pageok
pageok
pageok
Why Law Schools Generally Grade Based on a Single End-of-Semester Exam:

Most law school exams, to my knowledge, involve a single end-of-semester exam. Some comments in this thread argue that this is pedagogically unsound, and that having several exams — a final plus a midterm or two — would better measure people's knowledge. (It might also help students learn the material better.) I'm not sure that this is so, because I haven't looked into the research; but it seems plausible, and colleagues I trust have said that the research does support this. Let's assume then that this is right. Why then the single-exam format?

Some of the commenters identified one important answer: Professors hate grading exams, and would thus rather grade one exam than two or more (since presumably the two or more put together would involve more pages of exam answers than just the one). With few if any exceptions, law professors do all the grading themselves. "Law is the only discipline in which students select what's published in the journals and the professors grade the exams." Professors and professor-run bodies get to decide how many exams there'll be. Professors hate grading the exams. You get the picture.

I think there's a lot of truth to this, but let me suggest an extra factor: I suspect that most students prefer the one-exam structure as well, so that there's little pushback against the professors' one-exam preference, and there would likely be some pushback against professors' attempts to shift to the "better" two- or three-exam format.

In my experience, most law students are very serious about studying for exams, and thus spend a lot of time studying for them and stressing out over them. What's more, if you cut the significance of the exam in half, students won't think they have to study half as long or half as hard. Rather, they'll still see themselves as in hot competition with the other students, and will feel the pressure to study nearly as hard and long for each of the two exams as they would have for the one.

Naturally, some students might value the greater accuracy of the multi-exam approach, and the lower risk to them that this approach provides (since they don't need to worry as much about tanking a class because they screw up one exam). But for many students the downside of yet more studying and yet more worrying exceeds the upside of less risk and better measurement accuracy. And on top of that, grades are on balance mostly a zero-sum game — more accuracy of testing will just rearrange the As, Bs, and Cs, so as many students will lose as will win. Increased time needed to study for exams is a net loss for students as a whole. Therefore, you'd expect that more accurate measurement devices will please only a few students, while greater need to study will displease many more students.

So professors don't want to institute more exams, even if having more exams provides more accurate measurement. More students, I suspect, don't want to institute more exams, even if having more exams provides more accurate measurement. Who would predictably benefit from more accurate measurement, and lose nothing from it? Employers, and perhaps indirectly the legal system generally.

Employer pressure may indeed have some broad effects. For instance, while abolishing grades would make life easier for professors and would in some ways make it easier for students, most employers won't hire students who don't have grades. Even Yale, which some say "doesn't have grades" does have a "High Pass" and a "Pass" [UPDATE: and two lower grades which are rarely given], which selective employers definitely focus on; and schools that aren't as selective as Yale probably couldn't get away even with that (though Berkeley somehow manages with only three grades).

But on balance I suspect that few employers care about precision enough to make a stink about this. The grades give employers some decent signal of applicant quality; employers realize that no grading system would yield a perfect signal; it's hard to tell how much extra accuracy one grading system provides over another; the random noise in the grading probably averages out in considerable measure when you look at the student's entire transcript; so employers are unlikely to spend much time and effort on even thinking about the midterms-or-no-midterms question, much less trying to pressure stodgy academic institutions into changing their ways, or shifting their hiring practices towards schools that provide more midterms (which would provide indirect pressure on the law schools).

This is just a casual analysis, not based on any deep thought, and I might well be mistaken. Yet it seems to me that while professor self-interest is an important factor here (a free-market guy like me must surely appreciate the importance of self-interest in molding human behavior), student self-interest may in practice reinforce the professor self-interest rather than clashing with it.

Related Posts (on one page):

  1. Why Law Schools Generally Grade Based on a Single End-of-Semester Exam:
  2. Exams as a "Rewarding Experience":
Ilya Somin:
A minor note: Yale has "Honors" (H), "Pass" (P), "Low Pass" (LP), and Failure (self-explanatory). The latter two are rarely used by most professors. However, this is not much different from the way in which most professors will give very few Cs and Ds under the conventional grading system unless forced to do so by a curve.

The really big difference between Yale and its competitors is that Yale has a pass-fail grading system for the first semester of law school
5.8.2006 3:10pm
Eugene Volokh (www):
Ilya is right -- I oversimplified because LP and F are indeed almost unheard of; but I'll correct the post.
5.8.2006 3:12pm
gerald span (mail):
When I asked this question 40 years ago during my first year in law school, the response from my Contracts professor was that one test at year's end best simulated the pressure a lawyer faced on the day of trial, with no opportunity to make up for a poor perfomance later on.


