New Grading Scheme: Presentations Fail

A plus

Last week, I talked about how determine my students’ grades up to a C (by the way—the students are doing much better on the quizzes than I expected. Most students should finish up on Monday, the last day of grading. I expected many, many more students to be struggling to meet the requirements). This week, I will discuss how they earn a B or an A.

There are two components in determining grades above a C: presentations and a take-home final exam.

The exam was pretty standard. The one comment here is that I could make it fairly difficult, since it is really aimed at differentiating the C-students from the B-students from the A-students.

For the presentations, I essentially made a list of 100-200 homework problems for the semester, and doled them out to the students. I assigned 15 problems to be presented per class period for the second half of the semester, and I told the students to spend at least 10 minutes trying each problem before they are due. The purpose of this was so that the non-presenting students would get more out of the class (spoiler: I don’t think that students actually looked at the problems they were not planning on presenting).

The night before class, students request (via Moodle, our classroom management system) to present as many or as few of the assigned problems as they like. The next morning, I assemble an “itinerary” of presentations. The presenting student comes to class with the substance of the presentation written out on notebook paper, and then presents the problem with a document camera (I receive no money for linking to this camera. It is simply inexpensive and I have been happy with it).

Here is how the presentation grades were determined (I heavily borrowed from Ted Mahavier for this, who has a lot of experience doing this):

  • D – You attended every class, paid attention, and tried (mostly unsuccessfully) to present at least a few times.
  • C – You fulfilled the requirements for a D, and you had a few successful presentations.
  • B – You fulfilled the requirements for a C, and you had many successful presentations.
  • A – You fulfilled the requirements for a B, and you had many successful presentations of difficult problems.

My goal for the presentations was to create a Modified Moore Method-type atmosphere in the course. The problem is that the audience for each presentation zoned out after 2-3 weeks. Thus, I think that the presentations were probably very helpful to the students, but not very helpful to the students in the audience. I did a pre-test and a post-test using the Calculus Concept Inventory, so we will see if the data confirm my skepticism about this teaching method.

I like the overall structure of the course (Peer Instruction for the first half to give a good conceptual foundation, and then some sort of IBL thing in the second half for reinforcement and details), but I will likely not be using this presentation style again. The best format I have so far is to return to my <a href="http://en.wikipedia.org/wiki/Cooperative_learning"Cooperative Learning roots and do something like this:

  1. Do maybe 4-7 problems per day as homework.
  2. Have the students work in teams on one assigned problem at the beginning of class so that everyone really understands it (after getting a head-start on it from the homework).
  3. Randomly call on a team member to present the problem.
  4. That random team member’s presentation grade is the grade for everyone in the team.

This will help the students teach each other. I would love to hear feedback and other suggestions (and I apologize for typing and weird formatting; my son just woke up, and I probably won’t have time to proofread until Monday. Since I want this off of my to-do list, I decided to publish without proofreading).

(Image “A Plus” by flickr user s_falkow)

Advertisements

Tags: , ,

24 Responses to “New Grading Scheme: Presentations Fail”

  1. Andy "SuperFly" Rundquist Says:

    It’s so interesting how I can read about someone’s plan and think “this is awesome! There’s no way this can fail because it’s so well thought out!” I really did (and do) like what you did this semester. It’s so interesting to hear this postmortem on the presentation part. The biggest problem seems to be the audience one, right? What was the expectation of the audience? Did they tend to ask questions? Could you do Kelly O’Shea’s mistake game?

    • bretbenesh Says:

      Hi Andy,

      Thanks for having misplaced faith in me.

      Yep—the big problem was the audience. Something like the Mistake Game is needed. At this point, everything is on the table. I also thinking that giving credit for catching mistakes and making mistakes is possible: perhaps these could be standards.

      Thanks for reminding me of the Mistake Game. Do you have other ideas? Bret

      P.S. It could be that I am overreacting right now. Three students asked me about the final exam, and they all had the same huge misconception about the Fundamental Theorem of Calculus. Also, I graded some of the Calculus Concept Inventory post-tests yesterday, and the scores seemed really low.

