On eliminating lectures, a reality-check: Part 1

Today's The Atlantic contains an article entitled "Should Colleges Really Eliminate the College Lecture?" that has really inspired me to write, in a way that the pending deadline on my book has not. Ordinarily I just ignore pieces like this except for maybe a tweet or two about them. But this time, I feel like this article has so many factually incorrect claims, glosses over so much research, and has such potential to spread bad ideas to a very wide audience that I felt the need to address its points one at a time. This is Part 1 of that response.

Let's begin with the title of the article.

Should Colleges Really Eliminate the College Lecture?

There is a phrase that echoes throughout this piece: eliminating lectures or getting rid of lectures. This is a really curious phrase, and I cannot honestly say that I know what it means. As far as I know, there is only one college in the United States that is actually trying to "eliminate lecture": the branch campus of Benedictine University in Mesa, Arizona (the home campus is in Illinois) which opened in Fall 2013, billing itself as a "lecture-free" experience. I don't know how that experiment is going; perhaps the author of the article should check in on this, if the question in the title is going to be asked. But there is no such thing as "colleges eliminating the college lecture". There are individuals who are moving away from lecture and sometimes departments that are doing so. But entire colleges? I'm open to being wrong about this, but I believe there is no such thing.

And the critical question that the article fails to address is: Insofar as there are widespread efforts to move away from lecture pedagogy, why is this happening? Might there be a good reason for it? The article acknowledges this but doesn't really deal with it; keep reading.

So, if there is nothing like a widespread push for colleges to eliminate lecture, what could the author of this article possibly mean? Again, let's keep reading.

The article opens with a lament that, actually, I agree with completely: New Ph.D.'s do often lack the training in pedagogy that they need to be successful in their work. This training should include all forms of pedagogy, including lecture, and it should expose new instructors to the full range of pedagogies that are out there, as well as the research that informs their effectiveness (the concept of "evidence": hold on to this idea) and the skill of selecting a combination of teaching methods that best suits the learning environment they are tasked with creating. Many universities are wising up to this need for training, but more need to get on board.

However from here, things start to go downhill:

Despite the increased emphasis in recent years on improving professors’ teaching skills, such training often focuses on incorporating technology or flipping the classroom, rather than on how to give a traditional college lecture. It’s also in part why the lecture—a mainstay of any introductory undergraduate course—is endangered.

And here, we find the lede that was buried by the headline: The whole problem with lecture is that we're not well-trained enough in how to give great lectures. Training, insofar as it occurs at all, is focused on all these "modern" pedagogies and on technology. If we devoted as much training time to lecture as we did to the other stuff, then we'd see better results with lecturing. That is the claim as I understand it. It makes sense; but it's wrong, and I'll be explaining why as we go.

The article goes on:

For some years now, students in MIT’s introductory physics classes, for example, have had no lectures, and physics departments at institutions around the country have been following suit.

This is factually incorrect, and the very web page that the article links to contradicts the claim. This page is about the TEAL (Technology-Enhanced Active Learning) program at MIT in which the Physics 8.01 and 8.02 courses were redesigned to combine lecture with interactive engagement --- not to "have no lectures". The TEAL website says specifically that "A typical 8.01 or 8.02 class incorporates lecture, recitation, and hands-on experiments in one presentation. Instructors deliver 20-minute lectures interspersed with discussion questions, visualizations, and pencil-and-paper exercises." (My emphases.) That doesn't sound like "no lectures" to me.

What MIT has done is not "eliminated lectures" but put them into a context where they will work better. And the reason it did so, was because the evidence was telling MIT professors that pure lecture wasn't working. The TEAL website says: "Despite great lecturers, attendance at MIT's freshman physics course dropped to 40% by the end of the term, with a 10% failure rate. Even though MIT freshmen had good math skills, they often had a tough time grasping the concepts of first-year physics. Traditional lectures, although excellent for many purposes, do not convey concepts well because of their passive nature."

But apparently evidence that contradicts one's favorite traditions can be ignored.

