Playing fast and loose with the curriculum?

One of the weaknesses of the course I teach is also one of the reasons I like teaching it.

The curriculum revision committee I'm on has been discussing ways to maintain coherence between different sections of this multi-instructor course. Right now we have two courses to compare to. One of these, the one I'm presently teaching, has five or six different instructors, each teaching their interpretation of "Ecology, Genetics and Evolution". I kid you not, that's the full detailed curriculum.

I like teaching it because I can pick and choose freely among possible topics. So, for example, I've taught my students nothing about behaviour and almost nothing about DNA replication, but quite a lot about our local environment and about HIV in Canada and Africa. And in some ways this is good for the students, as I teach them the things I'm enthusiastic about. But, from the perspective of the goals of the Biology Program this is not a good situation, as different students learn very different things and sometimes learn nothing at all about some important topics. All sections do use the same excellent textbook.

The other first year course has the opposite problem. Instructors and students all use a common 200 page set of photocopied course 'notes' instead of a textbook. Coherence between different sections is thus not a problem (except if an instructor runs out of class hours before getting to the final topic). For inexperienced instructors this is a good thing, but there is little opportunity for experienced instructors to control what they teach. So it's hard to work up much enthusiasm.

So one goal of this committee is to come up with a curriculum document that's sufficiently specific to ensure that the important topics are covered, but sufficiently flexible to allow instructors to feel they control what they're teaching.

Are clickers worth the hassle?

I've spent most of the last 24 hrs wrestling with the software that goes with the Personal Response System (PRS) 'clickers' I have my students using in class. This is the third time I've assembled all the marks the students have earned for the answers they've given to the questions I've posed in class and posted the combined marks on WebCT. This time I think I've finally gotten it right. (Thank you, students, for your patience.)

The software isn't that bad, though it's a bit clumsy. Now I better understand how it works, it all seems quite straightforward. But getting myself to this point would have been easier if my university provided better support for instructors using PRS in their classes. The fuss I made about this last year produced some talk of a PRS Users Group, where we could help each other, but that seems to have been vaporware.

But I do think using clickers is worth the trouble. It enables me to push the students into doing 'active learning', rather than passively absorbing whatever I tell them. I really like to see the students talking with each other about what the answer should be.

'cumulative' exams

Students often ask me whether the final exam for my course will be 'cumulative', testing material covered both before and after the midterm. The alternative is that the final only tests material taught after the midterm.

I understand that some professors do give non-cumulative final exams, but I find it hard to think of a situation where this would be appropriate. In my courses, we build ideas onto other ideas. I select a particular order of topics (e.g. genetics then evolution then ecology) for precisely that reason. By first studying genetics, we develop the genetic underpinnings needed to understand evolution, and by understanding evolution we can better appreciate issues relevant to ecology. The exam questions I like best are the ones that ask students to pull together concepts taught in different parts of the course.

Telling students that material covered before the midterm won't be needed for the final is tantamount to telling them to forget that material - it's not even valuable enough to remember for another 6 weeks, much less beyond the final exam.

Researchers as teachers

I've been away from blogging for the past month, as I've been focused on getting two research grants submitted by a March 1 deadline. Yes, that's today, and they're both done.

I've been neglecting not only this blog, but also the students in my classes, as a consequence of caring more about my research than about my teaching. I originally wrote 'as much about my research', but that's not true - I do care more about research than teaching. But I'd like to think that caring about research makes for a good teacher, maybe better than someone who cares only about teaching.

Being taught by scholars (researchers in the science or humanities or whatever) is one of the supposed benefits of studying at a real university rather than at a college where the faculty have little or no time or facilities for scholarly work. The benefit is (should be) that the teachers are people who DO research. We care deeply about intellectual work, about scholarship, and we can communicate about the process from our ongoing experience. Active research also keeps us at the frontiers of knowledge, and gives us the perspective to make value judgments about what's in the textbooks.

The down side of this is that we're unlikely to be as dedicated to teaching as our non-researcher colleagues (at UBC, 'sessional lecturers' and 'instructors'). I, for example, have been skimping on my teaching responsibilities for the past couple of weeks, to get my grant proposals done.

But this isn't because we don't care about learning. I, and every researcher I know, care much more deeply about learning than the great majority of our students do. We LOVE learning - it's our favourite thing in the whole world. But we love learning as a concrete activity, experienced most rewardingly in the research we do, not as an abstraction. So it shouldn't be surprising that we value our own learning activities (our research) more than the learning activities of our students.

