College lecture classes have been around for more than 900 years. Lately, a handful of science and engineering professors have been experimenting with a more innovative way of teaching science, especially at the introductory level. The idea is to have students spend their class time solving problems and engaging in activities that are designed to help them think like scientists instead of listening passively to an expert.

A new paper in the journal PNAS offers some of the most convincing evidence that so-called “active learning” is superior to lectures. Scott Freeman of the University of Washington in Seattle and his colleagues analyzed data from 225 studies and found that students enrolled in lecture courses were 1.5 times more likely to fail than those enrolled in courses that had elements of active learning. Students enrolled in active learning sections scored about 6 percent higher on examinations than their peers in traditional courses, an increase that could translate to a median course letter grade of a B instead of a B-minus. Those improvements, the authors write, would go a long way toward encouraging more students to persevere in science and math. Of those who enter college planning to major in STEM (science, technology, engineering and math), less than 40 percent graduate with a STEM degree.

Carl Wieman, who earned the Nobel Prize in Physics in 2001 and served as associate director for science at the White House Office of Science and Technology Policy from 2010 until 2012, has been a tireless advocate of active learning. As a faculty member of the University of Colorado and, later, as the director of the University of British Columbia’s Carl Wieman Science Education Initiative, he transformed the teaching style of a majority of faculty in a few science departments, helping move them from traditional lectures to a system that more actively engages students. Last year, Wieman accepted a joint appointment at Stanford, where he serves on the faculty of the physics department and the graduate school of education. He’s written a commentary on Freeman’s paper that is scheduled to appear in PNAS this week. (05/23/14 Update: Read Wieman's commentary here). Below is a condensed and edited version of an interview we had over the weekend.

What is your reaction to the PNAS paper?

It makes a very compelling argument that if you’re going to college and you’re going to a whole bunch of lecture classes that require you to sit there and listen passively, you’re getting a bad education. By teaching that way, universities are just providing an inferior education, and the public ought to know that. The institutions really ought to be taking these results seriously.

How, ideally, should an undergraduate science course be structured?

At the most general level, the classroom is really the best opportunity for students to be interacting with the professor, who’s the expert in the subject, and their fellow students. How do you use that time and that interaction to get the most effective learning?

Headshot of Stanford physicist and Nobel Laureate Carl Wieman

Carl Wieman

What we know about learning from cognitive psychology is that people learn by practicing, with feedback to tell them what they’re doing right and wrong and how to get better. In this case, that means they need to practice thinking like a scientist in the field. They should do background reading that gives basic information before class and then in class they’re working through carefully designed problems that give them practice at a particular sort of scientific thinking, whether it’s how physicists think about forces in motion or biologists think about cells and how they repair themselves, and so on. This way, they get much more targeted feedback from the instructor, who can realize they’re confused about some basic point and can guide them much more directly. In this way, students spend all of their time in class being very actively involved, using their brains strenuously. They would also have homework problems that build on what they’ve done in class so they can practice more extensively.

The basic issue is practicing scientific thinking and getting guiding feedback on their thinking.

Is there any place for lecturing in undergraduate science and math courses?

That depends a lot on what you mean by lecturing. Is there a place for telling students something? Yes, absolutely. Is there a place for the traditional lecture, where the students sit there with no background and no preparation to understand why what they’re being told is important and what they need to glean from it? That’s just ineffective.

In these active learning methods, there’s actually plenty of telling, but it’s after students have been well prepared to learn from the telling. For example, after students have worked through a series of problems, they might all be wondering how you make the next step, and then the instructor, in response to questions, would explain things to them.

How prevalent is active learning among universities?

I’ve polled the leaders of all the major universities across the county, and I’ve found that no institution actually collects any data on what teaching methods are being used in the classroom, which tells you something about the level of importance attached to this. So, because the fact that no data is collected, it’s impossible to say how extensively it’s being used. But it’s pretty clear that it’s quite a small fraction of classes that are doing this.

What has to change in order for active learning to become more widespread?

I’ve been pushing hard for simply getting universities to gather data on the teaching methods that are being used. [As associate director of science at the White House Office of Science and Technology Policy] I pushed a specific step: that in order for universities to get federal research money, they had to collect that data and make it public. That way, prospective students could look and see which universities were using effective teaching methods and which ones weren’t, and then very quickly you’d see a big change.

Why do you think that suggestion met with so much opposition?

[NIH Director] Francis Collins doesn’t want to do it. Researchers would have to pay a little less attention to the research they’re doing for NIH and a little more attention to teachng effectively. It’s not his job to worry about the quality of education that undergraduates get; it’s to worry about how much NIH research gets done. The leaders of the major research universities are also against it.

What it comes down to is that you’ve got a system in place, a measure of quality in place, and anybody who’s at the top of that list, they don’t want to have the rules changed. If you pick your list of top universities in the country, most of their science courses are being taught using what science education research would say are pretty terrible methods.

They are considered top universities because the only thing that’s examined and counts for your success and prominence as a university is the research accomplishments you have, and the research dollars you attract. There’s $40 billion of federal research dollars per year that’s pushing people to focus on that. I’m a researcher, and I appreciate that it’s a very valuable thing, and universities have been wonderful at optimizing research, but the collateral damage is that they pay no attention to teaching, and it’s counted nowhere in any rating of universities.

Should lecturing be abolished in K to 12 science classes also?

The same principles of how people learn apply to all education levels. And I think most K 12 teachers would not expect students to listen quietly and passively while they were being talked at for an hour. But one of the particular challenges of introducing these kinds of effective teaching methods at the K to 12 level is that they really require more subject expertise from the instructor than a lecture. A lecture is basically a talking textbook. But in these methods I’m talking about, you really have to think about how scientists think about and solve problems in a particular area and then design appropriate problems that have students practicing and learning that thinking. Then you have to be able to give the students feedback on how they’re thinking. That is very demanding on your expertise in the subject. At the K 12 level, although there certainly are exceptions, teachers by and large do not have high enough content mastery to do this very well. In large part, that’s because they’ve been through college courses where the science is taught badly, so they didn’t learn it very well. So they’re graduating with a deficient understanding of the subject, and a deficient view about how to teach it. So before you can expect K-12 science teaching to get much better, you have to fix the science teaching in colleges and universities.

What are you working on now at Stanford?

At Colorado and at the University of British Columbia I was working on doing experiments on large-scale change in teaching and figuring out of it is possible to get a whole department to change its approach to teaching. Then I went to the White House and pushed to have the whole country change and beat my head on that problem and felt a lot of frustration.

Now I’m trying to get back to thinking on a smaller scale. I’m doing research on issues like, exactly what is the optimum way of designing these practice problems and what features of a problem are the most effective in terms of having students start more quickly and effectively taking an expert-like problem solving approach? How to help them more quickly develop an intuitive feeling and an effective use of mathematics, for example. We’ve also done some research with online interactive simulations, and we’re working on ways to use that new educational media to accomplish learning in new ways that has some advantages over other media and forms of instruction. .

But, I have to admit, it’s awfully hard to not get sucked back into worrying about the bigger picture.

Have you given up on making changes at the national level?

I’ve given up on getting something done by the government, but I haven’t given up on the idea of trying to persuade universities to report and improve their teaching methods. AAU [the Association of American Universities] is pushing for its members to start improving their STEM teaching. If I can just persuade a few top universities to start collecting data on the teaching methods being used in their courses, and making it public and making it a selling point that they are collecting this data and paying attention to the use of more effective methods, then I think there’s some hope that this will mushroom. So I’m still a proselytizer for trying to have better teaching put in place. I’ll probably never be able to get away from that.

IMAGE CREDITS: Top, Kevin Dooley via Flickr. Carl Wieman, courtesy of Stanford University.

More to explore:

"Online Courses Can Improve Life On Campus," by Robert Lue, Harvard University

"At MIT, Large Lectures Are Going The Way of The Blackboard," by Sara Rimer, The New York Times

Scientific Teaching,” By Jo Handelsman, Sarah Miller and Christine Pfund (Publisher W. H. Freeman, like Scientific American, is part of Macmillan Science and Education)