<p>Here’s the official statement on grading:
“Grades at MIT are not rigidly related to any numerical scores or distribution functions, that is, grades are not awarded solely according to predetermined percentages. As can be seen from the following grade descriptions, a student’s grade in a subject is related more directly to the student’s mastery of the material than to the relative performance of his or her peers. In determining a student’s grade, consideration is given for elegance of presentation, creativity, imagination, and originality where these may appropriately be called for.” Source: [MIT</a> Course Catalogue: Academic Procedures](<a href=“http://web.mit.edu/catalogue/overv.chap5.shtml]MIT”>http://web.mit.edu/catalogue/overv.chap5.shtml)</p>
<p>Yes, but how do the students describe the grading process? For instance, can the current students please describe an example of a grade that they received and why they received it? When I was at CPW I witnessed a conversation between students that basically described a grading process in which the professor felt that the student had written an good paper but he suspected that it was written the night before so he gave the student an A-; the professor then indicated that, had the paper been longer, it would have received a better grade. Is this the norm and how does this work with the more mathematical classes?</p>
<p>so im guessing it’s hard to get into the classes he teaches now?</p>
<p>Well, humanities classes are graded like humanities classes are – I mean, a professor might hand out a rubric at the beginning of the term detailing his or her expectations for A papers, or something, but it’s ultimately subjective. Professors are pretty open to discussion if you feel their grading was unfair.</p>
<p>In most of my science classes, tests were curved to approximately a B average; that is, the average score was set to a B, and the distribution was worked out from there. </p>
<p>I think curving gets a bad rep at most colleges, but at MIT it is a very good thing – I strongly preferred curved classes to straight-scaled classes, because curving guarantees that somebody will get an A. Most test averages at MIT are pretty low, because the professors want to make the tests difficult enough to show the true range of mastery. So curving a test at MIT means that the class average might be 60%, but that’s still a B. (One of my friends took a test in which getting any points at all meant you got a B. ;))</p>
<p>Most professors set the A and C levels according to the actual distribution of grades. Apparently there are usually several discernible lumps in the grade distribution, so they draw a circle around each lump and call that a certain grade.</p>
<p>I took an essay writing class where the prof said on the first day that a good essay with no major problems was a C, a <em>really</em> good essay with few/no minor problems was a B, and an essay that he was glad to have had the experience of reading was an A. He said that he would be happy to give the whole class As, or to give the whole class Fs. He lived up to all of this. He was also a really great teacher, one of the best I had at MIT.</p>
<p>In math/science classes, it seemed pretty common for curves to be roughly B- centered, in classes that used curves. However, they could be a little wonky because frequently the grade distribution was not Gaussian. Classes that didn’t use curves stated this up front, and had grade cutoffs. In 7.05 (biochem), for instance, I think the minimum for an A was in the low 80s (out of 100), the minimum for a B was in the mid 60s, C was something like a 48, and D was a 35 (this was three years ago for me, so these may be slightly off, but they’re pretty close).</p>
<p>Yeah, in my year, the cut-off for an A in 7.05 (established at the beginning of the semester) was an 85, and this year, it’s apparently 80.</p>
<p>My daughter’s currently taking Quantum Mechanics 1, and apparently the class average on the last exam was in the low fifties. Like Mollie said, some degree of curving can be a very good thing.
:-)</p>
<p>This semester’s 7.05 (in it now) is straight 20-40-60-80 (D-C-B-A).</p>
<p>
</p>
<p>One of my comments on this – I have observed this is the trend at some of the more difficult schools around, and one thing about it somewhat saddens me. Which is to say that I somehow do not favor the very concept of difficult <em>exams</em> – I’m all in favor of challenging problem sets, but not exams. The reason being, I’d much rather be assigned a tough problem set on which I’m expected to answer almost every problem, struggling through the weeks (kind of how I am now) than take a tough exam where the best students are likely to get relatively low scores. </p>
<p>Maybe it’s not practical in a larger class to do what I wish, but I wish things changed for good in the upper level classes at least. </p>
<p>
</p>
<p>A perfect example – there doesn’t seem a point assigning such problems on an exam. I’ve never taken anything of the form, and I sort of cringe that they exist.</p>
<p>Do people still get Wall-Street like jobs (forgetting about the current state of the economy) if they graduate from MIT but have a mediocre GPA?