      But I just looked at the pre-test scores, and they were even lower. So maybe it is not as bad as it seems. And things seemed to go well in my multivariable calculus class. Perhaps the biggest difference is that calculus I is essentially a class for freshmen, and multivariable is essentially a class for juniors (it is linear algebra-based). I will survey the multivariable class to get their opinions about the presentations on Tuesday.

      Regardless, I think that this can be improved upon in both class.

  2. danaernst Says:

    I’ve had many discussions about how to select presenters for an IBL/Moore method course. Strategies seem to fall into two categories:

    1. Students know in advance who is or might be presenting what.
    2. Students do not know who is presenting what.

    I’ve tried methods from both categories and in my experience, category 2 works out better.

    One obvious disadvantage of doing it this way is that you may have absolutely no idea what is going to happen when class starts. In particular, I often have no idea whether all the assigned problems will get presented/discussed. The first couple semesters, all the uncertainty was stressful, but now I love it.

    • bretbenesh Says:

      Hi Dana,

      Let me see if I understand this: on a typical day, you might assign, say, 8 problems for the students to do. When class starts, you might say something like: “Galen, could you please do the first problem?” Then Galen presents whatever work he has, and a great discussion follows.

      Then you might say: “Gaius, could you please present the second problem?” Repeat.

      Is this roughly correct?

      (I think you already know this, but) I have been having them tell me which problems they are willing to present. You do not do this at all? If they don’t have the problem you assign to them, they just don’t present it?

      Thanks for the help. Bret

      • Dana Ernst Says:

        On a typical day, I ask for volunteers for all the problems at the beginning of class (or possibly as we go) and if no one is up to the challenge for a particular problem, then any number of things could happen depending on how many problems fall into this category, how much time is remaining in class, etc. Here are some options for what might happen if no one volunteers:

        – I pick someone.
        – I ask if someone would be willing to come to the board and act as scribe as I nudge the class along via some Socratic-type questioning.
        – I break the problem up into 2 or more key lemmas and have the students work on them in groups.

        For the past few semesters, it has been extremely common for students to have a list of who is presenting what written on the corner of the board when I come in. I don’t require them to do this. Sometimes I modify the list they produce. For example, if Sally has been presenting all the hard problems lately and her name is on the board, then I tell her that she is off the hook that day. Also, sometimes there are 2 or more names written next to a problem. In this case, I pick the student that has presented the least (unless one of the students is specifically trying to redeem themselves after a poor presentation the class session before).

        I really like that the students don’t know exactly what is going to happen each day. However, they do know what they are supposed to get done each day and they seem to (try to) do it all. The audience is generally engaged since every student had been working on the problem that is being presented. Moreover, I often will ask members of the audience questions as the presentation is going on. This keeps everyone on their toes, as well, but I’m not doing it to stress them out.

      • bretbenesh Says:

        Hi Dana,

        I forgot that I already knew that. It seems to me like we are doing similar things, only I do things electronically (https://symmetricblog.wordpress.com/2012/11/02/matching-students-to-presentations/).

        The one difference: if no student knows how to do a problem, I leave it open and challenge them to keep working on it. I do not know if this is good or bad.

        I also do not ask the audience questions—that might be a good thing for me to start doing.
        Bret

  3. TJ Says:

    I have two thoughts:
    1) holy cow! There is a calculus concept inventory! Is it any good? & why didn’t I know this! (Last one is rhetorical…)

    2) What kind of questions arevyou asking during presentations? The level of suspense might be the issue. I am reading some of Guershon Harel’s work on intellectual need, and maybe that is what is missing? It is hard to tell without examples…but that would be a common reasons why students tune out.

    • bretbenesh Says:

      1) Yes there is! I don’t know what a CCI should be like, but my opinion is “It is okay as a test. It is great that it is a common tool for us all to use.”