Too easy or too hard?

We spent yesterday's class working very slowly through one genetics problem, and now I can't decide whether what we did was way too easy or too hard.

I wanted the students to appreciate how best to approach genetics problems. Normally this would be done in tutorials, but the university administration has decided that this course doesn't need tutorials (1500 first-year biology students! (not all mine)), so I did it in class. Using clickers makes this possible, because students are participating, not just watching.

The problem was a complex one. To solve it students needed to combine information from three different crosses, so we worked through each cross in turn, considering possible hypotheses and testing them against the data. I wanted to emphasize that doing the analysis slowly made the logical steps easier to appreciate, but I fear we went too slowly through the first parts. Almost every student got the PRS questions right, which indicates that we probably should have been spending less time on these issues. And we didn't really get to the last part of the problem, which is both the most difficult and the most rewarding - not getting to the answer leaves the students hanging, and leaves them with the most difficult part.

And, based on comments from students after class, most students will find finishing the problem harder than it should be because they missed the significance of information I gave them about why the problem would interest a scientist. They don't understand how two genes can both affect one phenotype, in this case that the genes each code for an enzyme that produces a pigment (pink and blue respectively) and that the pigments mix to produce the purple wildtype flower colour.

So I'm going to have to spend part of the next class providing this explanation, which means less time to spend on pedigrees and sex chromosomes and aneuploidy.

Why posting the solutions isn't the solution

Students always ask me to post the answers to the questions I give them, but I'm reluctant to do so. This makes them unhappy, as they sincerely believe that seeing the right answers is a good way to learn. But I think that seeing the answers often just gives a false sense of confidence.

Say you first try to do the problem without looking at the answer. Because you know the answer is available, you don't spend a lot of time on it. Instead you try do the problem quickly.

If you're able to come reach an answer, you don't spend a lot of time trying to decide whether the answer you've come up with is right, you just check your answer against the posted one. If it agrees with yours, you pat yourself on the back and go on to the next problem. If it doesn't, you look at how your answer differs from the correct one, say "I see how it's done; I won't make that mistake again", and go on to the next problem.

If you can't reach your own answer, you look at the posted one, say "I see how it's done; I'll be able to do it right next time" and go on to the next problem.

Using my "going to university isn't like going to the tanning salon, it's like going to the gym" analogy helps explain why this doesn't really teach you how to solve the problem. Looking at the correct answer is like watching a trainer show you the right way to do squats. You know you need to practice doing them correctly, so you do lots more squats, matching your moves to those the trainer showed you. If you just say "OK, I see" and go on to do bench presses, your squats won't improve.

But you can't go back to the same genetics problem again and learn to do it right, when you already know the answer. Working back from the answer is very much easier than working forward through the forest of possible answers. You need to build the skills that let you evaluate the candidate answers you come up with, testing each one against all the information you have.

It's comforting to think that seeing how a problem is done gives you the skills to do it. But it doesn't. Instead it gives you a false sense of security that can hold you back.

(next time - why going slow teaches you to go fast.)

colour-coding

A student pointed out on the course discussion board that I'd not used colours consistently in drawing chromosomes. I apologized for the confusion and tried to clarify it in a response.

But then I raised the issue directly in class yesterday, pointing out that on Monday I'd coloured the two chromatids in a pair differently (dark ad light blue), but on Wednesday I'd coloured them the same shade of blue but coloured their homologs pink. And the transparent strips I used to demonstrate meiosis had the three different chromosomes from one parent green, whereas those from the other parent were blue.

And I then told them that I was about to use yet a different colour coding in yesterday's class, with the homologs the same colour (maternal distinguished from paternal by a wavy line drawn on them) and the different chromosomes (those with completely different genes) different colours.

The class had been given strips of coloured paper to use as their won chromosomes in solving our first genetics problems, and I told them to pay attention to the colours they used, with the goal of having the colours a guide to the relationships between the chromosomes they were representing rather than a source of confusion.

The most important thing isn't that they get the colours right, but that they learn to think about what colours will be least confusing, and more generally about how to represent the factors that matter in any given problem. It's this thinking that leads to the most learning.

Acting out

Yesterday in one of the two classes I teach, I had some volunteers come the front of the class and model the process I had just explained. Why is this worth doing, given that it's time-consuming and chaotic?

