A grading epiphany!

As has become increasingly typical, I was up late last night grading exams, rather than reading a good book or watching a movie or going for a walk, like normal people do, and I was getting a little bit frustrated. This was an exam for an introductory biology course, all first year students, and it was fairly straightforward: about 40% multiple choice questions, the rest being short “essay” style questions that had to be answered with a coherent paragraph. I had questions like, “explain the difference between methodological and philosophical materialism” (yeah, there was some baby philosophy in this course) and “summarize the work of Peter and Rosemary Grant”, all stuff that we’d discussed in class, and if they’d missed class, it was there in the lecture notes I’d posted online, and which they should have studied.

You will be shocked and surprised to learn that some of them had not studied.

What annoyed me, though, and ate up a lot of my time, was when desperate students who had not studied tried to bullshit their way through an answer, throwing out vaguely recalled terms, hoping that some of them stick. The Grants, for instance, who actually did work on finch beaks and adaptation, were assigned to have worked on Galapagos tortoises or iguanas, and seemed to have compiled a taxonomic catalog of random, memorable animals on the islands. You don’t even want to know what kind of inventions they created to explain away the philosophy or history of science, or how the Cambrian was after the Cretaceous. It was ludicrous errors of fact and random word association games, and worst of all, I had to carefully read it all to see if there was a glimmering of an echo of a shadow of comprehension in there, and give them points for it. Ick.

I think I was muttering to myself something like “why don’t you just admit you didn’t know the answer” when I had my idea. In the very first lecture in this course I had talked to them about the value of asking good questions, and how it’s acceptable for a scientist to say “I don’t know” when they don’t have a good answer, and I thought, I should encourage them to admit when they don’t know the answer, especially since I have a pretty good bullshit detector. So I’ve invented a new policy I’ll announce to them.

If you don’t know the answer to an exam question, just write “I don’t know” and I’ll give you 25% of the points. It’s that easy! It’ll save me the agony of trying to interpret word salad, which generally earns 0 points anyway, and you’ll get a few points for honesty. Everyone wins!

Of course, 25% is not adequate to pass the course — 50% earns a “D” grade — so you can’t expect to slide through by answering “I don’t know” to everything, but if you hit one or two questions you’re drawing a blank on, it’ll spare you some anxiety and suffering, and me some exasperation, if you can just dismiss the question and move on. It’s also fitting with a major theme of the course, which is about how scientists do science and how we came to understand principles of evolution, genetics, and development.

I’ve only been teaching for a few decades before I thought of this simple solution to a chronic problem.


  1. strangerinastrangeland says

    Hmm, somehow your “I don’t know, still get points” feels wrong, even when I know how you feel regarding the bullshit some students try to sell when they don’t have a clue. On the other hand, I currently teach microbiology and your solution might fit better with the more philosophical parts of your course.
    I enjoy teaching in small doses (it is mainly voluntarily for me), especially of course with engaged students, but I do dread the exam that is part of it and coming up.

  2. says

    As educators, it’s easy to forget that learning is an acquired skill and that not all our students are competent at it. I tried something similar with my secondary (high school over there) English students – if you struggle with some of the longer, analytical or evaluative tasks, write me a brief note at the end saying what you found tough. A lot of students who would rather not ask questions out loud in class were quite happy to ask them in writing. There were patterns so I could address most of their concerns in class instead of giving detailed individual feedback. A little more work, a lot more value.

  3. adipicacid says

    Thats… brilliant. And that is coming from someone who most definitely would try to BS my way through stuff I hadn’t studied enough back in the dark ages when I was in school.If I had been given that option, I’d have taken it, and furthermore, knowing that I would have to admit that I hadn’t studied to get the points, I probably would have studied harder.

    I’d be fascinated to hear how it goes by the end of the semester.

  4. asclepias says

    50% is a D? Wow! where I went to school, anything below 60% was an F! Not that a Dis any more acceptable…or did you have fat finger syndrome?

  5. says

    A “D” is basically an acknowledgment that you showed up, but is a fail. You’re allowed one “D” on your final transcript, but of course, it does hurt your GPA.

    A C- is a marginally passing grade, and in my classes, that’s set at 65%.

  6. snarkrates says

    An old teacher friend of mine used to have a stamp that marked a paper as “BULLSHIT”, which meant that he had stopped grading at that point and simply dismissed the answer as worthless. On the downside, that meant a score of 0 points. On the plus side, he also graded for proper English, spelling, punctuation…, so it was possible to get a negative score. The Bullshit designation may have actually been a gift in some cases.

    He would also award the bled-out carcass of a red pen to the student on whole exam the pen had expired. Fortunately, he graded on improvement.

  7. birgerjohansson says

    A British teacher once wrote “I return these otherwise perfectly fine pages of paper because someone has been writing rubbish on them”.

  8. HappyHead says

    Well, if you want to speed up your marking, you could always do what one of the random grad students I had assigned to assist me with grading tests did – throw random numbers at the papers decided mostly by whether the student’s name sounded female or male. As in, female sounding names got grades in the 80-100% range, and male sounding names got grades in the 30-70% range, completely unrelated to how they did on the test. The worst part was, this was on the multiple choice section, since I didn’t trust him to deal with the long answer parts. I had to re-mark everything.

    Our D range was 50-60%, and I believe you were allowed more than one of those, but if you got two or more Fs, you were required to take a year off before being allowed to ask permission from the department to re-enroll. I had one student who had to re-take the course five times before he passed.

  9. birgerjohansson says

    Going off on a tangent.

    The British government is “solving” the predictable shortage of lorry drivers by …. making it easier to get the driving license.
    If you take someone with 20 extra hours of driving instructions and put him in a 40-ton articulated truck, the road will be crammed with accident sites. But the prime minister will be gone then so it will be someone else’s problem.
    The idiot government has two modes.
    1. Saying no problem exists until everyone can see it is a barefaced lie.
    2. Saying it is someone else’s fault.
    Letting distribution fail and recruiting people with inadequate training is so very Boris.

  10. blf says

    It was routine(?) practice, especially in my mathematics (and physics?) courses, when taking exams (and class- / home-work) to be told to “show your work”. Which meant, more than once, I got partial credit for a nonsensical / wrong / bullshite answer, because the instructor / grader could follow along with my reasoning until the point-of-error. And in those-ish courses, the error could be as simple as a mathematics / arithmetic mistake. In addition, if I was able to show my answer was wrong (e.g., by units analysis), there tended to be more credit assigned.

    This tended to only be the case when facts, not style, is what mattered. A notorious example which still rankles me now (many many yonks later) is the grad student TA (“teaching assistant”, commonly known as “grad turkey”) who insisted (paraphrasing) no-one would ever design level-triggered interrupts (a HW design massively preferred by SW engineers, and one which I managed to convince multiple (plural) (micro-)architects to specify during my career (and which happened to be what my first multiple industry jobs used!)); to this day I don’t understand how the TA could have gotten it so wrong. That same grad turkey also assigned a D(- ?) to a design (calling it too complicated), whilst their fellow TA assigned an A(+ ?, “brilliant”) — in that case, the “failing”-TA, I suspect now, had no understanding of OO-design, whilst the other TA presumably did.

  11. blf says

    birgerjohansson@10, They™ are also trying to recruit more drivers (up to 5,000 out of an estimated 100,000 shortage) by easing visa requirements until the end(-ish) of the year: Meaning a short-term job for lower pay and much worst conditions, plus extra expenses (compared to the EU as-a-whole), in a country that will then throw you out again… Needless (almost) to say, just about everyone has pointed out that isn’t going to work… plus there are shortages in some EU countries, with much less hassle for EU-resident drivers in filling those positions (and for higher pay and vastly better conditions than in teh “U”K). Even “U”K drivers are pointing out how much better the job is in the EU, ‘Getting into Europe is a relief every time’: an HGV driver reflects on UK crisis.

  12. Artor says

    PZ, you could also try something like the legendary Metallica contract, where buried deep in the pages was a stipulation that their Green Room be stocked with a bowl of only brown M&Ms. That made it simple for them to walk in and see if the venue managers had actually read through the contract and followed the instructions, rather than finding out the hard way halfway through setup that some vital equipment had been neglected.

  13. whheydt says

    Re: blf @ #7…
    My wife (who was a Linguistics major) and a anothter student in the same rooming house were in the same section of Physics 10 (aka “bonehead physics”) at UCBerkeley. They would get together to do the problem sets each week. My wife would set up the problems, and the other student–who knew how to use a slide rule–would solve them and both would put down the answer.

    Came the final. My wife, reasoning that it was a physics course, not a math course, went through the test setting up all the problems and then went back to laboriously solve as many as she could in the time alloted. She got a very good grade. The other student, didn’t so as well as she really hadn’t learned how to set up the problems.

    Side note… Physics 10 was, formally, something like Physics for non-science majors. It could be used as a breadth requirement course for those far removed actual lab courses, such as English or English Lit majors.

  14. blf says

    whheydt@14, I think you mean me@11… ;-)

    Slide rules… I am of an age I learned how to use a slide rule in high school (actually, somewhat beforehand, as my father was a rocket engineer (joke from childhood, when someone said “we need a rocket scientist for this” — “Ok, let’s ask dad”)). This was about the same time as serious handheld electronic “scientific” calculators became available for an affordable price, and my family bought me one as a going-to-University present. (It was quite useful!) Anyways, I had been using one of dad’s sliderules, a very good (and probably expensive) engine, but eventually the crosshairs broke, so I went to the local shop to get it fixed(? spare parts?). Sadly, they’d stopped selling / servicing slide rules by then, but me (a teenager) and the proprietor (an elderly woman) had a long discussion about the obviously ending Age of The Slide Rule.

  15. mcfrank0 says

    Don’t feel to bad PZ, my eighth grade science teacher insisted that we were living in the Pre-Cambrian, no matter how many ways the class tried to point out that she was mistakenly reading the chart.

  16. blf says

    Artor@13, I believe that was Van Halen, not Metallica, and the contract called for a bowl of M&M’s but no Brown M&M’s, albeit for the reason stated… they were touring with equipment that need extensive safely protocols, and wanted a check the venue had at least read the details… No M&M’s, or the presence of Brown M&M’s, was a quick check they hadn’t, and hence a very Very careful check of the rigging, etc. (They would be doing that anyways, but that was a clew to triple-check everything…)

    Metallica’s James Hetfield did suffer bad injuries when pyrotechnics were detonated immediately next to him to due to poor safety protocols / miscommunication.

  17. mcfrank0 says

    And, yes I can appropriately distinguish between to/too/two. However, the spellchecker on my new tablet overestimates its capabilities for grammar and usage and I forgot to proofread my post before publishing it.

    I did, however, manage to catch the spellchecker’s attempt to swap out “it’s” for “its” in the previous sentence. Could this be a case of deliberate sabotage rather than an AI version of Dunning-Kruger?

  18. blf says

    mcfrank0@16, Ha! That reminds me of my 6th grade teacher (a really nice and usually very much spot-on gent) who insisted rockets carried twice as much Oxygen as Hydrogen because (paraphrasing from memory) “half the Oxygen is burnt with the Hydrogen to produce thrust, and the other half is ejected as something for that thrust to push against”. I knew that was bullshite (as did a few other students, and as mentioned in @15, ironically my own dad was a rocket engineer), but we couldn’t convince him that was mistaken. (I have no memory of asking dad to speak to that generally excellent teacher.) The teacher was mostly talking about Saturn V’s and similar rockets of the era, which did burn Oxygen and Hydrogen, producing steam (H₂O) as exhaust.

  19. says

    Two side corollaries:

    (1) The correct answer to any question — including the multiple choice questions — in law school begins “It depends, but this is what I think to start with:”. Most professors (and bar exam graders) don’t enforce this, because their egos tend to be too big. As an example, consider a multiple-choice question asking what constitutional provision guarantees a right to counsel in criminal proceedings. The “correct” answer is “At its textualist core, the Sixth; but when something is a criminal proceeding depends upon the Fourth, Fifth, and Eighth, and ensuring that it actually happens depends on the Fifth, Fourteenth, and Fifteenth.” Yeah, I’ve seen a multiple-choice bar exam question get beyond the first six words… ok, I’m lying, no I haven’t.

    (2) When Aldous Huxley (grandson of T.H. Huxley) took his exams at Oxford in 1916 for his English degree, he chose not to answer the “set essays” at all. He filled his papers with snide and sarcastic remarks about the assumptions built into the “set essays” (which are a bit longer and more extensive than typical American “essay prompts” or “questions”). The Master of Balliol College was both highly amused and impressed with the result, which showed greater understanding (especially in the two pieces on nineteenth-century authors) than expected of an undergraduate… and ensured that Huxley was awarded a First with Honours (equivalent of summa cum laude).

  20. Dennis K says

    The best part of this brilliant scheme is that it rewards honesty — not to mention establishes trust between you and the student.

    Would that it were I had such great a professor back in the day.

  21. BACONSQAUDgaming says

    I’m afraid I have to disagree with your “IDK”, partly because it just encourages students to just give up.
    Instead I like to tell my students to try to think, and work out a possible answer to the question. I give an example of a midterm exam question I received in University where we were asked why an algae would make acid in its vacuole. We hadn’t covered algae, so I suggested that like monarch butterflies/milkweed, it did that to taste bad to deter predators. It turned out I was right, and the only one who guessed correctly.
    Part of the scientific method is coming up with hypotheses to explain observations, so I encourage students to do that on questions they don’t remember the answer to.

  22. rockwhisperer says

    I like your solution, it encourages students to be honest. I went back to school very late for my MS and changed fields (engineering to geology). I found being a student in my late forties/early fifties much easier than the first time around, and part of that was comfort engaging with my instructors as fellow adults. As a result, they shared some things about their work that they might not with random students, and one of those was frustration with plagiarism, cheating, and BS-ing on short answer and essay questions.

    My second university was a commuter school. Students–even those who went directly to college from high school–had jobs, families, and so forth. They might be the first generation in their families to attend college. Money was tight, and the tension between working enough to pay (family) bills and studying was real. I attended more than one class where a baby dozed in a carrier or a young child drew pictures quietly at the back table, because a parent’s regular child care arrangement had fallen through that day. So professors understood that students would sometimes simply not be able to study, and giving a little credit for a simple acknowledgement of “I don’t know” would have been a kind gift to students and themselves. (And yeah, there were the kids who simply wanted to party, but they tended to drop out in their first two years, and I didn’t encounter them much. My professors didn’t worry too much about them, either, figuring that they’d drop out, grow up, and try again.)

  23. betterkevin says

    This is basically just what the College Board does with the multiple choice questions on the SAT. 1 point for a correct answer. -1/3 or -1/4 for an incorrect answer. It’s better to not answer than to get it wrong.

    My own personal grading story. In this one, the professors are the difficulty. One year in college, I graded papers for two different classes of Calc 2. Same catalog number. Same textbook. Often the same homework problems (odds from the book!). But two different professors. Professor A was very strict, and Professor B was, well, not. I graded the papers for A as harshly as I felt comfortable with. I graded the papers for B as laxly as I could without crying. A always said I was too lax. B always said I was too harsh. GAHHHHH.

  24. JimB says

    blf @19
    Ahh man. That brings back a memory. When I was in 11th or 12th grade. In a speech class. This one day the teacher passes out a list of 10 items. We were to number them 1 to 10 in level of importance if you were stranded on the moon.
    It was stupid.
    But there were conversations around the answers. And somebody said something and I said something about the Moon’s gravity being 1/6th the Earth’s.
    And the teacher then spoke up and said “There is no gravity on the moon”.

    Me: What?
    Teacher: There is no gravity on the moon!
    Me: There are rocks on the moon. Right?
    Teacher: Yes, there are rocks on the moon.
    Me: Why do you think they don’t float away?
    Teacher: Because they are heavy.
    Me: We call that gravity.

    She didn’t believe me. She might be the dumbest person I every personally met in 62 years…

  25. Pierce R. Butler says

    mcfrank0 @ # 16: … my eighth grade science teacher insisted that we were living in the Pre-Cambrian…

    Doubtless she expected all her students to attend Cambridge University.

  26. dragon hunter says

    Wow, this is actually brilliant! When I graded essay questions at university, this was the one thing that frustrated me the most. The fact that I had to sift through what were clearly nonsense answers, to try and salvage some points for effort… for a student who clearly made no effort.

    My solution was always based on not having a fail mark. If a student is happy to finish their university with a 25% or less mark average, then let them do it (it’s up to the employers to decide whether they want this person). But your solution solves all the problems really, because mostly these nonsense answers would end up getting about 25% of the point available for an answer that was not an answer. I’m not sure pedagogues would agree with this strategy though, but to me this is actually a brilliantly elegant solution.

    Another related frustration, was that teachers were always hard pressed by students (and therefore admin) to have the support materials online ready before the lessons. But when we plotted access to the online material over time, there was virtually no access, until a massive peak a few days before the exam.

  27. wzrd1 says

    I’m an odd bird with education.
    Test me, if there’s a question I have zero clue on, giving a zero and a reference is good. Being dyslexic, that zero point is a net gain.
    Way back in ancient times, I got loads of zeros for homework, but always aceed all tests.
    The idiot’s argument “your grades follow you around”, long dispatched as nonsense.
    The laugh is, I’ve instructed college courses, including a radiology reading class by special invitation of the professor.

    Science isn’t magic, love is, education, well, that’s both art and science.
    And yes, did a class at Temple on reading an x-ray, after answering the professor’s frustrated “case of the week” film and getting the baby’s lifespan correct.
    Gassy abdomen is the hint, imperforate anus being the diagnosis that arrived hours of agony late.
    Hopefully, that might’ve saved a life.