I've spent a lot of time this week writing exams. I give my Principles students three midterms, with the third one falling on the next-to-last class (which would be this coming Wednesday). Then I use the last class meeting to review everything from the semester in preparation for the cumulative final; in other words, I try to remind them of everything they've learned, which I hate to say, usually ends up being sort of a laundry list of concepts. I'm still working on a better way to review "everything"...
I know that many of my students hate my exams - I suppose that they would say I try to "trick" them. That's because for many questions, I purposely try to make one of the wrong answers something that will seem right to students who only have a superficial understanding of the material. I talk to a lot of students who do worse on my exams than they thought and when they see the correct answers, they say, "Oh, that's the other answer I was thinking of. I was trying to decide if it was that one or the answer I choose - they both seem right." I'm constantly telling students that as they go over their past exams to prepare for the final, it isn't enough to understand why the right answer is right, they need to understand why the wrong answers are wrong.
This is one reason why I give a cumulative final exam. I know students don't like cumulative finals - it seems like an awful lot of material. I try to alleviate some of their concerns by pointing out how everything we did for the last two-thirds of the semester is really just various applications of the core principles we discussed in the first four weeks but there really is no way to get around the fact that yes, it IS a lot to think about. But I've read two things recently that have convinced me more than ever that cumulative exams are important, so much so that I am considering making all of my exams cumulative next semester. One is a post from the Teaching Professor about cramming. It highlights some interesting research that suggests that a) cramming does work in the short-term (i.e., students who cram do just as well on exams) and b) students don't retain what they 'learn' when they cram (surprise, surprise). This suggests to me that if I can't differentiate between who has crammed and who has really learned for the long-term by performance on a test, then I might as well give cumulative tests that at least give students some incentive not to cram.
I've also been reading Terry Doyle's Helping Students Learn in a Learner-Centered Environment, which I want to write about a lot more once the semester is over. For now, let me just quote Doyle on the benefit of cumulative exams:
I know that many of my students hate my exams - I suppose that they would say I try to "trick" them. That's because for many questions, I purposely try to make one of the wrong answers something that will seem right to students who only have a superficial understanding of the material. I talk to a lot of students who do worse on my exams than they thought and when they see the correct answers, they say, "Oh, that's the other answer I was thinking of. I was trying to decide if it was that one or the answer I choose - they both seem right." I'm constantly telling students that as they go over their past exams to prepare for the final, it isn't enough to understand why the right answer is right, they need to understand why the wrong answers are wrong.
This is one reason why I give a cumulative final exam. I know students don't like cumulative finals - it seems like an awful lot of material. I try to alleviate some of their concerns by pointing out how everything we did for the last two-thirds of the semester is really just various applications of the core principles we discussed in the first four weeks but there really is no way to get around the fact that yes, it IS a lot to think about. But I've read two things recently that have convinced me more than ever that cumulative exams are important, so much so that I am considering making all of my exams cumulative next semester. One is a post from the Teaching Professor about cramming. It highlights some interesting research that suggests that a) cramming does work in the short-term (i.e., students who cram do just as well on exams) and b) students don't retain what they 'learn' when they cram (surprise, surprise). This suggests to me that if I can't differentiate between who has crammed and who has really learned for the long-term by performance on a test, then I might as well give cumulative tests that at least give students some incentive not to cram.
I've also been reading Terry Doyle's Helping Students Learn in a Learner-Centered Environment, which I want to write about a lot more once the semester is over. For now, let me just quote Doyle on the benefit of cumulative exams:
"The rationale for using cumulative exams is that they force students to review and relearn much of their course material by continually retesting the important information that was to be learned in each section of the course. In addition, cumulative testing helps students to see the connections between the information they learned in the first part of the course and the material that comes later." (page 45)Basically, by 'forcing' students to engage with the material multiple times, you are giving them a better shot at getting that material into their long-term memory. Students may hate it but chances are better that ten years from now, they will at least remember the concepts that made them so miserable.
I fully agree with the idea of cumulative exams. We are teaching certain fundamental principles. We hope the students understand them.
ReplyDeleteSee "Teaching and Helping Students Think and Do Better" on amazon.