One reason is that it's just one more way of presenting information. Different ways work better for different people, and when a point is important I try to present it in as many ways as possible.

But there's a better reason for using students to model it. We're social animals. Almost from the day we're born, we find watching people to be more interesting than watching anything else. So, although many students will probably forget how I slid the model chromosomes around on the overhead projector, they'll remember the girls tied together at the front of the room, being tugged back and forth by boys with the yellow ropes and then released by the boy with the scissors. And maybe they'll remember that what happened to the girls is what happens to chromosomes.

What should biology students learn?

Our Biology Program has just been awarded a big 5-year grant (from the Carl Weiman Initiative) to improve how biology is taught. One issue that came up at our first meeting was "What should we be teaching our students?"

Students reading this may be horrified to realize that this is an open question. Surely professors decide what they should teach before they start to teach it! Well, we do try, but deciding what should be taught is a complicated problem and one we have no training for.

We university professors tend to teach a combination of what we learned as students and what we've learned since. This is bad for two reasons.

First, every time we learn something new and important we're tempted to add it to the curriculum, so the amount of information we're trying to teach keeps increasing. Most of us realize this, and keep trying to cut back on the information overload, but we never go as far as we probably should.

Second, the things we learned aren't necessarily the things our students should learn, because we were far from being typical students. Many of us were uber-geeks, and we were all the kind of students who go on to be university professors. But most of our students are nothing like we were. Their futures are likely to be much more diverse than ours, and many will have no direct connection to science at all.

There's another problem. We don't feel competent to teach many of the things we would like to teach, because we have no good ways to assess whether our students have learned them. We want to teach our students how to read critically, how to think creatively, how to write clearly. We want our students to really understand complex principles and processes, not just parrot back textbook explanations. But we don't know how to assess these abilities.

The Weiman Initiative grant will give us resources to develop the assessment tools we need. But that only addresses the second problem. First we need to decide what to teach. And these decisions need to be made in collaboration with our students.

We know what biology you need to learn if you're going to be a biology professor or a high school biology teacher, and some of the biology you'll need if you become a physician, dentist, or other medical professional. But many of you will go on to careers that have nothing to do with biology. So we'd like you to tell us how you might use your biology education when you're raising a family, or working in the family business, or selling real estate, or building furniture.

You can post comments to this blog entry, or if you're in my Biology 121 classes you can post them on the course's WebCT Discussion Board.

Finding the balance with clicker questions

Designing good clicker questions is tricky. I want them to be challenging enough that the students have to think quite a bit, but because correct answers count for marks, I want most students to get them right most of the time.

Usually I let the students discuss each question with each other before they answer. One way to improve both the benefits of the consultations, and the proportion of the answers that are correct, might be to first present the question not for marks, asking students to answer without consulting their neighbours. Then show them the range of answers (?) without indicating which is correct. Then ask them to consult their neighbours before answering again, this time for marks.

I have a couple of little books about using clickers in the classroom (gifts from the textbook rep). One is specifically about science teaching - I'll see what suggestions it has.

Ancestors

What we did:

1. Thought about how to infer properties of ancestors, when we know the properties of the descendants and the phylogenetic tree that connects them. This can be seen as learning to think as scientists, rather than as learning what scientists have found out.

2. Considered properties that might be shared by all cells. Not surprisingly, many students hadn't yet learned to include Bacteria and Archaea in their thinking about life. I wonder if the students who have taken Biology 112 were the ones who had?

3. Considered how the first cell could have evolved. It's difficult to be very specific, given how little scientists know about this, and difficult to be very thought-provoking, given the little exposure many of the students have had to molecular biology.

What we didn't do:

1. Use the clickers. The PRS software froze up when trying to create a PRS 'lesson' within PowerPoint even though it had worked fine on my computer this morning. To make matters worse, it also worked fine after class, when I tried to demonstrate the problem for the technician from Classroom Services. I suspect it was due to some changed setting on the podium PC that had returned to its default when I rebooted the computer. I'll come in on the weekend and check that Monday's questions are going to work.

2. Spend enough class time on student-thinking problems. I think this will be easy once we're into real genetics, but for next week I'll see if I can come up with a few thought-provoking ones.

big pictures

We tackled a number of very big issues in today's class. Probably too many, as none of them got the level of development they deserve.