      2) I built a foundation of more difficult problems, and then threw in easier problems in between. There were two reasons for the easier questions: I wanted to get more students involved, and I wanted students to review the material.

      Here are some sample questions:

      Hardest: Let f(x) equal x^2 on the rationals and 0 on the irrationals. Prove that f is differentiable at x=0.

      Medium (on the hard side of medium): Come up with a rule for the derivative of f(x)^(g(x)), where f and g are differentiable.

      Medium (on the easier side): Here is a graph of f’. Sketch a graph of f if f(0)=3.

      Easy: Find the derivative of x^pi + pi^x.

      I would love to know your opinion. I definitely think the problem set could improve. If I just wanted to tweak things for next time, I would cut down on the number of problems (from 180), and the problems I would eliminate would be the easiest ones. Bret

  4. identityelement Says:

    Here’s something I tried in linear algebra last summer that worked pretty well. Everyday, I had assigned “sign-up” problems due. These were in addition to ” hand-in” problems that were collected and graded once a week or so. At the beginning of class, I passed around a sheet with a list of students (labeling the rows) and a list of problems (labeling the columns), and the students checked off all the problems they felt they could present to the class. The important thing is that students earned credit for the number of problems they signed up for, not the number they presented. In fact, we didn’t have time to put the problems on the board every class. But when we did, I would go through the list and pick a student to present the problems I most wanted to have presented. I tried to balance the goal of getting the more challenging problems presented with the goal of giving every student a fair opportunity to present. I had hoped that students would be prevented from signing up for problems they had not actually worked on by the fear of humiliation. This seemed to work, with perhaps the exception of one student ocassionally.

    • bretbenesh Says:

      I hadn’t considered the idea of giving the students credit for _signing up_ for a problem. That is really interesting, and I will strongly consider that for later semesters. Thanks!

  5. TJ Says:

    I go with the most open-ended volunteering model possible. Students typically have a list of four to eight questions in front of them at any time. I simply ask for volunteers and choose from the hands that go up. Then I have to manage things a bit like Dana discusses above. It generally works. When I get no volunteers, I make something up. Dana has two good ideas listed. Sometimes I just pick one off the list and tell them to find a partner and try it for twenty minutes. This semester that worked well, we got several problems completed that way.

    I don’t have an opinion on your tasks as written. I just wondered about keeping things in the right zone. I am sure you have thought a lot about that already.

    • bretbenesh Says:

      I have already thought about the tasks; my concern is that I could be thinking about them wrong. That is something I will just have to learn from experience, I suppose.

  6. Vince Says:

    This has been a great read (thanks to DE for pointing it out on the IBL G+ community).

    This is all quite new to me. All the teaching I’ve ever received has been purely lecture based. As a lecturer I really want to emphasise pre learning, peer learning and student lead learning (there’s got to be a better use of “lecture” time then me just delivery information now a days)…

    I’m going to be teaching a programming course next term (the students will learn SAS and R in parallel). I have 4 full mornings worth of teaching (split in to 2 for both languages) and was planning on setting the students “challenges” before the session. The idea being that then the “lecture” would take care of itself:

    – Students would present
    – Students would discuss

    Ideally I wouldn’t actually open my mouth but I doubt that will happen. I guess the hard work is just making sure that the learning opportunities I create are the correct ones…

    My biggest worry is student participation/engagement. If students don’t buy in, I guess this falls flat on its face. I’ve been reading Theron and Dana’s posts on G+ about IBL with interest for a while now and they often mention how students come around after a while.

    Is the approach I’m describing actually IBL? I’ve had a hard time conceptualising what exactly IBL is (as opposed to project based learning for example).

    Also, with regards to groups, I will in essence have 8 “challenges” over the 4 weeks of teaching. I was thinking about splitting the classe in to 8 groups (of about 4) and getting the groups to present. I was planning on not telling the students which group would present when just so that as Dana says:

    “The audience is generally engaged since every student had been working on the problem that is being presented.”

