« State action and the Yankees lawsuit | Main | U.S. News and "assessments" »

Monday, April 20, 2009

A Modest Proposal Regarding Blind Grading

Why does blind grading rule the roost in law school? (Even Justice Stevens noted, in McIntyre v Ohio Elections Com'n, 514 U.S. 334, 342 n.5 (1995), that the practice of grading law school examinations "blindly" is "now-pervasive."). The most common rationale that I have heard is that "blind grading is meant to ensure, at least in part, that a student's actual performance on the exam, and not any improper influence, is the exclusive factor in determining her final score." Meredith Hattendorf, Comment, Theoretical Splits and Consistent Results on Anonymous Political Speech: Majors v. Abell and  ACLU of Nevada v. Heller, 50 St. Louis U. L.J. 925, 927 n.13 (2006).

Fair enough, but why is blind grading not nearly so prevalent at colleges and most other graduate schools? Is the implication that blind grading was introduced based upon the proposition that law professors were less able to remain unbiased during the grading process than other teachers? Well, historically, the answer is apparently, "yes," but not in the way that I expected. In One Law: The Role of Legal Education in the Opening of the Legal Profession Since 1776, 44 Fla. L. Rev. 501 (1992), Paul Carrington tracks the creation of special admissions programs for minority students in the mid-1960s. He then notes that "[t]o protect against concerns of favoritism for students specially admitted, most schools adopted some form of blind grading."  
So, it seems as if the main reason cited in support of blind grading is not (really) the same reason that it was created. And it seems to me that there are many possible problems with blind grading, with the foremost being (my belief) that it reinforces the one final exam to rule them all mentality. Of course, there are also many benefits to blind grading, not the least of which is the assurance it gives to students that they are being treated fairly (although it seems as if many students believe that blind grading is a myth). Moreover, regardless of the pros and cons of blind grading, it seems that blind grading has become embedded in routine law school practice to the point where it has become part of our law school culture. Well, 1L culture, anyway.

But, what about upper level classes, and especially electives and/or classes taught by more than one professor? Shouldn't professors teaching these classes have the option to chose blind or non-blind grading in the same way that they can decide issues such as whether to give an in-class or take-home-exam or an open book or closed book exam? Sure, many students would not want to take a non-blind graded class, but they could simply decide not to take the class just as students averse to closed book exams can now choose not to take classes with professors who give such exams. Conversely, I imagine that there are many students who would prefer classes where they could be evaluated based upon their performance over the entire semester and not simply their performance on one exam at the end of the semester. Indeed, many of these students already do so by chosing to take many seminars or other "paper" classes where grading is not "blind."

So, what do readers think? Should professors teaching upper level classes have the option of making their classes non-blind graded, or is blind grading different and thus something that should be required in every (non seminar/paper) class? If given the option, would you engage in non-blind grading, or do you prefer the blind grading approach? 

Posted by Evidence ProfBlogger on April 20, 2009 at 09:13 AM in Teaching Law | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference A Modest Proposal Regarding Blind Grading:


Two mechanical points:

1. Blind grading is fully compatible with take-home exams. I've been doing it for years; you have the exam administered by the Dean of Students Office, or anyone else at arms length.

2. The bump up/down system is easily adapted to avoid the problem noted in Colin Miller's note above: I give TWO kinds of bumps, "small" and "large". A "large" bump is good for one one grade step. A "small" bump is good for one grade step if the student was halfway there anyway. I grade numerically. If the two grade levels are, say, 3.5 & 4.0, a small bump is effective only if the student scored 3.75 or above.

Posted by: Michael Froomkin | Apr 22, 2009 10:46:29 AM

And if we think that seminar papers are graded fairly, that raises the question of why we think that non-blind exams would be graded less fairly.

My experience is that students take seminars despite the fact that they aren't graded blindly, not because they think there's no bias in the grading of seminar papers. Discussion about "such and such professor would never give me an A because of X" abound, warranted or not.

I graded the midterm and finals blindly in an undergraduate class I taught where it wasn't a requirement and was glad I did (I enlisted a neutral third party to hold the students' self-chosen ID number and name match-up until after I had graded the exam). I was relieved not to have to second guess myself as to whether I was grading an exam more leniently because I liked a student, and I think the students appreciated that I was approaching the subject of bias and favoritism with seriousness, even if they had to take my word for it that I wouldn't get the match-up key early.

Posted by: anon | Apr 20, 2009 8:14:53 PM

Thanks, Joe. You are certainly right that there are many reasons why law students take seminars/paper classes. You note some of these reasons, and another big one, which I will address in an upcoming post, is that they are usually not graded on the regular law school curve. I would say, though, that non-blind grading is one of the reasons students take seminars/paper classes, and from what I have seen as a student and professor, many students do not like blind grading and would prefer exam classes without blind grading.

Posted by: Colin Miller | Apr 20, 2009 3:17:34 PM

Hadar and C.E. Petit's point about law reviews is inapposite. Students run law reviews, and there are many article-selection criteria which relate to varying degrees to the abstract "quality" of the legal scholarship. If schools want to evaluate professors on the merits of their scholarship, they will not use placement as a proxy.

Posted by: joe. | Apr 20, 2009 3:14:37 PM

I think you're overstating the paper-seminar thing, with the suggestion that people take these classes because they are non-blind. I suspect that most students think seminar papers are still graded fairly (participation-based grade modification is separate, and consistent with blind grading of exams), and take these classes for other reasons: the subject matter is interesting, a paper is a better evaluation than an exam, small-group discussion is more engaging, etc. And if we think that seminar papers are graded fairly, that raises the question of why we think that non-blind exams would be graded less fairly.

Posted by: joe. | Apr 20, 2009 3:04:17 PM

Thanks, C.E. Petit and Hadar Aviram. As I noted in my post, "regardless of the pros and cons of blind grading, it seems that blind grading has become embedded in routine law school practice to the point where it has become part of our law school culture." The same seems to be true with non-blind submissions being embedded in routine law school practice. See Nathan H. Saunders, Note, Student-Edited Law Reviews: Reflections of an Inmate, 49 Duke L.J. 1663, 1677 ("In short, as a direct result of the firmly entrenched hierarchy of law reviews, collective action among law reviews to reform student editing and article submission processes is impossible.").

Posted by: Colin Miller | Apr 20, 2009 2:52:56 PM

C. E. Petit said exactly what I meant to say, only better. It's quite ridiculous that one should, by virtue of having been elevated to knighthood, benefit from the byzantine way in which we manage our publication system, and at the same time tout the importance of unbiased grading.

Posted by: Hadar Aviram | Apr 20, 2009 12:31:12 PM

I find it rather ironic (and disturbing) that law schools emphasize blind grading, but that law reviews don't use blind submissions. The number of bad pieces from "prominent" professors that crowds out more-thoughtful pieces from those who aren't (or are affiliated with lower-prestige schools) is probably at least as much a problem as the peer-edited/student-edited dichotomy.

By analogy, it's ok for law reviews to be accused of favoritism, but not for the schools themselves. And we wonder why practicing lawyers think the legal academy has no connection to reality...

Posted by: C.E. Petit | Apr 20, 2009 11:38:43 AM

Thanks, Rick. Here's another citation supporting the favoritism theory (at least for the University of Florida Law School):

Jeffrey L. Harrison, Race Lines, Immunities and Bonuses ... And Reader's Response, 48 Fla. L. Rev. 419, 427 n.28 (1996):

One of the quirkier experiences with efforts to avoid labelling evidently occurred at the University of Florida Law School and, no doubt, at other law schools as well. According to the “oral history” of the University of Florida Law School, the School has gone back and forth on the blind grading issue. Blind grading was once adopted at the outset of integration in order to protect those who opposed integration from accusations that their grades were a function of racist attitudes. While some may see this as benevolent, another interpretation is that racist professors favored blind grading because they believed that minority students would receive failing grades without the professor's “help.”

Posted by: Colin Miller | Apr 20, 2009 10:00:26 AM

Thanks, AnonProf. Here a couple of responses. The bump up/bump down approach combined with blind grading is a nice way to avoid students' fates being determined solely by the final exam. I just think that it is a bit too mechanical. Let's say that I decide to give STUDENT a bump up of one position. And let's say that STUDENT is close to the B/B+ border and now gets a B+. Well, that's great, and the system worked well. But let's say STUDENT is close to the B/B- border, and the bump up gives STUDENT a B+. That might not be so great, and if grading were not blind, I likely wouldn't have bumped STUDENT's grade up to a B+.

Or, let's say that, like your school, my school only allows bump ups/bump downs of one position. That means that I likely over-reward students who just make the bump up cut and under-reward students who just missed it. Non-blind grading introduces the possibility of bias, but I think that it also allows for more fluid grading.

Also, here are a few of the reasons why I think that blind grading reinforces the one exam to rule them all format. First, at many schools, blind grading means that students can't have contact with professors from the end of the final exam until grades are given. If I gave graded work during the semester, I imagine that I would have to do something similar, and this would obviously have some bad side effects. Second, if I gave a few graded tests during the semester, I imagine that I would get to the point where I basically knew which student wrote which exam by the time the final rolled around. In other words, it seems that the more graded assignments one gives, the tougher it is to keep grading entirely blind.

Posted by: Colin Miller | Apr 20, 2009 9:50:21 AM

It has always seemed to me that blind-grading was less about protecting students from bias than it was about -- as Carrington suggests -- protecting them from favoritism . . . and protecting professors from accusations of bias.

Posted by: Rick Garnett | Apr 20, 2009 9:42:18 AM

Personally, I do not think that blind grading -- by itself -- creates any problems. At schools where I have taught the blind grades can be adjusted by the professor after they are submitted to the registrar. One school only allowed the grade to be moved up or down by one position(so, from a B+ to an A- for example). The school that I am currently teaching at has no such limitation, which I believe is a better approach. I regularly modify student grades in this manner, and my syllabus alerts the students to this fact. I think blind grading with a bump up/bump down provides some transparency. Students can find out if you have adjusted their grade up or down for class participation or other reasons.

Blind grading is not what causes professor to only give one exam each semester. In first year courses, I think many feel that there are too many students (and maybe more importantly, no graduate TAs) to give a mid-term. In smaller upper-level classes I guess the logic is that students have already figured out how to take exams. I personally do not think that one test is an effective way of teaching, so I'll be investing the extra time it takes to at least offer and grade a mid-term next year. But I don't think this has anything to do with blind grading.

Posted by: AnonProf | Apr 20, 2009 9:32:46 AM

The comments to this entry are closed.