5.8.2006 3:29pm
Steve:
During my three years at Michigan I think about 95% of my classes were graded on the one-exam system. When I was a 1L, I remember a lot of students were very anxious about this, for the usual reasons, and they wished there were some kind of midterm so they could at least acclimate themselves to the idea of a law school exam without their entire grade being on the line.

One of the Con Law professors decided to offer a midterm to his 1Ls, although I'm not sure if this was in response to demand or if he had planned it all along. In any event, had it been a traditional issue-spotting type of exam, this probably would have worked out well. However, the actual question was some totally whacked-out scenario involving secession (you might recall that there are very few Constitutional provisions addressing secession!). Not only did this freak out the students who actually took the test and had no idea what to say, it freaked out the entire first-year class once word got out.

I assume, without really knowing, that the professor was being intentionally evil.
5.8.2006 3:32pm
Mr. X (www):
Anecdotal, I know, but this law student would prefer multiple tests to a single exam.
5.8.2006 3:42pm
Mr. X (www):
When I asked this question 40 years ago during my first year in law school, the response from my Contracts professor was that one test at year's end best simulated the pressure a lawyer faced on the day of trial, with no opportunity to make up for a poor perfomance later on.

It's a bullshit analogy used to justify a poor system of teaching.

First, a lawyer does not get jumped by a bizarre fact pattern in a trial.

Second, law school is training to be a lawyer, not some sort of bizarre lawyer-simulation game.

Like most things in law school, the reason for end of semester high-stakes exams is that things have always been done that way, so it must be the right way.

See also: Bluebooking.
5.8.2006 3:45pm
Aultimer:
My Civ Pro professor had a mid-term and final. It was truly helpful to me as an undergrad science major with little experience of humanities-type testing. Civ Pro was the least LiberalArty of the 1L curriculum - if I had to choose one class to have had a midterm, it would have been Con Law. I still couldn't get past middle B in that class.
5.8.2006 3:45pm
pallen:
Being a graduate not of law but of engineering--and coming from a notoriously difficult school: Caltech, let me say that you are dead-on about students prefering the one-exam system. Many classes have no exams; instead they have biweekly assignments. These assignments are often quite challenging and time consuming. Yet often so much so that they compete against actual learning. You spend so much time demonstrating your mastery of something you already understand that you are not left with adequate time to focus on those concepts that eluded you on their initial presentation.

I did have a chance to take some courses in constitutional law though where we had a single exam combined with a pair of papers. The striking difference here was the breadth of information which I needed to memorize. It seems to me (not exactly being in the field of course), that having a single exam helps ensure that the students have time to cover the breadth of a subject even if heavy end-loaded studying means they only have full command of the depth for a brief period of time.
5.8.2006 3:47pm
Public_Defender (mail):
I understand the value of a single exam. What I object to is the three-hour-long hand-written exam, especially if closed book. Never in my career have I been given a factual scenario and told to hand-write my evaluation without opening a book.

The most useful exam for judging a student's ability to be a lawyer is the take-home exam. As in practice, the exam measures the student's ability to think and write critically about the subject. The handwritten blue book exams don't give students sufficient time to revise their thoughts and grammar. The handwritten tests also don't evaluate the students' ability to research an answer.

Yes, take-home exams give students the chance to cheat, but the practice of law is both open book and on the honor system. And if you make the test open book, you've pretty defined most cheating out of existence.

One other factor to consider is that some students do better on some kinds of test than others. In one class of more than 100, we had a test that had multiple choice and essay questions on the same material. I was near the top of my class one part of the test and near the bottom on the other.
5.8.2006 3:51pm
Blindgambit:
As a somewhat recent law grad, I have to say I kinda liked the one-exam format. I think what 1Ls find most jarring is the stark disconnect between law school and undergrad in this manner. No one has ever had a class where one test is the be-all end-all, and professors aren't necessarily helpful in explaining its benefits or how to properly approach such exams. I, personally, would advocate for some form of sample exams in the first year to help the transition. After that, though, the single exam works wonders.
One strong reason I like the exam is that, in law, perhaps unlike in many undergrad classes, cases you study early in a semester may relate or tie very well into cases studied later, making for particularly challenging but rewarding exam questions. For instance, Con Law II at my school covered what could be called "individual rights." Thus, one by one, you cover due process, equal protection, etc. An exam that can present a fact pattern with EPC and DPC overtones is much more effective than one that can only raise EPC or DPC issues.
5.8.2006 3:55pm
Dave!:
I would prefer two exams: a midterm and a final. At least then, you have a *chance* of making up for poor performance due to unforeseen circumstances (like having a cold) versus the "one test tests all" aspect. I wouldn't want it to devolve into a regular course with assignments and multiple exams, etc.