    I wasn’t actually planning on any of this counting towards their grade. If that a mistake? With no incentives will the students just not bother? Should I offer bribes? Chocolats at the end of the 4 weeks for the best group??? 😦

    Sorry for my ramblings but given my lack of experience in this approaches I can’t offer much more 🙂

    • Dana Ernst Says:

      Vince, the bottom line is that you have to do what you think is right. Not having presentations count as part of the students’ grades is not necessarily a mistake, but it is worth thinking about. You know the type of students you have. The “right thing” to do depends on the teacher and the students.

      Here’s a thought. If you tell them that presenting is part of their grade and it ends up not being the right call, you can just give everyone full marks in that category, which will likely resolve any issues. However, if you it isn’t worth anything and they aren’t playing along, what do you do then?

      My goal is to play Goldilocks with this part of their grade. I want to make it just enough to make them play by the rules, but not so much that it causes more harm than good. The percent that presentations are worth in my classes varies from class to class. In my lower-level courses, presenting/participating is worth as low as 5%. In my upper-level courses, it is worth as much as 30%.

      Lastly, chocolate is always a good idea.

      • Vince Says:

        Thanks Dana, I like the idea of going with credit and simply giving everyone the credit as long as they ‘play along’. I could possibly start out by saying that they have the credit and all they need to do to keep it is a minimum level of engagement: not even necessarily successfully completing the challenges.

        I imagine certain issues arising with regards to some group members not participating but I don’t see how to get around that… (I’ve been posting on G+ about a game theoretical approach to that issue so that might help…)

    • TJ Says:

      Vince, I would say that you just want to play. if students have the idea that you want to help them and you aren’t going to be a hard-ass at grade time, you will get a lot of slack.

      are your students graduate students? if so, my adviser always used to say, “everyone here will receive the standard grade for this course. those who do an exceptional job will get the standard grade+” that quite effectively got the point across that he wasn’t there to play games, and you were responsible for your own learning. That flies for motivated graduate students, who know that GPA is nothing, but a good letter of rec is important.

      • Vince Says:

        This course is on a 1 year taught MSc program and very year the students are very different. The motivations are always all over the place. With this particular year I think incentives (‘hardass-ness’) are needed… I think I’ll go with giving them credit at the beginning that they can lose if they don’t do a minimum…

    • bretbenesh Says:

      Hi Vince,

      Thanks for commenting. First, don’t worry about your lack of experience; the guy who writes this blog isn’t all that experienced, either.

      I will sort-of-kind-of disagree with what Dana and TJ said: ideally, I think that you would not grade the presentations (I generally regard grades as a nuisance that is forced on me, although I grant that it is complicated). I think that there are two key questions to ask to determine whether you can get by without grading the presentations:

      1. How motivated are the students? As TJ said, you can very easily get away with doing this with graduate students. With first year students, it might be a disaster, and you might want to follow Dana’s grading advice.
      2. How interesting are your questions? (It is a positive sign that you are calling them “challenges”). I find that I have no need to grade my students on the “clicker questions” I do in my class because the activity is interesting enough, I know how to “sell” the idea to students, and I am at the point where I can write interesting enough questions. I am NOT at the point where I can write interesting enough presentation questions (otherwise, I doubt I would have written this post). I leave it up to you to determine how interesting your challenges are.

      I suppose the safe thing is to follow Dana’s advice this semester; if things go really well, then you can choose to un-grade the presentations the next time you teach the course (if you do).

      P.S. I like the idea of teaching two languages in parallel. Keep us posted.

      • bretbenesh Says:

        Also, I have no idea whether your course would be IBL or not. TJ, Dana, and I had a sizable discussion about what “IBL” means earlier this semester. I still don’t think that I know, but I am two steps closer now.

        Regardless of whether it is IBL or not, what you are doing sounds interesting, and you should blog about it.

