Monday, April 19, 2010

Lessons in Frustration

I recently gave my second test of the year. The students did not do well. Even the students who did well on the first one posted only mediocre scores. It was a blood bath.

Part of this is undoubtedly my fault, but I went out of my way to avoid any subjects that were not discussed at length in class. I was not out to trick them or ask things that are only in the book, I kept it lecture-based and right from my slides that I provide them. While clearly I need to be better at getting some of these concepts across to them, there has to be some level at which I can feel like I have done all I can. I mean, they are the ones who are supposed to be learning, right?

As a case in point, I had a question on the exam worth 10 points (the exam was out of 90 points and only two questions were worth this much) that asked the to explain a method and how it works. I had covered this method in class the same week they performed in lab. As a follow-up, I discussed it again in the context of their results during the class after the lab. Based on all of that coverage, I figured it was fair game to ask about and have it worth so many points.

Virtually no one in the class got better than half credit on that question. WTF? They did it with their own hands! They walked through the whole procedure and the answers I got could have been the same answers they would have provided before I introduced it to them. To make matters worse, they were incredulous that I would ask a question about something they did in lab (and class, mind you). The class material and lab is planned out to be in concert and all of them felt that the lab was helpful for their understanding when I asked about in the mid-semester evaluation. But... they shouldn't need to incorporate the class concepts with the lab exercises?

I'm now struggling with how to go forward. I'm not going to hold their hand here. I will do m best to provide them with organized lecture material to help them understand the concepts I want to stress, but how much is on me?

Honestly, I don't think these students have any idea how to study (At least two have told me this), which leaves me feeling like I'm teaching in a foreign language sometimes. By the time I see these students, they are in their last year or two. If they don't know how to study it doesn't really matter what I tell them.

9 comments:

  1. Maybe it was partially a misunderstanding about what was testable material?

    My thoughts as a recent undergrad and now TA for a lab-based course: In my program, we often had classes with both a lecture and lab component. They complemented each other well and it was easy to see the connection as a student but, in the minds of the students, these are two separate components with their own time for evaluation. Perhapes the "method" type of question may have been something they were associating with the lab and did not realize it was possible testable material for the lecture component of the class.

    ReplyDelete
  2. I taught molecular biology in the fall and asked the same EXACT question on all three exams. No change in wording. Nobody improved on that particular question. You'd think that seeing it the second time might have prompted some of them to look it up or ask about it before the third exam.

    ReplyDelete
  3. On the long question.. was it worded as "explain this method... go" or did you articulate more what exactly you were looking for, along the lines of "give all the steps of performing this method, and note the scientific concepts behind them"?

    Yes, students SHOULD be able to organize an answer by themselves, but from experience I know that short-question-at-the-top-of-a-blank-page, please-explain-this-whole-process-to-me questions tend to produce underwhelming responses. Of course, it's a hard balance to strike... you want them to know what you are looking for, but you can't do too much hand-holding either.

    I'd say this is probably mostly on them, and not you, but if you want ideas about how to help them... 1) be as explicit as you can reasonably be about what you want to see in an answer when you write your test questions and 2) go through a few test questions in class and lay out exactly what you want to see in an answer.

    ReplyDelete
  4. If it was only a lab thing I probably wouldn't have put it on the test, but I was discussed twice in class and performed by them in lab. And I was very clear about what I wanted from them in the answer. It wasn't that they underwhelmed me with the response, they completely talked about the wrong thing. A few got it, but most blathered about something related, but not at all what I asked.

    There needs to be a term for then "answers" that students write when they don't know what the actual answer is. You know, the ones that are the test equivalent of verbal diarrhea.

    ReplyDelete
  5. Yes, but how did the distribution look? It's okay if the mean is low, so long as it's more or less a bell curve-?

    A few things can be helpful (assuming you haven't tried these already?):

    - Give them a review sheet with a list of material they are responsible for on the exam. On this sheet, you can list references to books/websites with advice and other resources (like campus services) re: how to study. This saves you from having to lecture on study skills, but it points them in the right direction if they are lost.

    This is also where you make clear, in writing, that lab class material will be covered on the exam. Fair warning. If no other professor has done this, they're just going based on past experience and you can see why they would assume your class is the same unless you say otherwise.

    - stick with short-answers, simple lists, multiple choice, or calculations for tests. Save your full-on essay-type question for homework, extra credit, or use in a 1-question pop quiz.

    Personally, I think lab classes are too cook-booky and usually students just follow the directions (or don't) but aren't forced to understand how things work.

    I've made the rogue suggestion before that maybe lab classes should involve experiments designed NOT to work, so the students are forced to troubleshoot and/or discuss what they think might have gone wrong.

    I think it's so much more instructive when they're not just following blindly along and getting the expected perfect outcome. And so much better as preparation for real research.

    ReplyDelete
  6. PLS- you did your job. They did not do theirs. Let it go. Or, take it as a teachable moment. If they really do not know how to study, and you are willing to donate 2 hours to their cause, the hold a exam review session and go over the old test. BUT, start this session by inspiring them to understand and link concepts in order to gain full understanding of the material. Only when this is accomplished can students begin to generate new insights and think creatively about XYX. Then, let them ask you about the exam material. BUT, refuse to answer questions that are 'what was the answer to question x?' Instead, ask them to rephrase the question so it deals with the concept you were trying to teach. Then dazzle 'em. BUT, realize only 5-10% of the students will actually appreciate your efforts.

    ReplyDelete
  7. For what it's worth, those long compound answers are not the best way to test students. They end up testing on a very small percentage of overall material, so it's really just a gamble of how many students missed that one particular thing. That's one reason why many people use a more diverse set of short-answer questions instead.

    ReplyDelete
  8. I'm in my second year and it still amazes me how poorly many students manage their studying and exam-taking time. As far as I can tell they really are studying, they're just putting their effort into understanding all the wrong things---and then when they take the exam, they spend all their time on the problems they don't know how to do and fail to complete the ones they do know how to do. Many of them seem to have no sense of which things are important and which aren't (no matter how much I emphasize certain things in class or in one-on-one meetings), as well as no sense of how much a single problem is going to affect their total grade (even though the points are right there on the page).

    ReplyDelete
  9. For what it's worth, those long compound answers are not the best way to test students. They end up testing on a very small percentage of overall material, so it's really just a gamble of how many students missed that one particular thing. That's one reason why many people use a more diverse set of short-answer questions instead.

    Off the top of my head I'm going to disagree with this. If done correctly an essay/short essay question tests a student's knowledge deeper and further than a set of multiple choice questions. You don't have the room you need to ask 10 MCQs/subject item, but one essay question can make students cover those points in varying levels of detail, depending on how you phrase the question.

    I would let mine know that partial credit would be awarded, not for blathering, but for at least trying. If they brain fart and mix up sympathetic & parasympathetic nervous systems, I can still determine if they have picked up the subject; they've just mixed two similar terms. If they get every pathway wrong, or every neurotransmitter wrong, i know they haven't learned a damned thing.

    ReplyDelete