Wait, there's more:

But while the movement to eliminate the college lecture first gained traction among physics professors, including the Stanford Nobel laureate Carl Wieman and Harvard’s Eric Mazur (a proponent of “peer instruction” who has compared watching a lecturer to learn physics to watching a marathon on TV to learn how to run), it has expanded beyond the sciences. Getting rid of the college lecture entirely is the mission of a broad group of educators.

Here the article shows its lack of understanding both of Wieman's work and the concept of peer instruction. Let's break it down.

First of all, Carl Wieman: Carl Wieman is a Nobel-prize winning physicist who changed his research area from physics to science education in an effort to improve the way science is taught to undergraduates. In this article from 2007, Wieman provides an extensive critique of the effectiveness of lecture on even tasks as simple as basic retention of the information that was supposedly conveyed during a lecture. This finding has particular relevance to today's Atlantic article:

Zdeslav Hrepic, N. Sanjay Rebello, and Dean Zollman at Kansas State University carried out a much more structured study. They asked 18 students from an introductory physics class to attempt to answer six questions on the physics of sound and then, primed by that experience, to get the answers to those questions by listening to a 14-minute, highly polished commercial videotaped lecture given by someone who is supposed to be the world’s most accomplished physics lecturer. On most of the six questions, no more than one student was able to answer correctly. [My emphasis]

Note that this not only provides a point of critique against lecture in general, it strongly refutes the point made by the Atlantic article that if lecturers just had more training on how to give good lectures, then lectures would be more effective and they wouldn't be so endangered. Even when students get lectures that are made by professionals using professional-grade technology and editing to make the lectures as good as possible, the results for learning are dismal.

The Atlantic article is right about at least one thing: It does appear that Wieman would prefer to get rid of lecture-based physics courses. But what the article misses -- and it misses this same point over and over again -- is that this desire is based on evidence both that lecture is not effective when used as the primary means of instruction in physics and evidence that other approaches work better. It is not merely a witch-hunt perpretrated by a crank who was harmed by lecture as a child, or by someone who holds active learning as a sacred tradition.

The same cannot be said unfortunately for most of the argumentation for lecture that is given in this Atlantic argument: The lack of evidence in favor of lecture, and the veneration of lecture purely on the basis of tradition rather than evidence, is painfully obvious. In fact even the little snippet in the browser tab for the article tips its hand:

But more on that later.

Next, about Eric Mazur and peer instruction: A cursory attempt to understand just the definition of peer instruction would show that lecture is not "eliminated" in peer instruction. Here is that definition, taken straight from the source:

In order to address these misconceptions about learning, we developed a method, Peer Instruction, which involves students in their own learning during lecture and focuses their attention on underlying concepts. Lectures are interspersed with conceptual questions, called ConcepTests, designed to expose common difficulties in understanding the material. The students are given one to two minutes to think about the question and formulate their own answers; they then spend two to three minutes discussing their answers in groups of three to four, attempting to reach consensus on the correct answer. This process forces the students to think through the arguments being developed, and enables them (as well as the instructor) to assess their understanding of the concepts even before they leave the classroom.

So contrary to the article's claims, peer instruction doesn't "eliminate lectures". In fact, lectures are very important to the peer instruction process, as anyone with the tiniest understanding of this teaching method knows. But what lectures are not, is the end-all-be-all of the introductory physics experience at Harvard. They are targeted, brief, focused, and intended to fit into a larger landscape of a very effective learning environment, as almost 25 years of controlled research studies on peer instruction at a variety of institutional settings has shown.

This article so far is not faring very well when held up to the light. But at least we might be getting an idea about what the author means by "eliminating lecture": Apparently, "eliminating" or "getting rid of" the lecture means displacing lecture as the way, truth, and life of university education. There can be no nuanced approach to teaching and learning, apparently, that seeks to place lecture in a context and scope where it will help students the most. There can be only lecture. In fact, the most damning thing about this article so far is not the complete misunderstanding of the examples it attempts to use to prove its point which turn out to refute those same points, or its willful ignorance of research evidence, but rather it's the total absence of the voice, needs, and well-being of college students that is the most disturbing.

Part 2 of this response will focus on the dubious claim of the article that all those research results that show lecture pedagogy being inferior to active learning pedagogies would just go away if we trained lecturers better.