      • Vince Says:

        I think the challenges are interesting but I guess its hard to say. They are designed to be relatively ‘easy’ as long as student read through the notes, look at the videos and generally think a tiny bit out of the box. As it’s a programming course, it’s all about finding the syntax at a minimum and possibly finding better ways of doing things. For example one of the last challenges is to plot the daily number of tweets with a particular hashtag over a year. If the students just look through the lecture notes (which as well as screencasts is all availabke to them before any of the scheduled classes) they can do it all. If they look into it a bit they would find a particular way of handling dates that make it more efficient. What I’m hoping would happen in class is that the basic approach would be presented and some students in the group would point out the better way, or perhaps the better way would be presented and students would ask what the hell is going on. At that point ideally the students would discuss without me actually getting involved and teach each other. That’s the optimistic POV I currently hold..

        I completely agree re grading but as its a bit brand new to me I think I will make a token percentage of the mark if only to see how things go this year and have a bit of a safety net…

        I can’t say I blog regularly (enough?) but I certainly post on G+ (too much?) and will post about this stuff on there. I might also blog about it … I’m in the middle of a higher education teaching certification process. I have to put together a portfolio over the 4 modules I’ll be doing. I’ve completed module 1 and that bit of the portfolio is up on my site if you’re interested:http://www.vincent-knight.com/home/teaching/pcutl

        Nothing too interesting in there at the moment but I’ll put my module 2 stuff up there which will probably have some stuff to so with this conversation in there…

        (Thanks for all your comments guys!)

      • bretbenesh Says:

        Hi Vince!

        I said “blog,” but that is mainly because I still think that it is 2010. Posting to Google Plus is perfectly fine; just make sure you let us know about it! Bret

      • Dana Ernst Says:

        Fundamentally, I agree with Bret when he says, “I generally regard grades as a nuisance that is forced on me.” Since we are in the predicament of having to assign grades, I want enough data to justify the grade that a student earns. Moreover, grades act as a carrot for the students. It’s unfortunate that my last statement is true. Perhaps if there was a critical mass of teachers that resisted this paradigm, we could change things. Of course, as TJ mentioned about graduate students, it becomes less of a problem as the funnel narrows. When assigning grades, my goal is for this to provide useful feedback as much as possible (which is why I am so interested in standards-based assessment).

        As for presentations, while I do jot down a grade, I don’t take the grade very seriously and I don’t show it to the students unless they ask (and they usually don’t). The most important aspect of students presenting is being willing to share something, even if they think it is wrong. I emphasize over and over again that what the students present doesn’t have to be perfect. The best presentations are the ones that contain small mistakes. We learn a lot in this situation. I make it clear to the students that for most of them, their grade will predominately be determined by how many times they came to the board. I say “most of them” because if a student only presents 6 easy exercises during the semester while another student attempts (but isn’t necessarily successfully) 3 of the most difficult problems, I want to reward the second student. I used to stress out about how to have a rigorous system for assigning grades to my presentation category, but I don’t anymore. Managing all of this is a bit like jazz.

        One strategy worth mentioning is that I tell my students that they must present a minimum number of times before each of the exams (generally, 2 midterms and a final) in order to obtain a passing grade in the presentation category. This isn’t a perfect system, but I think it works for most of the students. I worry that some students do their minimum (maybe plus a couple) early in each block and then say, “ok, I’m done presenting for a while.” However, this doesn’t seem to be as big an issue as I imagined it could be. It helps that I emphasize that the number is a minimum for a passing grade. The number that I set depends on the number of students and the amount of time we spend presenting.

      • bretbenesh Says:

        Fundamentally, I agree euthanized what Dana says, too. I don’t really grade presentations at all, save for three categories: “majorly wrong, solved a really difficult problem, and Other.” All three categories help students’ grades, but in varying, unspecified amounts. Students only need to do “difficult” problems if tey want an A (I do not tell them in advance which ones are really difficult).

        Merry Christmas, to those who celebrate it! Bret

  7. Assessment Idea for Calculus I: Feedback desperately wanted! | Solvable by Radicals Says:

    […] fall. I used a combination of Peer Instruction and student presentations in Fall 2012, and I was not completely happy with […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: