![[IOP]](http://images.iop.org/cws/icons/themes/phw/colophon-small.png){kind=icon}
By Margaret Harris
I came to physics very late by UK standards: I had already started my freshman year of college. For scheduling reasons, I therefore had to take introductory mechanics with engineers rather than physics majors. Supposedly, this meant I had roughly 300 classmates, but in practice, attendance at any given lecture hovered around 50 students – half of whom sat slumped in the back of the room, muttering “God, I hate physics”.
It seems that my experience was far from unique, and according to an article in yesterday’s New York Times, the physics department at MIT has decided to do something about it. Their new mechanics and E&M courses for undergrads employ something called Technology Enhanced Active Learning (TEAL that does away with the traditional professor-in-front-of-blackboard lecture format in favour of students working on physics concepts in small groups at round tables. Various high-tech gizmos let the students answer questions posed by the professor, who wanders around the room with a few teaching assistants giving presentations and answering questions.
The result? Attendance at these non-lectures has shot up from less than 50% under the old format to over 80%, and the failure rate has dropped from 12% to 4%. The NY Times article quotes a number of experts who think the new system is just great – including atomic physicist Carl Wieman, who’s become deeply involved in changing physics education since winning the Nobel Prize in 2001.
There’s just one fly in this ointment: the students seem to hate it.
Picking up bad vibrations
Although a number of NY Times readers have posted comments wondering, in effect, where this kind of thing was when they were struggling through a boring lecture course 20 years ago, the comments from students themselves are overwhelmingly negative. They point out that a hefty chunk of students’ grades in TEAL come from attendance, so it’s not surprising that both grades and attendance have jumped.
As for the “personal response clickers” that allow students to beam their answers to a central computer, “Most students would rather hurl them into the Charles River,” says one recent graduate. They’re also expensive: although a $2.5 m donation paid for the course’s initial IT outlay, students still have to shell out $90 for their own personal gizmos on top of textbook costs. Many complain that the room layout means they can hardly hear the professor, and one even says that in order to learn the material properly, they’ve turned to downloading old-style lectures off YouTube.
Plus, the course is an awful lot of work: up to 21 hours a week to do all the homework assignments, quizzes, online assessments, lectures and office hours required for this single course (MIT students typically take four courses a term, plus labs).
Is this really the future of physics education? Maybe all those absent engineers in my lecture course nine years ago were on to something…
The sad thing is that because it makes the university look good (ha! we have 80% atttendence and 4% fail rates, bet you can’t beat that!) the enjoyment of the students probably won’t affect whether it continues. I think the whole approach is nonsense. A 12% fail rate in physics is not a bad thing–some people just can’t do it, and by forcing students to attend class and reducing how many flunk out you breed a generation of mostly-mediocre physicists, because all they had to do to pass was come to class. Tuition and books are already expensive enough, forcing students to pay for something they don’t even want is just nonsense! If this is the feature of physics, I’ve really lost all faith in the university education system.
I went through MIT just a little too early to catch TEAL in its full, ahem, glory. My options for freshman physics were basically the ordinary lecture course or the lecture course with more mathematics (I took the more-math option). When I was a senior, though, I got to hear a lot of younger folks griping about TEAL. Now, the uniformly negative impression I received of it may have been an artefact of sampling bias: as a physics student, I hung out with people who were themselves physics students, or at least the dice were loaded that way. Therefore, if a programme were good for the non-majors but frustrating for students who already had some interest in and knowledge of the subject, then we older kids would’ve gotten a skewed view.
On the other hand, I can think of a few reasons why a drop in the failure rate might not be evidence that more physics is being learned. As you point out, if attendance counts, and if TEAL motivates warm bodies to show up at lecture, then the numbers will show a success without an actual effect. Also, classes at MIT do not exist in isolation. The proverb has it that an MIT education is “like drinking from a firehose”: there’s simply too much knowledge blasted at the student in too short a period to handle it all. Consequently, even the “best and the brightest” are typically forced into gamesmanship, learning what can be sacrificed and how to play the system’s unwritten rules. (Benson Snyder, who ran the psychological services of MIT and Wellesley in the late 1960s, wrote a great book about these problems, The Hidden Curriculum. Highly recommended reading.) So, anything which makes it easier to balance the time which freshman physics demands against the time required by introductory chemistry, first-term calculus and so forth will likely look like a beneficial change, even if the students’ brains are not acquiring substantially more knowledge of physics.
