i-have-been-thinking-about-adding-some-multiple-choice-questions-to-my-current-all-essay-exams-in-recognition-of-the-fact-tha.html). I guess one solution would be to give these MC questions as a midterm and then my usual short essay questions as the final exam. I could do this and still give an ungraded short essay midterm so that there would be (1) multiple points of grading, (2) little concern about grading on the final not being "blind," and (3) the opportunity to take an (ungraded) exam that looks like the final exam.

Posted by: Colin Miller | Sep 13, 2011 10:01:38 AM

I taught a class with blind grading (of only the exam component of the grade) and a graded midterm, albeit for undergraduates not law students. It was also a very small class and I had interacted significantly with all the students. If there was any class where I should have been able to know whose final exam was whose, it was that one.

And . . . I didn't. Yeah, I had a guess here and there, but even as I was grading them, it was clear to me that I didn't know which exam was which. I don't think blind grading has to be "perfect" in the "professor will never have a guess about who the exam is" sense. That's impossible; professors are humans and will have a sense. It just needs to introduce enough doubt into the equation to remove the belief that you know whose exam you're grading and thus the associated unconscious favoritism. In my experience, it did even given significant familiarity with the students' work.

Posted by: Katie | Sep 13, 2011 9:58:49 AM

All of our Fall 1L courses must include a midterm (up to the professor as to whether it "counts"). I make mine worth 10% of the final grade. It's a combination of MC and essay, the latter of which I grade very quickly using a checklist, which I later give to the student. The overwhelming emotion I get from students post-midterm is gratitude. They are so thankful for getting the opportunity to "mess up" when it only counts for 10% of their grade and, hopefully, still have the time and intellect to fix their mistakes prior to the final exam. At final exam grading time, I never once have thought to myself, "Ooh, this paragraph sounds like something Student X would write...I wonder if this exam is from Student X?" And if I did, I can't imagine it affecting my grading given how regimented my grading checklists are. I suppose it's possible but it's never happened to me.

Posted by: Heidi Anderson | Sep 13, 2011 9:50:35 AM

Great issues you are exploring here, Colin, though I think the most simple explanation for "One Final Exam Mentality and Blind Grading" is that they both make our lives as law professors somewhat easier, especially given that we generally have to fit our class grades to some pre-determined curve which requires some hard-working earnest students to get (much?) lower grades than they wish/want.

I do not mean to suggest/assert that the one exam + blind grading approach lacks other possible virtues for students and other legal education constituents. Indeed, I doubt many students AFTER their 1L year are still eager for more tests/graded assignments in traditional doctrinal courses. I think busy upper-level students --- especially those who have figured out the "game" of doing well on traditional law school exams --- end up really liking the fact that they can coast most of the semester in some/many courses and still get an A thanks to one exam + blind grading. And, especially for upper-level students who are sensibly spending time on job-hunting and journal work and other useful professional activities, this allocation of time/energy may serve everyone well.

For 1Ls, however, the story is much different (especially in the Fall semester), and I know they like and benefit from feedback that is neither blind nor after the course ends. But I wonder why that needs to be in the form of an exam? How about graded mid-term papers? These would not necessarily disrupt your ability to do blind grading at the end. Of what about computer-administered assignments that use technology and/or the crowd-sourcing to potentially provide students with helpful feedback without shredding the veil of ignorance?

In other words, I think these issues can/should be nuanced further by distinguishing between 1L and upper-level classes, and also by not falling prey to a more general "exam" mentality. And, perhaps more importantly, I think we ought to question whether and when students truly do benefit from blind grading as much as we think they do (largely because I think we benefit from it).

Posted by: Doug B. | Sep 13, 2011 9:35:46 AM

The comments to this entry are closed.