The clicker questions didn't work. This was my own fault; I had assumed (hoped) that deleting the PRS logo from a slide would delete the associated clicker question, but instead it just created a mismatch between the questions and the slides that sent the PRS software into a tizzy. So we did the science questions by shows of hands, which was fine.

In both classes students raised a point I hadn't anticipated, that phylogenetic trees usually have the deepest branches on the left. I wonder if they learned that in high school. I'm pretty sure this isn't an explicitly-stated convention, but it is commonly done.

Learning about and with clickers

We didn't do anything with clickers in the first class, but I want to have some clicker questions in the next one, and to spend a bit of class time on the mechanics. So this requires two kinds of preparation.

First, I need to have a series of steps that get students started with clickers, because close to half said they hadn't used them before. 1. How to program your student number into your clicker. 2. How questioning works. 3. How answering works. For this I need to have a few very easy sample questions, and to allow time that would otherwise be spent on the science.

And there should be have at least one interesting thought-provoking clicker question about the science we're doing. This can come at the end of class, but I should allow at least a few minutes for it.

The pedagogical challenge is to move one or more concepts from lecture-style presentation, in which I tell the students the concept, to question-plus-thinking presentation, in which I raise the question, students evaluate possible answers, then I tell them the answer. The latter takes a lot more time, but gives much more real learning. So this is another pedagogical problem - because I need to spend less time lecturing on other concepts to create time for thinking about the most important ones.

Why have a 'teaching blog'?

Classes start on Monday, and I've put a link to this blog on the BIOL 121 WebCT homepage (only open to students), so curious students from my classes are likely to start visiting this blog. I doubt that any of their other profs have teaching blogs, so I'd better explain what I'm trying to accomplish here.

This is where I'll be reflecting a bit about my goals for the course. I'll discuss what I'm trying to accomplish in each class and the logic behind the different things I'll ask students to do. I'll probably also consider how to deal with problems (both practical and pedagogical), and now to improve approaches that are not working as well as I'd like.

I'm making it easy for students to read this blog because I'm a big believer in open information. People who study teaching often write about 'meta-learning' (learning about the process of learning) and argue quite convincingly that students who are encouraged to think about how they learn will learn better.

The comments are open (anonymous comments are allowed), and I'll always read them though I probably won't directly respond. Students who want responses should post their questions on the WebCT Discussions Board; I'll create a 'topic' there for posts about this blog, and allow anonymous posting.

And no, the blog material won't be on the exam.

Papers for students to report on

One of the Biology 121 project options is writing a report on a peer-reviewed scientific paper. Students are free to choose any paper that interests them, but I'm providing a list of suitable papers as well.

It's nice if the papers are by local researchers, so I've emailed my colleagues asking for suggestions of papers they've written that might be suitable. The paper shouldn't be overwhelmingly technical, which rules out a lot of molecular biology papers, but this course isn't about molecular biology so that's OK. And they shouldn't be too long. And they should be sufficiently well written that students can understand what the research question was and why it's interesting.

When I discussed their papers with last year's students, many said "I had to read my paper five times before I started to understand it!" But they weren't complaining - rather they were proud that they'd eventually mastered such difficult material, and felt that this new skill would be a big help in their future courses.

Preparing for January

I pulled up last year's detailed course outline to start deciding what to cut and what to rearrange.

Given the shorter term this year, and the timing of the mid-term break, I think I'll move more of the fundamental evolution material into the first two weeks. This will give us a solid grounding in evolutionary principles and processes before we get into the nitty-gritty of genetics.

That will also let me schedule the midterm in the last class before the break, before we get into the technicalities of how natural selection works. And I'll be able to cut a week from last year's post-midterm material on evolution, leaving us enough time to get into sustainability and ecological principles.

But the above assumes that I'll still teach the same content, just crammed into fewer classes. That's not what I want, so I still need to work on cutting factoids to give more time for thinking. And on the classroom activities (clicker questions) that get us all thinking and discussing, rather than just transmitting information.

What to cut?

The powers-that-be have left us with only 12 weeks of classes this term, rather than the usual 13. This is probably due more to the dates that various holidays fall on, rather than to a fiendish scheme by the bean-counters to give the sudents 10% fewer classes for their tuition dollars. But it means that I need to cut 10% of the content from my classes.

