some thoughts on exams

It's getting to be time to start thinking about final exams again.

I'm also giving two "in class" exams this week (I give four "in class" exams plus a comprehensive final. In some classes I drop the lowest "in class" exam score - partly because this saves a great many headaches re: missed exams. I can simply tell someone who slept through the exam, "Okay, that will be your drop exam then" and I neither look too much like an ogre nor need I listen to a litany of woes designed to make me relent and go, "Oh, okay, I'll write a make-up." Because writing make-up exams really stinks. It's time you have to put in that you didn't bank on, and invariably it takes longer to write the make-up because you have to avoid questions you used on the original exam.

It especially stinks if you have someone with responsibility issues who then fails to show up to take the make-up, as I had happen twice the semester before I instituted the drop-test policy.

I have gone in some classes to the policy of "if you want a make-up, it will be all essay" but I can't do that everywhere).

Anyway. I'm getting better at making up exams, I think.

I do provide "review sheets," I kind of hate doing it but it is sort of expected. (I rarely got them when I was a student). Mine are NOT an "exam before the exam" - I merely list topics that the test will cover and make particular reference to topics that are either not in the textbook or that I did not discuss in class but that the students were expected to read in their textbooks.

One thing I've learned about these sheets is that they can help ME - I take an extra copy of them and scan down and assign "point values" to each topic so I know how much to weight questions. I also make notes of topics that lend themselves well to multiple choice questions.

Yes, I use multiple choice questions sometimes, and I feel kind of guilty about it, because it's hard to write good (that is, challenging but not misleading) multiple choice questions. But I don't have a grader. And it gets awfully long to grade through even 25 all-essay tests. So I use multiple-choice where I can.

I'm also fond of using computational questions (if the students have learned how to calculate an index, or things like survivorship). Those are nice because they're a challenging question but they are objective to grade: the answer is either right or wrong.

I vastly prefer grading objective questions to grading subjective ones. I do make every effort to be fair: removing the student's name from the exam, grading all the same question at the same time, grading the exams in randomly shuffled order. But I still feel sometimes like I either get harsher or more lenient on certain questions as I either get angrier, or more tired and worn down, as the grading wears on.

I do include a number of essay questions on every exam (short essays and more rarely, a "long essay" that is worth 10 points - on a 100-point exam). I do this partly because some topics lend themselves better to that type of question, and also partly because some students are simply better at doing well on essay questions. I'm inclined to think a "mixed test" where you have a variety of question types is more fair.

I have learned on the essay questions to be as specific as possible. I often list subquestions I want them to answer, or I say, "Include in your answer a description of...." and then list several topics I want to see them address. I find if I leave the question more open-ended, some folks will go on for pages (and waste valuable time on minutiae) and others will give a one-sentence response that isn't complete enough - and then complain when I take points off for lack of detail.

I do make up "grading rubrics" before I read the finished exams - I make a list of what topics I need to see covered in the essay questions, with a number of points per topic. If everyone blows something, or if everyone leaves something out, I may reconsider, but starting from a "list of points" helps again with fairness of grading.

(One thing that makes me resist putting too many essay questions on? Some of my students have HORRIBLE handwriting. Worse than mine, and that's saying quite a bit. And they write SMALL. And so, I'm left using my hand lens to try to decipher some of the words. And I hate the feeling that I may grade them more leniently - or more harshly - just because I can't make out what they've written. I don't know that I could do "please come and explain this to me; I cannot read it" on the exam grading, it's too likely that some people might abuse that and just write a scrawl, and then take time to look it up after the exam).

I do write my exams new each year. I sometimes recycle certain questions - but I know exam files exist, and I know that they exist for my exams, so I don't do too much recycling. Or I reconfigure the question. Or I change the choices, if it's multiple choice. (I have one colleague who seems to have given up on the putting-effort-in thing who won't give exams back. This person lets students know their grade but they cannot keep a copy of the exam - I'm not sure they're even allowed to see the exam. I tend to go to the other direction - I'd like students to see where they made errors, see my comments and corrections, and then have the exam to study from for the comprehensive final).

Testing is challenging. I used to like it when I was newer to the teaching game; it meant a day I didn't have to prep before class. But I think writing the things and grading them takes me longer now than prepping for class takes.

Another thing: just as I hate with the passion of a thousand flaming suns when a student jokes (on a low-attendance day) "Do we get extra credit for showing up?" I also hate the person who pipes up, when I ask if there are any questions they have before the exam, "Can we do this as an open note test?"

I mean, yeah, to them it's funny - because they haven't heard it 20 times before.

Another thing that gets me - and that I wish people wouldn't do - is the whole B.S. line restate-the-question, fill-the-page-but-don't-say-anything. If you don't know, leave it blank. Or write what you do know, I give partial credit. But don't force me to wade through a pageful of B.S. just to find that you don't really know anything about the topic. (I realize for some people that might be, "If I start writing, I'll remember" but I think also - because I've heard it happens at some schools - in some classes the prof just glances and says, "full page - full credit" without reading the answer).

I will be honest: I wish I never had to test. But we don't have any other good way of assessing how much knowledge has been gained (and no, I don't believe "portfolios" are a valid replacement, especially in the sciences at the college level). I try to make the tests fair and as close to a real-world application as possible: I do not require students to memorize complex formulae; I assume they'd have access to those on-the-job. So I give the formulae on the exam but expect them to be able to apply them correctly - which I do think you'd have to do on a job. And my essay questions tend to be over the big, important, "capstone of the chapter" material.

But I still get complaints. I suppose every teacher does. In some cases, I want to tell people (but don't), "This is the world's tiniest violin...."

Tests were generally harder when I was a student, or at least I think they were.