How much do scientists lecture, and why? Poll results and commentary

Thanks so much to everyone who completed my little poll on how much science faculty lecture, and why. I conducted the poll because of my admittedly-anecdotal sense that much of the vigorous online debate about how to teach–in particular, whether to lecture–is a bit disconnected from the practical decision-making of many faculty. The online discussions I’ve seen tend to focus narrowly on what pedagogical research says, with the implicit assumption that pedagogical research is or should be the most important determinant of how people teach, and that the main reason people still lecture is because they’re ignorant of pedagogical research. In practice, there are many considerations that go into how to teach, and their relative importance seems likely to be sensitive to individual circumstances. For instance because most faculty have other duties besides teaching.

Here are the full poll results. Below the fold is a summary of the main results, and some comments. The bit I found most interesting is that people who mostly lecture and people who mostly don’t are making their pedagogical decisions for very different reasons…

We had 184 respondents. They’re not a random sample from any well-defined population, obviously. But I think it’s a sufficiently large sample to be worth talking about. It’s certainly a larger and more diverse group of people than “people I happen to know”. So I think the poll is an improvement on–or rather, useful complement to–discussing one’s own anecdotal experiences. And as I’ll discuss below, the results are broadly in line with more scientific surveys, so I’m reasonably confident that the respondents are roughly typical of science faculty.

Respondents vary a lot in how much they lecture, but only a minority lecture almost all the time. 33% of respondents spend 25-75% of class time lecturing. 28% spend 75-90% of class time lecturing. 21% spend >90% of class time lecturing. And 18% spend <25% of class time lecturing. This surprised me a bit, I’d have thought people who spend >90% of class time lecturing would be the most common category. Turns out most science faculty are not just standing in front of their classes talking all the time. Rather, most people spend some non-trivial portion of class time doing something other than lecturing, though that portion varies a lot.

People teach as they do for lots of different reasons. Most of the many options I presented were chosen by at least a few respondents.

People who mostly lecture, and other people, teach as they do for different reasons. This was the most interesting result to me. The most common reasons why people who lecture >75% of the time teach as they do are:

1. want/need to cover a lot of material (chosen as one of their most important reasons for teaching as they do by 35% of respondents who lecture >75% of the time)
2. lack the TA support to do anything but mostly lecture (31%)
3. want to use existing prep (29%)
4. know how to lecture but don’t know how to implement other approaches (25%)
5. feel the class is too big to do anything but mostly lecture (23%)
6. enjoy lecturing (22%)
7. risky to switch away from mostly lecturing; it might work great or might fail badly (20%)
8. class has labs, which are active learning. Don’t see need/benefit for much additional active learning in lecture (19%)
9. habit (17%)
10. lecturing less would be too much work (17%)
11. current way of teaching seems to be working fine (13%)

People who lecture <75% of the time teach as they do for very different reasons than those of people who lecture more often:

1. knowledge of the pedagogical literature (chosen as one of their most important reasons for teaching as they do by 51% of respondents who lecture <75% of the time)
2. want to teach/reward skills and behaviors that lecturing doesn’t teach/reward (40%)
3. want the students to attain a deep understanding of less material (39%)
4. formal pedagogical training (23%)
5. wanted to try something new/different (20%)
6. enjoy not lecturing (18%)
7. class lacks labs; have to get some active learning in there somehow (17%)
8. lectures wouldn’t suit course material (e.g., it’s a field course) (16%)
9. non-lecturing works better for my students than lecturing (15%) (note: by “works better” I meant “improves learning outcomes”, but some respondents didn’t interpret it that way and instead wrote in that they don’t lecture because they get better learning outcomes that way. So you should probably consider this option to have been chosen by rather more than 15% of respondents who lecture <75% of the time.)

Overall, I think the survey results point to four big differences between people who lecture >75% of the time, and people who lecture less often:

1. Breadth-depth trade-off. People mostly lecture when they feel it’s necessary to cover all the material they want or need to cover. People teach in other ways when they want the students to attain deep understanding of less material.
2. Constraints imposed by the class. People mostly lecture to big classes because it’s hard to do anything else, especially without a bunch of TAs. In contrast, one big reason people don’t mostly lecture is because they’re teaching courses such as field courses, for which lecturing isn’t even a live option.
3. Risk aversion. As a group, people who mostly lecture are risk averse: they’re reluctant to make a big investment in new prep for an uncertain payoff, preferring (perhaps without consciously thinking about it too much) to stick with what they know because it seems to be working fine. In contrast, people who don’t mostly lecture include a substantial fraction of people who are less risk averse and happy to try out new things.
4. Training. People with formal pedagogical teach differently than those lacking that training.
5. Enjoyment. People who lecture a lot enjoy it. People who don’t lecture enjoy not lecturing.

Of course, #1, 2, 4, and 5 on my little list are in a sense similarities rather than differences. They’re different sides of the same coins. Many people teach as they’ve been trained to–but different people have different training. Many people teach as they think best, given how much material they want or need to cover–but different people want or need to cover different amounts of material. Many people teach as they think best, given the constraints imposed by the nature of the class and the available TA support–but different people teach different sized classes covering different material and with different levels of TA support. Many people teach in a way they enjoy–but different people enjoy different things.

The most widely criticized reasons for lecturing are rare reasons for lecturing. Anecdotally, many criticisms of lecturing that I’ve seen include attacks on faculty for ignorance or skepticism of the pedagogical literature, for not caring about teaching, for lecturing because they were lectured to as students, for teaching in a way that caters to the strong students over others, or for believing without evidence that the pedagogical literature doesn’t apply to them. Without wanting to deny that faculty like that exist, and that some of them might deserve criticism as teachers, such faculty are relatively rare. Few respondents mostly lecture because they are unaware of the pedagogical literature, skeptical of the pedagogical literature, think the pedagogical literature doesn’t apply to them, don’t consider teaching a high priority, think active learning holds back strong students to cater to weak ones, think choice of pedagogical approach isn’t important compared to all the other things affecting student learning, or because lecturing worked fine for them as students. None of those options was chosen by more than 10% of respondents who mostly lecture. (Aside: when I say these reasons for lecturing seem to be widely criticized, I don’t mean to imply anything about my own opinion of those criticisms. This post is descriptive, not prescriptive.)

The experiences and advice of colleagues and superiors aren’t primary determinants of people’s pedagogical choices. I was slightly surprised by this, because a big reason why I now teach intro biostats as a flipped class was the advice and example of my colleague Kyla Flanagan. But apparently that makes me somewhat unusual. Few people cited the example of colleagues (positive or negative!), advice/mentoring from colleagues (or lack thereof), or approval or disapproval of their chair/head of department, as among the most important reasons for teaching as they do. Just to comment a bit more on that last item: I really sympathize with people who’ve been told by their heads of department to lecture more than they think best. Unfortunately, it probably behooves you to know how those in authority over you think you should be teaching. But at the same time, I’m glad to hear that such incidents apparently aren’t all that common (while wishing they never occurred).

Student happiness doesn’t drive people’s pedagogical choices (plus, what makes students happy varies). Only 9% of people who mostly lecture said they do so because that’s what makes their students happy. And only 10% of people who don’t mostly lecture said they teach as they do because that’s what makes their students happy. I find this reassuring, because an instructor’s job is to help students master the material, not make them happy. Plus, what makes students happy often is hard to predict, and often doesn’t correlate with student learning, or even correlates negatively. (Aside: I recognize that at some places, you need to keep your students sufficiently happy for the sake of tenure and promotion.)

Comparison with previous data

The survey results are broadly in line with those from other, more rigorous surveys in the peer-reviewed literature (Ebert-May et al. 2011, Lund and Stains 2015; ht @ErinSBecker, @meganbarkerase). According to those surveys, a large majority of science faculty are aware of and use at least a few non-lecture teaching techniques, often in combination with lecturing. And the main reasons people teach as they do are actual or perceived external constraints, particularly time constraints, class size, the physical layout of the classroom, and availability of TAs. In part because of those constraints and in part for other reasons, people also tend to stick with whatever method of teaching they adopted early in their careers.

A caveat

One caveat to all this: I’m assuming that people are reasonably good at estimating how much time they spend lecturing vs. not lecturing. But of course, as readers of this blog know, people who don’t formally track their time are crap at estimating how much time they spend on pretty much anything. That’s true of time allocation to lecturing (Ebert-May et al. 2011). I think and hope the poll results are robust to this, since I only asked respondents to place their time allocation to lecturing within very broad categories (<25%, 25-75%, etc.).

The implications

Your mileage may vary, but here are some implications I draw from this poll and reading a bit of related literature:

• Don’t overgeneralize from your own example. It’s great to share your own teaching experiences and advice with others. But if your goal isn’t just to share your own experiences for others to take or leave, but to convince others to teach as you do, you need to recognize that your audience is probably mostly comprised of people whose training, circumstances, and goals are quite different than your own.
• Ethics have nothing to do with it. One occasionally sees claims that “lecturing is unethical“. Fortunately, I think most people don’t buy such extreme claims (including, as she later clarified, the person to whom that first link points). As this poll and previous surveys make clear, people consider all sorts of factors when deciding how to teach, of which the pedagogical literature is just one. Accusing people of being unethical because they’re not teaching as the pedagogical literature suggests they should amounts to saying that it’s unethical to consider any of those other factors. Which is just silly.
• Practical advice matters. The main reasons people mostly lecture have to do with constraints of prep time, TA support, class size, material coverage, and previous training. So what I bet a lot of people would find really helpful is practical advice on how to do something other than lecture in big classes, without TA support, in classrooms with chairs bolted to the floor, without having to invest a ton of time in new prep, and without having to cover much less material. And a pony. 🙂 Except that I don’t think this is an “and a pony” situation, at least not entirely. Clickers, for instance, work with classes of any size and room layout, needn’t take a lot of class time, and many uses of them don’t require lots of extra prep or any TA support. Same for think-pair-share, a technique I for one should really use more often. Giving lots of little multiple choice quizzes is pretty scalable if the quizzes are on scantron sheets, though the time required to write quiz questions adds up. Meg has two posts on clicker use. Terry McGlynn has a couple of handy posts on doing active learning without having to invest scads of prep time. The other nice thing about techniques like clickers and think-pair-share is that you can implement them just a little to start with, so as to avoid the risk of breaking a class that basically seems to work fine. Of course, those scalable, easily implemented, non-risky techniques won’t necessarily be a massive improvement over pure lecturing–but that’s true of any teaching technique. And based on the survey results and the other data I linked to, I suspect that many people are already using those scalable, easily-implemented techniques.

As always, looking forward to your comments.

Related: Brian’s curmudgeonly musings on modern pedagogy