A few questions about final exams.

Someone forgot to tell our department photocopier that finals started today; rather than being a vengeful photocopier toying with the pitiful mortals in its thrall, it was a happy photocopier that photocopied my final exams beautifully. And since I wasn’t clearing any cryptic paper jams, my mind wandered into the question of how others approach final exams:

  1. Multiple choice, essay, something in between, or a combination of question formats?
  2. Scantron forms? Blue books? (If so, do the students have to buy them or does the prof provide them?)
  3. In-class or take-home?
  4. Open book or closed book?
  5. Clever cheating-deterent procedures?
  6. Plenty of time to finish or barely enough time to finish?
  7. Pedagogically useful or a necessary evil?

My answers below the fold.


Multiple choice, essay, something in between, or a combination of question formats?
I hate multiple choice tests with the heat of a thousand nuns, so if it were just up to my preferences, I wouldn’t put any multiple choice items on my exams. But, I have some students who swear up and down that multiple choice is the only test format they’re good at, so I usually have a handful of multiple choice questions, some true/false, and some fill-ins (with a “well” of words or phrases from which to choose that is always slightly larger than the number of blanks that need filling). But I also have “short answer” questions that require a few sentences, plus your standard essay questions that require a couple handwritten pages.
As it turns out, doing well on the multiple choice and true/false items on my exams is strongly correlated with doing well on the essay questions.
Scantron forms? Blue books? (If so, do the students have to buy them or does the prof provide them?)
To me, it feels like piling on to make students sit for an exam and to make them buy the instruments of their own torture. (Not that I set out to torture them, obviously.) So even though we now have vending machines on campus from which one can buy Scantron forms and blue books, I make a point of providing all the necessary space for answering the questions within the exams themselves.
Actually, I’ve heard stories about students “preparing” their blue books (by, say, writing out their answers to pre-announced essay questions) prior to exams. My colleagues who administer bring-your-own-blue-book exams head this off by requiring students to contribute their blue books to a stack at the front, from which blue books are randomly distributed to exam-takers, along with the exams.
In-class or take-home?
In-class. For one thing, it cuts down the amount of time during which students are actually writing (and thus the number of words that I need to read). I like to increase the ratio of time thinking to time writing. For another thing, take-home exams seem to bring with them pleas for extensions (especially from students who manage to forget what the due date is). In my experience, students have a better shot at getting the exam in on time if they’re actually showing up to take it in a specified exam period.
Open book or closed book?
Mine are mostly closed book, but I allow students to prepare a set of notes (usually a single 8.5″ by 11″ sheet, front and back) that they can use during the exam. I’ve seen some students approach open book exams like scavenger hunts, taking a lot more time to rifle through books and notes than to think about the questions or write sensible answer to them. However, my memory and I have not always been on the best of terms, so I understand the anxiety a completely closed book exam can provoke. My students seem pretty happy with being able to put whatever information they think they’ll need on their set of notes. And, having devoted some thought to what they should put in those notes, many of the students end up not needing to refer to them much during the exam.
Clever cheating-deterent procedures?
Aside from being in the room while the exam is happening? Not really. Of course, essay-type questions (even “answer in a few sentences”-type questions) don’t seem to lend themselves so well to cheating.
Plenty of time to finish or barely enough time to finish?
I try to design my final exams so there will be plenty of time to finish, even if one has a mini-freak-out during the exam. Judging by how long it actually takes students to finish the exams, I usually achieve this design goal pretty well.
Pedagogically useful or a necessary evil?
I used to view exams as a necessary evil, but it turns out that some students are much better at showing what they know on an exam than in papers (which I also assign, of course). For these students, exams not only give me more information about how well they understand various concepts (and how they fit together, and how to apply them), but often give a significant boost to the grades of students who have been struggling with the papers.
Now, as to whether grades are pedagogically useful or a necessary evil, that’s a question for another post.
So, what are the exams you’re giving (or taking) like?

facebooktwittergoogle_pluslinkedinmail
Posted in Academia, Personal, Teaching and learning.

8 Comments

  1. …the heat of a thousand nuns

    I must lead a sheltered life … this is not a metaphor with which I’m familiar.
    All I can say is that I’m glad that I’m a good number of years past exams …

  2. 1. MC questions out of necessity (huge lecture class), but I also like MC questions because they ensure fairness, and a reduction of begging for one more point. Also, since intro (discipline) is more of a facts than ideas course, it’s an easier format for the students
    2. Scantrons — see above. I hadn’t heard of this forcing students to buy them idea until this year; it seems like a bad idea all around (prefilled, forgetting, etc.), so do you know why it’s done?
    3. In-class, because of size & because of the purpose of the course
    4. Closed book — see #3
    5. Not clever, but yes: multiple forms, me & proctors walking around the room. I still had someone take the test for someone else, though; didn’t really work when s/he put the own name instead of the student’s name on the exam.
    6. Plenty of time. Just works out that way.
    7. Pedagogically useful. No papers, so only way to grade. (And students here aren’t going to even try to learn anything if they aren’t going to be graded on it…)