I met some people at MIT who party like I want to but (to my knowledge) most of their GPAs are at a mediocre level mostly because they don’t get a lot of their work done…
I plan to party a good amount in college. My friend, who goes to Duke, finds a lot of time to party and still gets his work done.</p>
<p>
</p>
<p>How do you want to party? I know people who go to parties every weekend - but they do their work during the week. You can’t party on weekends and be lazy during the week and do well.</p>
<p>I’m not expecting to party every night but I expect to party hardcore when I do.</p>
<p>
I do think there’s a point to it, though – they want to make a test which will create a distribution, which is to say they want to make a test that will display something like the true range of mastery in the class.</p>
<p>I’m in favor of these kinds of tests, although I was usually solidly in the middle of the grade distribution at MIT. I felt too much pressure on easy tests – easy tests mean you have to be perfect to get a good grade. Hard tests give you more wiggle room. I also felt there was something of value in difficult tests as a learning method. My human physiology class (7.20) gave notoriously difficult tests, but, man, they were fun! Moreover, you learned something by taking the tests.</p>
<p>
I can assure you that there are many, many people at MIT who party hard on the weekends, but buckle down and get their work done during the week. At that point, it’s about you, not about MIT.</p>
<p>Most people don’t have a problem taking Friday night and Saturday off from school. Even in my most difficult terms, I did homework on Saturdays only rarely.</p>
<p>
</p>
<p>Mollie, this is definitely a good point, and I may not have made myself clear – I definitely think <em>easy</em> exams are horrible, but I feel like debilitating ones seem to be horrible for similar reasons. On an easy one, well you have something like the SAT – miss one point, and it’s like missing a hundred. However, in a debilitating exam, kind of the opposite scenario occurs – your goal is to do as much as you can in the assigned time, given nobody basically can complete very much. But then, is that really a good reflection of the student’s mastery of the subject? I’d think not. </p>
<p>So again, I hate easy exams too. I took plenty of high school classes where I was very happy the teacher gave exams that made you think, because frankly there were 1000 people who could do long division more carefully than I, but fewer who could actually legitimately solve decently interesting physics problems or something. </p>
<p>I just feel like material gets complex enough that exams themselves seem not so great a way to assess students. Generally, I spend all week thinking about my problem sets, in the shower, on the way to class, on the way back, while talking on the phone, while posting here, etc. The number of ideas that occur to me, and that I toss around is invaluable in attaining whatever level of mastery of the subject I’m studying. And I feel like the pages and pages of work I sometimes turn in being replaced by a 2-hour exam covering 3 weeks of material or something (at least!) seems not so great. </p>
<p>So in short, if exams <em>have</em> to be there, I’d favor what I know one professor does – make them the sort of thing that doesn’t take much creative thought, and make them ask the sorts of things “every mathematician who’s taken this class should know” (replace “mathematician” by whatever you like) – not awfully easy, but the sort of thing you’d only know if you understood your stuff.</p>
<p>EDIT – one thing about having a totally problem set based class is that you never stop thinking about them…<em>EVER</em>. I have not taken a single weekend off; I end up taking time off during the day to recuperate, and then go back to work – i.e. breaks are self-created and come in short bursts. If my classes were exam-based, I’d probably be much more fine taking a weekend off, because I’d have a chance to balance my time over a larger period.</p>
<p>
Then I think we’re in agreement. 
To be totally fair to professors, I don’t think most devastatingly difficult tests are intentional. My husband’s aerospace engineering tests were often written with the express goal of getting a 70 class average – sometimes the class average was actually 70, and sometimes it was definitely not.</p>
<p>I guess some professors are just better at running classes than others. It’s kind of amusing – I’m taking a math class that is the second half of a sequence of math classes. All but 5 students dropped out of the first half, because the professor basically does not know how to phrase problems. There are only 4 students in my section, and <em>none</em> of them took the previous section – we all learned it by ourselves. </p>
<p>The two professors’ styles are orthogonal. The first assigned 13 problems per week (which is basically impossible to complete for problems at that level unless you do <em>nothing</em> but that class), but graded easy, and gave exams with 20-30% averages. The second one gives no exams, but expects <em>ridiculously</em> good quality work on substantial homework assignments, but not as bad. I think the second one’s style is precisely ideal.</p>
<p>Sometimes I feel the best solution to the exam situation is just to make difficult exams, but give longer time periods to solve these exams.</p>
<p>I feel like a lot of 50-minute exams aren’t difficult per se, but it’s very challenging to finish within the allotted time. What exactly then are professors trying to measure with the exams then? People who can think under pressure? I really think that for it to be fair, students should always have more time than they need and then the exam will measure whether you’re able to come to a conclusion with your thoughts. I mean, the people who don’t know their material will still don’t know their material if you give them 2 extra hours, but I don’t think it’s fair to discredit people who might know their stuff but just got caught up in some things and didn’t get to finish the exam on time.</p>
<p>OK 50-minute exams are RIDICULOUS. UTTERLY AWFUL. 2 hours is the minimum time I would accept as worthy. </p>
<p>My philosophy is: 2-hour exams for lower division classes, maybe a solid combo of a midterm or two + good problem sets for lower level upper division classes, and for upper level upper division classes, avoid exams, or at least make them take-home entirely. </p>
<p>Either way, echoing Oasis, let there be good time! Let ideas pickle, and let deep thinkers be rewarded.</p>
<p>…you guys have way too many opinions about exams.</p>
<p>
</p>
<p>Uh, what?</p>
<p>Most classes, appropriately IMO, schedule their non-final exams DURING CLASS TIME. This is because people have things to DO during not-class time. And most class periods are not two hours long. In fact, 55 minutes is probably the “standard”.</p>
<p>There’s no reason you can’t have a good exam in 50 minutes. It just has to be shorter than a two-hour exam, obviously.</p>
<p>Like LauraN, I’m confused as to why prefrosh who have never taken an MIT exam have so many opinions about them.</p>