On eliminating lectures, a reality-check: Part 2

In case you missed it, on Friday The Atlantic published an essay titled "Should Colleges Really Eliminate the College Lecture?". Essays like this come around several times a year, just like articles that call for the banning of laptops in college classrooms or something similar, and generally speaking they can be safely ignored, because the only persuasive power they have is to further entrench the beliefs of those who feel that lecture, despite all evidence to the contrary, is the best possible means of teaching in higher education. You ignore these essays and life goes on, until the next round of essays. But this time, with this article, I felt that I needed to respond --- because of how thoroughly wrong it is on basic and easily-checked facts, because I'm tired of my colleagues in higher ed making teaching decisions based on their own interests rather than students', possibly because it's getting near the end of the summer and I'm getting punchy. Whatever the reasons, here was Part 1 of the response in which we found (by actually checking the articles to which the original linked) that many of the claims about "eliminating lecture" in the first 1/4 of the article were flat-out wrong.

Why we should continue to pay attention to the article after this point is a good question, but in the interests of being thorough, here is Part 2. There will be a part 3, perhaps a part 4 because there is just so much here to refute and a pressing need to definitively put some of these specious arguments to bed once and for all. (Although, I am not kidding myself that "once and for all" is realistic.)

Let's pick up the action where we left off, with this sentence:

Educators and administrators alike argue that active learning yields superior results to the lecture. [Carl] Wieman recently issued a fresh plea to educators to stop lecturing.

(Recall that we introduced Carl Wieman in part 1.)

Here we go again. Wieman neither said nor implied any such thing in the NPR piece that the Atlantic article links to, or in the 5-minute audio that is embedded in that piece. (Both of which you should go and check out.) In fact, the NPR article and radio segment confirm the use of lecture in Wieman's classes. Here is what the NPR piece says:

He [Wieman] starts the class with some slides. "We've talked about how to get even one wave packet like that if we just have a single value momentum," Wieman says, offering up a kind of mini-lecture.

So, there's lecture. In fact, although it doesn't say this directly, it's clear from the description of Wieman's teaching that he is using peer instruction in his classes, or something very similar to it, which not only does not "eliminate lecture" but actually is predicated on lecture --- short, targeted lectures intended either to address student misconceptions or to set up clicker questions that drive student discussions.

There is no point, either in this NPR piece or in any article ever written by Wieman to the best of my knowledge, where he "issues a plea" to people to "stop lecturing". What Wieman does suggest is that, just like we saw in part 1 with peer instruction and the TEAL initiative at MIT, lecture is not effective when used as the primary or only means of instruction; and that the research evidence suggests that major improvements in student learning can be had by just putting it in a context where it will support active learning. Then again, just as we saw in Part 1, the Atlantic article has no problem using terms like "peer instruction" without having any idea of what they mean.

This is yet another instance of one of the worst things about this Atlantic article: The stubborn insistence that teaching in any way other than pure lecture is the same thing as "eliminating lecture". If you are using peer instruction, for instance, and you replace 50-minute lectures with three 8-minute lectures followed by clicker questions and class discussion, you are "getting rid of lecture". Reading this over and over again is like listening to a child who, when told that he should only have sweets once or twice a day because studies show sugar is bad for you, flies into a temper about how mom and dad "never let me have sweets". This all-or-nothing mentality about teaching is just as annoying and just as damaging, and frankly it's unbecoming to professionals in higher education where, you would think, there is value placed on nuance and the capacity to find middle ground between opposing ideas.

Moving on, we have this:

For Wieman, who sees himself more as a kind of cognitive coach than the traditional “sage on the stage,” the college lecture is like bloodletting -- an outdated practice that has long been in need of radical reform.

Focus on that word: "outdated". Here we have a subtle semantic attack that is dishonest and dangerous. First of all, Wieman never says that he uses active learning because lecture is "outdated". He is saying he uses active learning because lecture doesn't work as well as modern methods. Go to the NPR piece and start listening to the audio clip around 1:17. Here is a transcription starting at the 1:40 mark:

Eric Westervelt: You're saying that for a thousand years, people have been teaching in a completely ineffective way.
Wieman: No, no, not completely ineffective. What I'm saying is....
Westervelt: Bloodletting was pretty ineffective.
Wieman: No, look at the impact though. You let some blood out and go away and watch them and they get well. OK? So, was it your bloodletting that did it, or something else? And I would say that in this case, it's the same thing. You give people lectures, and they go away and learn the stuff. But it wasn't that they learned it from lecture — they learned it from homework, from assignments. When we measure how little people learn from an actual lecture, it's just really small.

When you bring up evidence, both experimental and anecdotal, about the issues that lecture has, almost inevitably the defenders of lecture will try to reframe the argument in terms of time: "You don't like lecture because it's old. But lecture has been around for 1000 years and so it's a proven commodity. Meanwhile all this active learning is a fad."

But keep this in mind: The discussion about active learning and lecture is not about what's "new" or "traditional", "modern" or "outdated". It is, or at least ought to be, about what works best for student learning. If we have evidence that lecture works well for student learning, then we should use it. If the evidence points away from active learning in terms of how well it works for students, then we should stop using it. Or at least, use the evidence to try to find the best mix of pedagogies. (To be fair: Arguments in favor of active learning that rely on "getting with the times" or some other appeal to modernity alone are just as invalid as those arguing against active learning because it's "new".) We need to be scholars not only in our research but also in our teaching, and constantly let facts drive the discussion. But any attempts to hijack the discussion about teaching and learning and turn them into preferences about new versus old have to be repudiated at every turn, starting with this part of the article.

Moving on: In the next paragraph we get a question that seems to be the central one:

But is it the college lecture itself that’s the problem --- or the lecturer?

As I mentioned in Part 1, this is the lede that was buried by the headline. The next two paragraphs explain why this question is important, but the first of those two might make you question why we lecture:

Concerns about the lecture derive from anecdotal impressions as well as research data, including one meta analysis of 225 studies looking at the effectiveness of traditional lectures versus active learning in undergraduate STEM courses. That analysis indicated that lecturing increased failure rates by 55 percent; active learning—meaning teaching methods that are more interactive than traditional lectures—resulted in better grades and a 36 percent drop in class failure rates. High grades and low failure rates were most pronounced in small classes that relied on active teaching, supporting the theory that more students might receive STEM degrees if active learning took the place of traditional lecturing.

First of all, every person reading this blog post right now should stop, take 20 minutes and read the entire study that is linked in this paragraph from the article. It is a recent article published in none other than the Proceedings of the National Academy of Sciences. I have written about this study before and the ethical issues it raises when we choose not to use active learning. It is perhaps the most important study to date on active learning. A meta-analysis of 225 studies, written by scientists and published in one of the world's leading scientific journals, is no joke.

I'm not sure what your reaction will be when you read that PNAS study. But I will go out on a limb and say that any college or university professor who gives half of a damn about the well-being of his or her students will read that study, and then stop and at least think for a moment about whether his or her teaching in the classroom is part of the problem or part of the solution. If you care about students, then I find it incomprehensible that this PNAS study would not cause you to wonder for a moment if you could be teaching in a better way. If it doesn't do that --- if you just look at the study and dismiss it without any serious consideration of the results, even if you don't intuitively believe the results --- then I think you might want to have a gut-check as to whether higher education is the right career path for you.

Moving on in the article, it says:

Still, although proponents of the movement to move away from the lecture cite data on its ineffectiveness, the debate has failed to take into account the fact that academics are rarely, if ever, formally trained in public speaking.

That word, "still", is painful. Here we have a meta-analysis of 225 existing studies that cuts across a wide spectrum of institutional types, student demographics, and instructional styles and shows a profound impact by active learning techniques on student learning and achievement. There is a suggestion that millions of dollars in students' tutition could be saved each year through the use of active learning because of lowered failure rates and improved retention in the STEM disciplines, particularly among women and minorities. This ought to be convincing to any rational person that, at the very least, there is something to the idea that active learning improves student achievement over and above pure lecture. But the article looks at all this and says: "Still..."

However, this is research and we are scholars, so it's completely fair to ask questions about methodology. Is it true that the studies we have seen have failed to control for the quality of the lecture given? And if so, and a similar study controlled for lecture quality, would we see significant changes in the results?

This is where we'll pick up in Part 3.