  3. 1. Mixed format – short answer plus problem-solving (writing spreadsheet formulae, evaluating expressions, coding database queries, converting numbers between bases, etc.). Also a large lecture class, but plenty of TAs to grade. Each TA grades one page and becomes an expert on that page, so even if they make mistakes they tend to be consistent across 400 exams.
    2. Neither scantrons nor bluebooks. I print the master copy on my laser printer at home as a cover page plus 6-7 pages of questions (around 20-25 questions overall), then give to the department to make 400 photocopies, single sided, stapled. Students answer on the exam form itself. Takes the TAs 7-10 days to grade the midterm, slightly less for the final.
    3. We reserve big lecture halls in the evening for the midterm; one of the gymnasia at the final exam (the only time all 400 are in the same room at the same time).
    4. Open book. I did closed book for a few years, then realized that open book more closely matched reality (when working on a real-world problem you generally have your reference books nearby) and lowered the overall stress level without unduly affecting the class average.
    5. At the midterm students are required to sit with one empty seat separating each pair. At the final the gym has the desks already configured with 8-10 feet in between. Wandering eyes have to really stretch, and become obvious. After one warning they are very publically moved to a more remote seat.
    6. Some finish early (two hour exams), others have to be removed with a cattle-prod. Many find my exams to be pretty long.
    7. Useful, and evil. Part of the game for which (nearly) everyone understands the ground rules. Computer projects constitute a major portion of the final grade, so people who do not test well can still come out OK overall. What I have never really understood are the people who show up at the final exam with so few points gathered across the semester that they would need 300 points on a 100 point exam to pass with a “D”.

  4. I never learned what mom preferred in finals, other than essay questions. Straight essays up and down. To make it real fun, she insisted not only on getting it right, but on supporting your conclusion with a good argument, and insisted on good grammar. Came with being a dual threat biologist and English teacher. She was known for docking points for poor composition, and really penalized students for bad grammar and spelling.

  5. I’m an undergrad student, and a trend I have noticed is that many professors give all multiple choice finals due to time constraints on grading, especially in large classes. For example, in my genetics class all of our unit tests were a mixed bag of MC/problem/essay questions. They always involved complicated problems that had to be worked out and diagrammed, requiring all the work to be shown for full credit, plus essay/short answers on concepts. On the final, though, all the questions were more general and were done on scantrons, because the professor only had 48 hours to score 250 exams before the grades had to be logged into the system.

  6. Coming from a completely different tradition (the other side of the ocean, over the hills and far away and further still) I do not have the same problems as you, as most of the formats are not known here.
    1. Never MC – the 25% random factor is discouragement enough. Usually it is either a combination of ID/essay questions or essays only, depending on the course.
    2. Neither – I always prepare my own exam and answer sheets.
    3. In-class only; take-home exam would not work here, as students would most certainly take unfair advantage of it.
    4. Usually closed book as this is the traditional way here. This year I’m experimenting with the open book format to see if our students can cope with it.
    5. Atmosphere of terror in the examination hall usually works just fine. Sometimes I am tempted to use a “mystery student”, whom I would remove from the hall early on :), but normally our students seem to be an ok bunch in this respect.
    6. Plenty of time, definitely.
    7. Useful, both for them and for myself.

  7. I have only ever sat exams, not set them, but here are my answers.
    1) I dont recall ever seeing a multiple choice in my undergrad or Masters courses.
    2) Whats a scantron? And whats a blue book (apart from a book that is blue)? Is it a book in which you write the answers? Do students REALLY have to buy their own of those? Thats pretty incredible to me.
    3) Take home exams? Thats not an exam, its a piece of coursework that I would take to the library and use all the knowledge the world has to offer upon.
    4) I never had an open book exam.
    5) Exam conditions in a large hall.
    6) The tradition seems to be 1 mark per minute in any exam. Which for me has always been just about perfect. I tended to finish the exam and have just enough time to do a cursory check at the end.
    7) Vital for me.

  8. Whats a scantron? And whats a blue book (apart from a book that is blue)? Is it a book in which you write the answers?
    A scantron sheet is a standardized form with preprinted bubbles for answering multiple choice questions. You fill in the bubble of your answer with a #2 pencil, and after the exam the machine scanner can read and score all the answers for all the forms in just a few minutes. Very efficient for grading, but embodied with all the flaws of multiple choice questions.
    A blue book is a small standardized booklet, about 9×9 inches, with a blue cover, containing a few blank pages of lined low-quality rag paper. I used these in college in the 1970s for writing essays in my English literature exams, but I don’t ever recall using them in science or math classes.

Leave a Reply

Your email address will not be published. Required fields are marked *