I actually want to cut more than that, because I'm hoping to have students spending more of their class time thinking and less time than copying down things I tell them. This means I need to come up with thought-provoking classroom problems and activities, but also means I need to eliminate even more of the 'lecture style' content.

Cutting content is hard. What criteria should I use to decide what students don't really need to know about? Which of the lovely PowerPoint slides I slaved over last year should I consign to the trash? Should I cut the cool new frontiers of science stuff, or some of the classic concepts? Should I just not bother to teach the parts that everyone forgets right after the exam? Do I cut the hardest concepts, or the time I spend reminding students of the basic principles?

To make this even harder, I want to include more about ecological sustainability this year. I fear that this means some genetics will have to go.

Getting ready for the new clickers

Today those of us who will be teaching BIOL 121 next term got together for a demo of the new 'radio frequency' (RF) clickers we'll be using, and of how to integrate the clicker questions into our PowerPoint slides. If you don't already know, clickers are remote response gadgets used to collect students answers in class to questions posed by the instructor (like in "Who wants to be a millionaire").

These new RF clickers are much better than the clunky old infrared ones we used last year. Now the students don't have to worry about whether the system received their answer, so they can relax a bit and think about the question.

Last year the clicker questions we did in class were worth 5% of the total marks for the course. This year I'm going to set it up a bit differently, allowing each student to choose, at the beginning of term, whether they want their clicker responses to count or not. If not, the midterm and final will together count 5% more. I can do this because I've figured out how to set up WebCT and Excel to automatically use the appropriate mark calculations for each student.

I like the idea of giving students more choice in how they will be graded. But I also think that clicker questions help students learn, and that students will take the questions more seriously if they count for marks, even though the mark value of each question is only about 0.05% (there will be about 100 questions over the term).

I like to think that most students will agree with me and choose to have their clicker questions count. Before making their decision they'll sensibly want to know the effect of the clicker marks on grades in last years' classes. So I just checked - on average students did a bit better on the clicker questions than they did on the midterm and final, so the clicker marks pulled their grades up a bit. Note that this doesn't address the question of whether doing clicker questions for marks helped students learn biology. Rather it reflects my decision to avoid giving clicker questions that were very challenging.

Planning...

Yesterday my colleague and I planned our workshop on open-book exams.

Giving a workshop is itself a form of teaching, with the difference that the 'students' are our peers. In any teaching, the students learn best the ideas they come up with themselves, so the participants are going to spend much of the 90 minute workshop doing activities that will raise in their minds the ideas we want to develop.

So they will first write down why they currently give closed-book exams, and what problems might arise if they just let students bring their books and notes to such an exam. Then they'll come up with possible advantages and disadvantages of open-book exams. And finally they'll evaluate the suitability of various exam questions for use in an open-book exam, suggest ways to convert unsuitable questions into suitable ones, and from this generate a set of principles to use in evaluating other questions.

Our roles will be mainly to direct the activities and discussion, and to record the ideas on flip charts. We've developed our own lists of ideas, but we'll use these only if the participants miss something we think is important. We also have a page of sample exam questions that participants can evaluate (I had fun making this last night), but we've encouraged them to bring
questions from their own exams.

Open book exams

A colleague in the Faculty of Education just reminded me that, in a couple of weeks, she and I are going to give a 90 minute workshop for faculty on open book exams. We'll discuss why we think open book exams are a good thing, and easy ways to transform conventional closed book exams into open book ones.

All the exams in my courses (finals, midterms, quizzes) have been open book for the past ten or twelve years. My usual rule has been "You can bring anything except a cell phone or a friend", though lately I'm having to change that to prohibit anything that might allow wireless communication, such as a laptop.

I made the switch to open book exams because I want to be testing students on what they understand, not what they have memorized. This is real learning; it's how the world works. "Life is an open book exam." Of course, in the real world you don't always have the time and resources to look up anything and everything, and you don't on open book exams either.

I also want students to know that what I value is the understanding, not the memorization. Educators always say "Assessment drives learning", meaning that the abilities we reward (by giving marks on assessments) are the abilities students will master. This applies to the way we test as well as what we test. No matter how many times an instructor says "This course is about concepts and understanding, not memorization", a closed book test says "This course is really about memorization".

The colleague and I worked out a very rough outline of this workshop last summer, and found some great resources about non-traditional ways of testing. Lots of people have signed up for it, so now it's time to get our acts together and plan it properly.