However, the 'students appreciate the pressure' of the one exam approach is, in my experience, complete b.s. I have yet to meet a law student who says they prefer it. I would suggest that if you say you have, you're either lying, or not asking in the right context to get an honest answer. (A survey of hands, in class, means _nothing_).
5.8.2006 3:59pm
steve k:
I always felt the big exam at the end of the class was fair--everyone knew about it, everyone prepared for it. To put it another way, I question if two or three tests throughout the term would show significantly different results than one big test at the end.

It would actually be an interesting experiment to see if the two different systems created far different grades for the same group. I would guess 90% or so of grades would hardly change, while the biggest difference would be found amongst those who either get extremely nervous with all their eggs in one basket, or those who can rise to the occasion.
5.8.2006 4:00pm
Mr. Mandias (mail) (www):
AT Notre Dame, 1L's get a mid-term practice exam that the profs grade, but that doesn't count. Loved it.
5.8.2006 4:07pm
EricRasmusen (mail) (www):
The sociology of this is interesting. I teach economics to econ and business students. They would rebel if there was just one exam, because (a) they want to know how they are doing before the end of the semester, and (b) they don't like the pressure of one big assignment.

I would much prefer having just one for-credit exam at the end, because:
1. It de-emphasizes grades as opposed to learning.
2. It reduces the opportunity for whining and grade-grubbing.
3. It is easier for me, while producing pretty much the same grades as having one or two midterms besides.

A downside is that I can't identify failing students and get them to start coming to office hours, as I can after a midterm.
5.8.2006 4:11pm
Stryker:
Someone correct me if I'm wrong, but I believe that most European college level classes have one test at the end. This test may be retaken if the result is not satisfactory, however.
What is the consensus on this? How about re-taking classes? If a student gets a "D" and then an "A," should his GPA be lower than someone who got a "B-"?
--Stryker
5.8.2006 4:28pm
Marcus1 (mail) (www):
There's also the question: accurately tests what?

As I understand, law school testing is only very loosely tied to knowledge of the subject. For many professors, it seems to be much more directly aimed at measuring a specific type of intelligence, or perhaps combination of intelligences.

In that sense, one could say that with 4 or 5 exams per semester over 3 years, a person has plenty of opportunities to overcome one bad test. After all, they're all testing the same thing, and besides, it's not like most employers care about your grade in one class anyway. Doubling or trippling the number of tests, thus, really wouldn't increase the overall "accuracy" much at all.

On the other hand, I think take-home exams could be a big improvement, since the absurdity of the three-hour jam-packed test seems pretty clear. If person A does better on a 3 hour exam than person B but person B does better on a 24 hour exam, I think I'd hire person B every time. In the end, though, the degree of accuracy for any type of testing seems pretty limited, since people are different in so many ways.
5.8.2006 4:39pm
ox:
There's an easy solution here: poll your students. And if the response is what I think it will be (especially among 1Ls), then the only remaining excuse will be that professors don't want to do the extra work of giving their students feedback. And that does seem unjustifiable.
5.8.2006 4:41pm
Stamboulieh (mail):
We are in the middle of exams right now. I'm a 1L at Mississippi College. I actually prefer the one test per class. Our Torts prof. gave us a midterm last semester, and it wasn't indicative of how I did on the final at all. However, that being said, it does give 1L's a chance to see how a test will be taken.

I second the commentor above about the biweekly assignments. They are extremely time consuming (much more than the read/brief day to day dealings of school) and I don't feel like I "learned" as much from them as I did the reading/briefing I did.

As much as I dislike having my entire grade hinge on one exam, I think it's still a better idea than multiple tests throughout the semester.

S
5.8.2006 4:43pm
Jeremy (mail):
I've always much preferred the one-exam format, because I believe it gives a natural advantage those of us who are smart but lazy.
5.8.2006 4:55pm
JK:
One's opinion on this issue most likely turns on one's success under the one exam system. If you, as I did, excelled on end of the semester exams, you support the system.

Adding more exams per class would only cause serious students to spend more time in the law library.
5.8.2006 4:59pm
Steve:
AT Notre Dame, 1L's get a mid-term practice exam that the profs grade, but that doesn't count. Loved it.

I think I remember someone trying this at Michigan. The consensus was that it was a waste of effort, because no one wanted to take the time out of their regular study schedule to prep for an exam that didn't count. But I'm sure it was of value to some.
5.8.2006 5:03pm
BossPup (mail):

I think I remember someone trying this at Michigan.



Professor Cooper does this with Civ Pro. I found it incredibly useful, although horribly frightening at the time. Since it was my first sesmter of law school, I had no idea what to expect as to finals. Once I saw that the point of the exam wasn't to come up with the "right" result, but instead was to find and discuss those interesting wrinkles that the question posed, I felt much better prepared to take those exams for real.
5.8.2006 5:14pm
Mr. X (www):
One's opinion on this issue most likely turns on one's success under the one exam system. If you, as I did, excelled on end of the semester exams, you support the system.

I excel at end-of-semester exams (enough to grade-on to law review) and still think the high-stakes terminal exam system is bad.
5.8.2006 5:25pm
Reinhold (mail):
But if some students will be unhappy with the system either way, why reward the students who are good at cramming and performing well on one three-hour exam over the students who would learn the material better through more exams? I think it's because most professors don't want to tamper with the status quo--probably because they performed well on traditional law school exams and thus think that all students should be filtered through the same system, with the gifted ones also performing well, naturally.
5.8.2006 5:36pm
Former Lee Warmer (mail):
Assuming what you say is true, I think this reflects poorly on law schools. Basically the argument is: professors and students prefer only one exam (mostly professors) so we'll have only one exam. Under this scenario, no value is given, by student or professor, to the quality of the education, accuracy of testing results, or improvement of the legal profession. We're just all doing what feels good.

If it's true that employers and the legal profession would benefit from the relatively painless step, all things considered, of multiple exams, it's a disgrace that more schools don't do it.
5.8.2006 6:03pm
anotherlawstudent:
I, as a law student, would also vote for multiple exams over one exam.

it wouldn't make for more studying, it would merely spread it out. having to learn the stuff a little bit at a time means that you don't have to cram...which means that you retain the stuff better.
5.8.2006 6:04pm
SLS 1L:
As a 1L, I hate the one-exam system. I find it notable that nobody has mentioned other possible forms of evaluation, like homeworks, quizzes, and writing assignments. What if students were asked to draft a contract for Contracts, or take a ten-minute true/false quiz on the Rule Against Perpetuities in Property? Write a paper for Con Law? Take an oral exam on a fact pattern provided in advance?

The real problem with having just one three-hour exam is that it rewards students who are good at a very specific constellation of skills: spotting every legal issue in a bizzarre fact pattern quickly; writing essays without time to think, do research, or ask questions of others; performing under extreme time pressure; etc. These are useful skills, but neither necessary nor sufficient for understanding the material or being a good lawyer. A better system would use multiple forms of evaluation so as to disentangle understanding and the ability to do well on a very specific kind of test.

For that matter, it would also be better to get rid of the time limits. Make the exams 8-hour, so students have time to think and consult their notes, and impose word limits, not time limits.
5.8.2006 6:11pm
WAL:
Since you're writing in response to comments on the other thread, you already know my opinion on this. I'd toss in a couple things, though:

It makes a difference if you're referring to after your first year or during it. After your first year and after interviewing, chances are you already have your summer position and your future job lined up. At that point, you're just going through the motions. If you offered students a 15-minute in-class test at the end of the semester, there are probably a ton who would go for it.

I'd argue it makes a difference if you're talking about your 1L year.

I obviously can only use my school as a reference, but 1.) I don't remember having a lot of free-time anyway. Since your grades counted for a lot more and you could be cold-called in class, everybody studied for the entire semester. In that case I'd either be studying by reading cases or-in the hypo here-studying for tests. Adding in a few tests wouldn't have affected the amount of time I spent studying, at least not much.

2.) If employers acquired students at graduation or in their third year, I think you could make a better case that everything will even out (From my understanding this actually used to be the case if you go back a few decades). Since interviewing is almost always after the first year now, you'd have to go by 6-8 classes-I think that case is harder to make.
5.8.2006 6:19pm
Been there:
I'm currently a second-year law student (evening program at a good school) and empirically, I'm not so sure about Prof. Volokh's argument that students prefer the one-exam system.

My Criminal Law professor this semester considered having a midterm, so asked the class; as I recall, there was pretty strong support for having a midterm. Personally, I supported it because I like the opportunity for a "status check" and course correction before the big roll of the dice.

However, it turned out she wasn't able to do this for some administrative reason, so we ended up with a traditional (and, I thought, very fair) 3-hour final.

Perhaps a more interesting question for the Conspirators is the merits of closed-book versus open-book exams, and their various hybrids (e.g., those where you may bring a personally-prepared outline only, or only some primary source such as a copy of the Constitution). I've now had all of these types and haven't really formed a strong preference.

It seems to me that each method mirrors some aspect of legal practice but I'm not sure which most inspires students to actually learn the material.
5.8.2006 6:22pm
test taker (mail):
Why do law profs do one test?

Because law is the most methodological conservative academic discipline is the US? Because today's law profs were themselves successful at this very method and have internalized it as being the only proper way to teach? Because unlike medicine or business, where the goal is to teach the professional arts, the law schools sort for future academics and leave the professional arts to the law firms? Because that's how Harvard did things beginning in 1880 and everyone's been in lockstep ever since? Because ritual is more important than training?
5.8.2006 6:26pm
Marcus1 (mail) (www):
I'm reconsidering my previous comment that the averaging done in a GPA (which seems to be what employers look at) is enough of a scientific sample. After all, it's really the first year, or even first semester that is critical.

Having more tests would probably be more fair, in that it would not effectively slam the door on students like the current system does with those who happen to have a bad first semester. It might be significantly less arbitrary in that way.
5.8.2006 6:43pm
Patrick (mail):
I love one-exam, but my tax has three assessed pieces!!

Short policy essay marked by lecturer 1, short advice marked by lecturer 2 and 2 hour exam marked by all three lecturers (three Qs; one booklet per Q; one Q per lecturer).

I hate it, but I can see the merit. OTOH, my advanced Con law has no exams, but one mid-length advice and one 20 minute oral argument.

But this is the first time I have had this sort of stuff - previously I have done umpteen 3 hour exams and written two essays, each time the assessment was 100%.
5.8.2006 6:50pm
Steve:
Professor Cooper does this with Civ Pro. I found it incredibly useful, although horribly frightening at the time.

I wonder if that is a new development. Prof. Cooper was my Civ Pro instructor, lo those many years ago, and I really don't remember having any kind of ungraded midterm. Of course, given Prof. Cooper's teaching style, the subject matter, and the fact that the class was at 8 a.m., it's possible that I took the test in a less than fully conscious state.
5.8.2006 7:13pm
Eugene Volokh (www):
An anonymous "3L who only has one more law school exam to take" writes:
On your exam post, I didn't want to leave a public comment, however, I am one of those folks that hate multiple exams in a semester. It makes life a living hell. I barely have enough time to get everything done in normal circumstances (i.e., one exam per class), and multiple exams make the semester more stressful and busy. When a student has an exam, he ignores the world around him including other classes.

I had a law school class where the professor gave two midterms, seven pop quizes, and a final exam. I never worked hard in all my live for four credit hours. I easily did more work in that class than all the other classes combined that semester. And all my other classes suffered (sp?) because of the demanding class too. It was hard to keep up with them and the demanding class.
5.8.2006 7:42pm
Mike BUSL07 (mail) (www):
There is a lot to be said for the one-exam format. Prof. V is right - if I had a midterm, it would just be another frantic week (or 2, or 3) with 17-18 hr/day of studying. When I'm in finals, I don't shave, or do laundry, or go to the gym. (I do shower though, to be sure). Then I spend the week after exams, which is this week for me, actually, sleeping and doing nothing, more or less unable to do anything productive. The last thing I need, is to go through this 4 times a year, instead of 2.

I understand that the current system favors "us sprinters," over "them marathoners," but a multi-exam system would just reverse the preferences, not get rid of them.

Also, law practice, in my limited experience, is punctuated with intensely stressful periods of a few weeks here and there. The one exam system replicates that, without sabotaging the learning that actually takes place during the semester. Midterms would distract from classes where no midterms are given, and generally "cramp my style." Studying for tests and learning concepts are completely different things and require different mindsets. The less back-and-forth in that regard, the better.
5.8.2006 8:26pm
Rob Johnson (mail):
Eugene,

I'm glad to hear that this is just your cursory analysis because I think it is especially poor. You completely ignore the fact that humans are risk averse, which would suggest a preference for multiple tests.

You also do not address what I think is the primary reason for single end-of-the-semester exams: they mask the extent to which grading is arbitrary. Professors faced with the prospect of multiple (blind) grading events would face the very real prospect of giving inconsistent marks to the same student throughout the semester.
5.8.2006 8:59pm
another law student:
Rob,
I agree that human beings are inherently risk averse, and thus drawn toward a multi-exam format. But I also think that human beings are inherently lazy, which draws them towards the one-exam format? If both of these are true, then what trumps what?
5.8.2006 9:36pm
Rich K (mail):
I'd like to weigh in on the professor's side here. In my ConLaw class this semester, we had group papers that were 35% of our grade due halfway through the semester. It may have been the format of the assignment, but the paper was universally reviled in my class. My property class had bi-weekly (on average) paper assignments that were only worth as much grade as the professor wanted to give it (so, pushing you from a B+ to an A- if you wrote really good papers). The papers got done, but I think for the majority of people it was more of a hassle than a great pedagogical tool.

It's interesting how fervent some people are about HOW HORRIBLE ONE EXAM PER SEMESTER IS, OMG. You end up covering all the same material -- what's the big deal? Why not give law students more free time?
5.8.2006 10:49pm
James D. Miller (mail) (www):
The primary purpose of law school is to reduce the supply of lawyers. The accuracy of law school exams are, therefore, far less important to the legal profession than the accuracy of, say, medical school exams are to the medical profession.
5.8.2006 11:30pm
Elliot123 (mail):
Both the multiple exam system and the final exam system measure a learning style. One style breaks up the course material into defined segments which are mastered in sequence, with a test at the end of each segment. The other takes the course material as a whole and relies on a single segment and a single test. Some people use one learning style. Some use the other.

Consider the student who masters the course material by the end of the term. If he doesn't master it on somebody else's schedule, he will be penalized by low midterm grades. What's the point?
5.8.2006 11:50pm
Lostingotham (mail):
I'm dubious as to the assumption that the random noise balances out in the end. Law student applicants to top law firms typically interview at the beginning of their 2L year on the strength of a single year's grades, so the grade sample is pretty small--as few as seven data points.

Those who get a summer slot are all but guaranteed a job on graduation. In my 1L year, the GPA spread between the 10% percentile and 50% percentiles was about 35/100 of a point. At the time fully a quarter (I counted) of firms who participated in early interview week stated that applicants must be in the top 1/3 of their class (granted this was at the bottom of the post-9/11 slump, so things may have changed a bit), and many specified top 10%. This was at Georgetown, so I imagine it was at least as bad if not worse at lesser-ranked schools. Now it is true that many firms will relax their stated grade standards on the strength of a strong interview (or good family connections, or whatever), but the fact remains that a student is in a much happier position interviewing with a 3.45 than with a 3.10, and the difference can be a bad flu on any given exam day.

That said, I agree fully with James D. Miller, above. The major function of law school seems to be to imbue a tedious and not-terribly-challenging (as usually practiced) profession with an air of exclusivity. In that regard it succeeds admireably.
5.9.2006 12:15am
Eugene Volokh (www):
Rob Johnson: You write "You completely ignore the fact that humans are risk averse, which would suggest a preference for multiple tests."

But I thought I had written, "Naturally, some students might value the greater accuracy of the multi-exam approach, and the lower risk to them that this approach provides (since they don't need to worry as much about tanking a class because they screw up one exam). But for many students the downside of yet more studying and yet more worrying exceeds the upside of less risk and better measurement accuracy. And on top of that, grades are on balance mostly a zero-sum game — more accuracy of testing will just rearrange the As, Bs, and Cs, so as many students will lose as will win. Increased time needed to study for exams is a net loss for students as a whole. Therefore, you'd expect that more accurate measurement devices will please only a few students, while greater need to study will displease many more students."

It seems to me that this takes into account people's risk aversion; suggests that it might be outweighed by people's laziness (see another law student); and points out that even risk-averse students might not prefer the multiple-exam format, since having everyone's raw scores clump together closer to the middle won't keep there from being the same distribution of grades as before.
5.9.2006 1:17am
Tom Tildrum:
Prof. Volokh, I have to wonder about your notion that there has been little "pushback" from students seeking a change in exam structure. What form could such "pushback" possibly take? When do students ever have any influence over course structure or content, let alone exam format? When does student pushback ever make itself apparent to a law professor, and when, to your knowledge, has student pushback ever had an effect upon a professor or a course?
5.9.2006 1:19am
Brian G (mail) (www):
I hate one-time exams, that is why I have taken several paper courses. If my GPA were based on my 6 paper course, I'd have higher than a 4.0 (thanks to all A's and one A+). If my GPA were based solely on graded exam classes, it'd be a 2.9. I have a 3.21 so it isn't the end of the world, but still, I HATE EXAMS.
5.9.2006 1:21am
WAL:
My guess you heard enough from me on the last thread. I'll try avoid much more in this one, but:

When you write


"even risk-averse students might not prefer the multiple-exam format, since having everyone's raw scores clump together closer to the middle won't keep there from being the same distribution of grades as before."



My problem is that it seems you're implying the outcome for students would be the same (people would have the same rank) and what would happen would merely be a case of the GPAs for students being closer together. If I'm reading too much into this and am about to come off as patronizing, my apologies, but I think that this implies the exact opposite.

The fact that raw scores would clump closer together would imply you're going to have people who are in different positions with multiple exams than if there was one. Say you have a few tests in a class where an A is assigned to the top 5%, A-s to the next 20%, B+s to the next 25%. On one test some guy gets A. The only way that guy will get a final grade that's closer to the mean is if on other tests he performs worse and another student performs better. It could be he winds up with the same ranking he would have received on his test. It could be that he winds up with the same ranking he had received on his second test. But it's impossible to receive the ranking at the end of the day that you would have received on all of them unless you perform the same on each. If you do, cool. If not, if you do great or poorly on the first one and then do average on the rest, then that's you'll wind up closer to the mean. You'll have a different outcome then you would have otherwise and that's the benefit of it. That's when you have an anomaly ironed out.
5.9.2006 1:46am
TFKW:
I had multiple professors justify the one-exam format by saying that their courses only really came together at the end, and testing students before the last few weeks would yeild fairly random results. I think there is something to this, at least for some 1L classes.
5.9.2006 4:37am
Public_Defender (mail):
Been there writes:


Perhaps a more interesting question for the Conspirators is the merits of closed-book versus open-book exams, and their various hybrids (e.g., those where you may bring a personally-prepared outline only, or only some primary source such as a copy of the Constitution).


Brian G writes:


I hate one-time exams, that is why I have taken several paper courses. If my GPA were based on my 6 paper course, I'd have higher than a 4.0 (thanks to all A's and one A+). If my GPA were based solely on graded exam classes, it'd be a 2.9. I have a 3.21 so it isn't the end of the world, but still, I HATE EXAMS.


Papers or take-home open-book (or partially open-book) tests do a better job of measuring what a student will be able to do as a lawyer than three-hour, in-class, closed-book, handwritten tests. I'd be interested in why the professor thinks so many law profs stick by the blue book model.

All that said, the real purpose of tests should not be to create a good curve or help employers. The real purpose of tests should be to encourage students to study in such a way that they will retain what they have learned.
5.9.2006 8:32am
Mr. X (www):
Anonymous 3L quote:
On your exam post, I didn't want to leave a public comment, however, I am one of those folks that hate multiple exams in a semester. It makes life a living hell. I barely have enough time to get everything done in normal circumstances (i.e., one exam per class), and multiple exams make the semester more stressful and busy. When a student has an exam, he ignores the world around him including other classes.
I had a law school class where the professor gave two midterms, seven pop quizes, and a final exam. I never worked hard in all my live for four credit hours. I easily did more work in that class than all the other classes combined that semester. And all my other classes suffered (sp?) because of the demanding class too. It was hard to keep up with them and the demanding class.


Umm, this just makes the point that 3Ls are lazier than final-year students in other disciplines and that the third year of law school is a joke.

In the same way that we shouldn't cater the law school exam process to what's easiest for the professors, we shouldn't cater it to please 3Ls who can barely manage to make it to class and aren't paying attention anyway.
5.9.2006 9:52am
Bruce Hayden (mail) (www):
The problem is clearly 1L. After that, I would suggest that the one exam format does a better job at tying everything togeter, and is probably preferred by the students, esp. the 3L, who are sliding by on less and less work, as the work gets easier and easier. One exam, for a 3L, is most likely preferable, as he only has to study once, and can blow off most of the rest of the term.

Having been burned by 1L grades, though I like the one exam format, I do think that multiple exams that first year make some sense. They give you a chance to recover. For me, my worst was Contracts where the prof graded on memorization of barely relevant cites, giving points for case name, date, and court name, but also rewarding for citing cases that any judge would throw out as obvisously irrelevant. This, of course, conflicted with all my other first term classes, as well as all classes I took from other profs in law school. A mid term would have done wonders for my grade (as evidenced by how I did with him in Contracts II), as I would have known how to study for his tests (i.e., ignore the theory, and memorize the couple hundred case cites).

Multiple exams each term for 1Ls would also be an easier transistion for most coming from their undergraduate schools. So, my proposal would be to implement a mid-term for 1Ls, and then go to the traditional one test per term for 2L and 3L.

As for papers, my memory is that you tended to have to work a lot harder for an A in those classes, but in my case, I hooked up with a prof who liked them, and essentially gave bonus points if you chose to write chapters for his upcoming books. So, I wrote a lot of papers for him, worked hard, but got A's because I wrote some of his book chapters. Also, when grading papers, a prof invariably knows who wrote it, since most often he worked with the student on choosing a topic, and may have helped along the way (and when you write chapters to a book, you work with him closely). One's relationship with one's prof shouldn't matter in the case of a paper versus the usual anonymous test, but I think often does affect grades there, at least at the margins. In short, they seem to be graded a little more subjectively. And this too ties in with work load, since that implies that blowing off class can affect your grade, at least a little bit.

I also had a couple of short answer type finals, and thought that in their context, they were a good way to go. They seemed to work well when you had a fairly broad area to cover that didn't all tie together into the type of issue spotting that you got in most classes. I should also note that my corporations prof used this sort of test to give bonus points for attending class, by throwing stuff out there that was topical (usually out of the WSJ that week) and then having a couple of those as short answers. I liked this format for Corporations, but think the regular issue spotting tests were better for Con Law, Contracts, Torts, etc.
5.9.2006 11:15am
Bruce Hayden (mail) (www):
The problem with pass/fail is that it probably doesn't grade finely enough for any schools out of the very top, like Yale. I am not sure of the exact relationships, but there, it is probably equivalent being in the top 1/4 or 1/3 (or 1/2?) to being in the top 1/10 or 1/20 of most law schools. And, I doubt that the system they use at Yale could really distinguish that closely.

I should note that when I was an undergraduate, we went to a pass/fail system (with honors a lot more rare than A's) my sophomore and junior years. But it turned out to be a major problem for getting into graduate school, and the school went back to grades my senior year, and has stayed there since. The school was forced to go back to the profs to calculate real grades for the pre-meds.

Finally, the solution to grade creep is mandatory curving. We had such, and the curve for any required class was fairly well mandated, say 20% A's, 35% B's, 35% C's, and 10% D's and F's. It was actually applied to all non-paper classes above a certain size, and, again, was one reason to take seminar/paper classes, which typically required more work, but also had a better chance for an A.
5.9.2006 11:33am
Marcus1 (mail) (www):
One other caveat: I'd say there's pretty little support for the idea that law exams aim to test ability as a lawyer. The only way they make any sense, in my view, is if they are deciding who should be a legal academic. Anything that makes it closer to a raw intelligence test, thus -- which a one-time deal probably does -- is probably considered good.

Law exams make sense to me as an intelligence test for those who pass the threshold of having spent the time to learn the material.

It's kind of interesting, though: It seems that the one end-of-semester exam is pretty scary for those coming into law school, but then after doing it, people seem to think it's the only acceptable way. The biggest factor may be that people like the devil they know.
5.9.2006 12:08pm
Mr. Mandias (mail) (www):
"Anything that makes it closer to a raw intelligence test, thus -- which a one-time deal probably does -- is probably considered good."

But don't you get more noise with a one-time test? Multiple exams that are closed book issue-spotters probably approximate raw intelligence more accurately. Though as a lazy student, I would rather take my chances with just one test.
5.9.2006 12:44pm
Marcus1 (mail) (www):
>But don't you get more noise with a one-time test?<

Yeah, more tests might improve overall accuracy, and would have a big impact on the few who have that bad day. Another factor, though, is that the change in quality of the top 10 or 20% -- the group for whom grades are considered to really matter -- would be very small no matter how many tests we had. E.g., a few smart people may get screwed by the current system, but I don't think more tests would weed out any more dumb people from the top 10 or 20%. So as far as the noise, I think schools and employers don't really care.

I think my basic assumption, though, was that more tests would increase the emphasis of diligence and studying over intelligence.
5.9.2006 1:22pm
Stryker:
as an Engineer:
1.) Why isn't anyone mentioning the BAR exam? What part does that play in grades (or is it taken enough later in career to be irrelevant?)
2.) I had a prof that planned on 3-4 tests (all formats, but mostly 1/2 untimed open internet) a semester, with the average around 50% on each. The top person had a 76%, and the number two person a 69%, with several people under a 30%. It was REALLY easy to see how well people understood the material. He also had notecard to call on people in class for questions--that'd affect grades if you were close.
5.9.2006 1:54pm
Different River (mail) (www):

In my experience, most law students are very serious about studying for exams, and thus spend a lot of time studying for them and stressing out over them. What's more, if you cut the significance of the exam in half, students won't think they have to study half as long or half as hard.


Right, but if there are two exams and they study the same amount as they would for one exam, but each time for half the material, won't they learn more?

Also, don't underestimate the information value of an exam partway through a course -- a midterm gives the student information necessary to adjust his/her thinking and learning in time to use it productively. The test is not just a measure of the student's ability; it is part of the learning process.

When I teach economics, I give two exams. Scores of good students increase substantially from the first to the second, since the material in the latter part of the course builds on the material in the earlier part, my policy is that if a students gets and A on the final they get an A for the course, regardless of what they got on the midterm. One risk of this is that students might not take the midterm exam seriously, but my experience is that this is not actually the case.
5.9.2006 5:26pm
Steve Sandberg (mail):
Maybe I'm idealistic, but I'd like to think the purpose of law school is to train law students to think, analyze and write better than they could before they started law school? If legal skills—including exam-taking skills, which typically translate into some type of enhanced legal ability—can be learned at all, then why not incorporate proven pedagogical techniques from other endeavors? For example, why not incorporate some type of feedback (perhaps immediate feedback is impractical) into legal learning? I think Different River has the right idea.

I would argue that good lawyers (and good clerks and academics) are made and not born, or at least made much better by practice. And maybe I'm naive, but I would hope that lawyer-training would begin in law school. Of course there are those that are naturally better at seeing issues or thinking clearly or writing. But getting better at those skills takes practice, and the most effective practice incorporates regular feedback. Sure it's a hassle to study for more exams, or to prepare and grade more exams. And there are pros and cons to different test types and formats. As a general matter, though, more exams will produce more feedback and more ability to practice and learn from mistakes. My personal-anecdote two cents: the content and skills I've retained the best from law school are from those classes I had to work at throughout the semester, and not just for two or three weeks at the end.
5.10.2006 6:45pm