Before getting too wrapped up whether students like this approach to learning, it is worth looking at the results of the previous approach. The video “Can we beleive our eyes”, part one of the “Minds of our own” series available for free (registration required) from http://www.learner.org, shows students graduating from MIT being asked to light a bulb with a battery and wire. Most of them cannot. Similarly the video “A private universe” shows students graduating from Harvard being asked what causes the seasons. Something to do with the variation in distance between the earth and the sun is the most common answer (even for physicists).
These videos illustrate a phenomenon which can be assessed more formally using standard tests (concept inventories) to show just how many students pass physics courses without understanding the most basic of physics concepts. Doing this, and the shock of the results, was the stimulus for reforming teaching at MIT and some other institutions.
Phil Barker‘s comment raises an interesting question: what are the results of TEAL graduates on those same concept inventories? Doubtless some of these studies have been done (though with TEAL only up and running for a few years yet, one would expect the results to be somewhat preliminary). Those of us with an interest in science journalism would then like to know, “Why didn’t the NY Times care to mention the numbers which actually matter?”
Those of us who had well-stocked toy bins in childhood would also like to know: “Couldn’t the same result be achieved more simply by purchasing a Capsela set for each student?” Ah, Capsela. . .
For more “meat” on this subject, see Carl Wieman’s “Why Not Try a Scientific Approach to Science Education?” http://www.cwsei.ubc.ca/resources/files/Wieman-Change_Sept-Oct_2007.pdf (mentioned in the New York Times article that Margaret Harris is writing about). While I’m an economist, I had thought that such was common knowledge in the physics profession.
Students complain anytime they encounter a course setup that is substantially different from what they are used to or expect. So do profs who are wedded to the status quo. And if a course arrangement is new, then it is open to criticism simply because it “ain’t what it spozed to be”. Who would actively complain about lectures, when they are established as the Common Teaching Method in colleges?
The problem is that the traditional lectures, in physics at least, are about the worst way to teach physics to all but a small percentage of students. If you study physics education research you’ll find a lot of data showing that regardless of the brilliance of the lecturer, the vast majority of students (even at MIT) don’t get a very good understanding. So they don’t show up, or they sit in the back of the room and gripe. (But of course they usually don’t complain formally).
A lot of research in classrooms has shown that pretty much anything that gets students active and involved is going to have better results than lectures. MIT is to be applauded for taking a big chance and employing new methods that are way more effective than traditional lectures.
Before you complain about something new in physics classrooms, you should learn what the issues are. In this case, one of the major issues is whether the majority of the students gain a useful understanding of the basic ideas. And if they can’t light a bulb with a battery and a wire, they’re missing something pretty darn basic.
On the other hand – students, even good ones, often have a difficult time knowing if they understand something, and typically, like most young people, are also ill-suited to knowing what’s best for them, opting more often for what’s easy or popular. If gaming the system (see, for example, Fried, Robert L. (2005). The Game of School: Why We All Play It, How It Hurts Kids, and What It Will Take to Change It. San Francisco, CA: Jossey-Bass.) counts more than understanding, are we surprised that kids adept at the game (like MIT freshman) resist attempts to change the rules? The main point here is that data from objective measures like Phil mentions shows these interactive classrooms really do work better – that the students can not only answer conceptual questions about the material, but solve problems even better than the traditionally-taught ones that focussed on problem solving but can’t tell you that the acceleration of the ball thrown up at it’s highest point isn’t zero.
As one of the driving forces behind the introduction of interactive physics at MIT, I want to point out that the comment that “the students seem to hate it. ” should be more accurately phrased “the students who were motivated to post a comment at the New York Times web site seem to hate it.”
One has to ask about the correlation between the negativity of the comments and the kind of student who will be motivated to go the NYT site and post a comment. As some measure of that, the student run over-all course evaluations that MIT students do here do not show anything like this level of negativity. Now (but not in the early days), the course evaluations for the interactive method are about at the same level as when we were doing lecture recitation. To show this, I put up a graph of the course evaluations scores and the response rate for a few years before and after the transition from lecture recitation to TEAL in 8.02, electricity and magnetism.
http://web.mit.edu/jbelcher/www/PhysicsWorldBlog/evaluationscores.jpg
If you look at the averages, when we were doing lecture recitation the overall course evaluation level was at about 5.3/7.0 and the interactive version has dropped to 4.5/7.0, with scores for indivividual terms rising well above that. I would characterize that as the students don’t like it as much as before, not as the student hate it.
I would also note that these are end of term evaulation passed out in the last week of classes, and because things done in TEAL classes count toward the grade, the number of responses is considerably higher in TEAL than it was in lecture recitation (see the bottom graph on the above jpg). These increased numbers are from students who were not previously coming to class in the lecture format, and I think one could posit that at least some of the drop from 5.3 to 4.5 is due to the fact that the additional students ranking the course are not a neutral group–if they were not coming before to lecture then they didn’t think much of coming to lectures period. That is, you are sampling different populations before and after the introduction of TEAL, and it is hard to compare them directly. I don’t think that accounts for the entire difference between 4.3 to 4.5 but I think it accounts for some of it.
Let me also say something in general about student opinions of a course. They are really important, but they tend to be the only thing promotion and other committees in academia look at as evidence of teaching performance, at least at MIT. My favorite example of what the significance of these popularlity scores mean is as follows:
When I was lecturing 8.02 in the old format, 700 students, I got one of the highest “I like this course” scores on the student evaluations ever recorded at MIT, not as high as Walter Lewin, but high, and the students who came to my lectures thought they were just great. I give the entire text of the student run comments about the course in 1994 at the bottom of this comment as evidence of that (note I had 175 responses with 700 students in the course–I don’t know what the other 525 students thought, since they did not fill out the evaluation). Clearly they thought I was a good lecturer. In contrast, the first term we did TEAL on term, the over all course evaluation was terrible, the lowest of any course I have been associated with at MIT, but I can plausibly argue that that term the students learned twice as much as under the lecture system, using assessment based on Hake normalized gains. You can see the study that supports this argument at
http://web.mit.edu/jbelcher/www/PhysicsWorldBlog/JLS.pdf
I summarize this by saying that the term I had the worst course evaluations in my career was also the term that I had objective evidence that the students learned twice as much as they historically did in this course. Affective student opinion is very important; it is not the whole ballgame, or even the majority of the ballgame.
Finally, there is a dissonace in the student comments as you paraphrase them:
“….They point out that a hefty chunk of students’ grades in TEAL come from attendance, so it’s not surprising that both grades and attendance have jumped….”
and
“…Plus, the course is an awful lot of work: up to 21 hours a week to do all the homework assignments, quizzes, online assessments, lectures and office hours required for this single course (MIT students typically take four courses a term, plus labs)….”
The last comment is wildly off the mark about the time, and it is not reflected in the student gathered overall student self-estimate of the time they spend on the course, which is about 12 hours.
The implication of the first comment is that the grades have jumped for basically bookkeeping reasons, the implication of the second is that the students are working a lot harder. You can’t have it both ways. I think they do work harder than lecture recitation, but if that is what it takes to increase the learning (and as a by product lower the failure rate), so be it.
John Belcher
Summary of Student Comments from the MIT Student Course Evaluation Guide for Spring 1994
(175 responses out of ~700 students)
Professor John Belcher is highly praised by most of his 8.02 students. “He was one of the best professors I have had here — interesting, relevant, and a good teacher. He is funny too!” Three students claim: “Everything about him is effective.” Over half the class remarks on his “awesome board technique,” noting especially his excellent use of colored chalk to keep diagrams clear, and respondents refer to his in-class experiments as “awesome demos.” Belcher also receives high marks for his ability to explain concepts clearly, for the outlines he uses in lectures, and for his reviews of previous lectures. Most class members praise his attitude toward teaching and toward his students: “He definitely knows how to teach,” and “He cares about his students.” One individual states that Belcher is “phenomenal in his organization.” Another student writes: “Belcher obviously prepares his lectures ahead of time and is the best professor I have had yet. He makes it a point to be structured and organized.” The only negative comment to be raised is the fact that Belcher’s constant throat clearing during lecture is distracting and irritating.
Whenever a new educational initiative is undertaken, students often are the most resistant to change. There is no question that many MIT students are unhappy with TEAL for a variety of reasons but there are also just as many who really like it. However there are several points that are important to clarify.
There are two different subjects taught using the TEAL approach at MIT, 8.01 Newtonian Mechanics (taught only in the Fall semester) and 8.02 Electricity and Magnetism (taught both Fall and Spring semesters). Every term we ask the students to anonymously evaluate the course. The overall subject evaluation has steadily increased for 8.02 Electricity and Magnetism. Using a scale where 1 represents very poor and 7 represents excellent, the subject evaluation for Spring 2007 was 5.1 (364 out of 624 students reporting) and Fall 2007 was 4.8 (100 out of 170 students reporting). For 8.01 Newtonian Mechanics the overall subject evaluation for Fall 2007 was 4.6 (348 out of 580 students reporting). We also ask students to report how many hours they spend on the class including class attendance. For 8.01 Newtonian Mechanics Fall 2007, the reported time was 10.5 hrs/ week (N=348). MIT gives credits for each subject based on the weekly amount of hrs the subject should take and the TEAL classes are 12 unit subjects so students actually spend slightly less time than MIT expects. The data for Fall 2006 was identical: 10.5 hrs/ week N= 327. The data for 8.02 Electricity and Magnetism is similar: 10.1 hrs/week for both Spring 2007 and Fall 2007.
Using a scale in which 1 represents a light workload and 7 represents a heavy workload, students reported that the workload for 8.01 Newtonian Mechanics was 4.5 for Fall 2006, and 4.4 for Fall 2007.
The second point is that the all first term subjects for first-year MIT students is pass/ no record. There is no official grade but there is a hidden grade that can be requested. So the 8.01 Newtonian Mechanics has only been taught on the basis of pass/no record. In order to pass the class, students must pass the exam portion of the course (two midterm tests, five quizzes and a final exam). If you do not pass the exam portion, then no matter what your attendance record is, you will not pass the subject.
All the 8.01 Newtonian Mechanics material and course information can be viewed on the website http://web.mit.edu/8.01t/www/.
Peter Dourmashkin MIT
As a middle school biotechnolgy teacher, one of the most important aspects of my job is to engage and inspire students to consider careers related to math and science.
As an elective teacher in competition with other elective teachers for the same limited pool of students, I totally agree with the notion that courses must appeal and be liked by students.
As Dr. Pavlica, founder of the Authentic Science Research in the High Schools program, was fond of saying, “You catch more flies with a jar of honey than a barrel of vinegar.”
Having said that, I am a strong advocate of interactive engagements (IE). I understand that IE methods are foundational to creating a TEAL-like experience. So let’s not throw out the baby with the bathwater.
In the interest of full disclosure I am a middle school biotechnology teacher who uses as much technology as I can get my hands on in my classroom.
I have no clickers, but I do as many hands-on laboratories and interactive computer simulations as I can squeeze into the curriculum.
In end-of-semester surveys, the majority response of my students is overwhelmingly positive. The type of instruction I design incorporates interactive engagments. It differs dramatically from TEAL in the use of clickers and the level of rigor and expectations. For example, there is no homework.
Despite the lack of “rigor”, in assessments of the rather sophisticated concepts of the particle nature of matter and the central dogma of molecular biology respectively, my sixth and seventh grade do surprisingly well. Sixth grade students are able to create illustrations of particles in motion as they migrate from an open perfume bottle to a nose. Seventh grade students are able to illustrate their thinking on the relationships between DNA, RNA and proteins using simple cartoon models. In one seventh grade class 9/10 students illustrated the central dogma of molecular biology correctly. One of these students even included tRNA, ribosomes and nucleotides. His illustration indicated that he had developed a notion of processivity. I find that very impressive.
To achieve this result, the students saw several videos, worked with hands-on models, and used interactive computer simulations that animated each step of the process from DNA to proteins. I haven’t had the time to set up a controlled study, but I can’t imagine I could have had such success with a straight lectures, textbook readings and homework. In previous years, I have used those traditional methods to teach about the process of DNA to proteins to ninth grade honors students and had much less success.
So it seems to me a no-brainer that the interactive engagement and the use of technolgoy to foster understanding is definitely on the right track.
Looking back to one of the original intentions that motivated the use of clickers reveal that they are designed to allow instructors to make rapid instructional decisions based on student responses. That is a very good thing. The alternative is to blindly lecture on without a clue as to whether or not the students really “get it.”
Let’s raise the level of this discourse.
Will the critics of TEAL, please offer positive alternatives to both TEAL and to straight lecture that have been demonstrated to be more effective?
Will the propoents of TEAL, please offer research on the issues such as the one raised by this article and its commentators?
Thank you. I look forward to your reponses.
I still don’t understand why millions of education R&D funding goes to physicists and engineers who don’t have any training at all in how the mind works and how people learn.
It’s not surprising that you see an ‘every problem is a nail when all you have is a hammer’ kind of solution, and see the confirmation bias of researchers highlighting only the positive data, and not the negative.
hmm…seems the course reflects the assessment.
if exams single point in time are to be the measure, then no point going to lectures if the professor just reads from the text..to get a better pass-find a better text, or a better reader!
if communication is now a priority, many exam passing doctors/physicists/lawyers cannot speak to save themselves..then the course AND assessment must reflect that.
in the end, the question must be asked..what’s the point!? IF the purpose is to produce backroom research genius’ then choose more male asperger candidates…if you want communicators, get a few females into the course BUT value the language and discussion components..otherwise we will continue to get 80% male dis-interested know everything, explain nothing graduates..opps..include engineers in this as well…
don’t mention trains, stamps, cricket/baseball scores, chess or memorisation of some obscure ‘facts’…maybe stereotypical, but it is as good an indicator as the point in time testing of years past
call me cynical…but the ‘new’ methods are no longer new, having been described over 40 years ago…only now when the survival of colleges and schools is under threat from students who want 21st century skills are the piece-meal catchup attempts being made.
many colleges and universities engage their students..and get a genuine enthusiasm for learning..not what is on the exam!
have a look at digital enthnography from KSU(kansas state university) Michael Wesch has identified where the needs are, and has more than a few inklings on how to address the issue..btw michael was US professor of the year 2008, his Youtube videos have about 7million views in 18 months..Machine is us/ing us and Information R/evolution
get with it or get out of the way…
As an Engineering Prof. who had to take physics as a student , and as a past dean and Provost who had to hear complaints by students about the physics courses, I want to add one factor- Most Physics profs look at engineers as glorified mechanics, and feel punished for having to teach them. This sentiment becomes mutual one or two weeks into the course. The same subjects, ( Mechanics, Fluids, Thermodynamics, etc. )when taught by more amenable profs., usually in the engineering departments themselves, suddenly become interesting, and the success rates , and enjoyment increase.
Unfortunately, in many universities one has to find work for the overstaffed physics departments…
I was also an MIT undergraduate and got through the usual freshman calculus and physics courses. To be honest, I only attended classes for the first term of physics (more math option). I read calculus and physics II on my own. All you had to do was pass the tests back then. I learned how to wire up a light bulb in grade school and still do occasionally.
Having to endure an interactive learning experience strikes me as the height of tedium. Who the hell wants to interact? It’s work, and it’s work that distracts from the subject matter. Let’s face it. There are MANY different learning styles. I tend to get a lot from reading and lectures, especially if the lecturer has the sense to let me think. I don’t like interacting with people when I am trying to learn something. Having to answer some silly question when I’m fitting a new piece of knowledge into its matrix makes learning more difficult and shallower.
Forcing everyone into a single learning style is a very bad idea. MIT has two tracks for freshman physics because some people learn more easily with vector calculus, while others work better with physical intuition. When I went there there were at least two other options, along with some experimental versions of the introductory courses. I’m not surprised that many students at MIT, even today, are not all that fond of compulsory lectures and compulsory interaction.
garybau picks out the heart of the matter: What’s the point? What are the objectives of the course that is being taught?
Is it to ensure that more of the students walk away with a better understanding of the concepts, or to have the students show up like clockwork physically present but mentally vacant, to encourage group problem-solving or independent intellectual rumination?
To have highly scored student evaluations on the feedback form because they had an entertaining time listening but not participating, or to have students somewhat miserable because they actually had to put in effort and -work-?
Knowing one’s goal in teaching allows one to choose the teaching style/method that gets more of the desired results.
Students (or people in general) are always going to complain when they encounter something new and different. Especially if it’s hard and pushes them out of their comfort zone. But sometimes, that’s exactly the point of making them uncomfortable.
Whether they should be forced to pay exorbitant fees for the remote clickers or textbooks, or threatened with bad grades for non-attendance is another matter altogether.
“Having to endure an interactive learning experience strikes me as the height of tedium. Who the hell wants to interact?”
My word, so the old lecture style teaching was creating people not willing to interact when solving problems and absorbing information? How far is that going to get a person in the working world?
We don’t have to eliminate lectures completely. Some people may very well learn better from lectures. In fact, I believe MIT is at the forefront of making lectures digital and accessible to all in their electronic archives, which can be listened to at any convenient time.
Realizing that there are indeed other learning styles and coming up with one method to involve those who learn better in interactive groups offers a bit more hope for the future.
Now we just have to wait a little longer for more methods to be developed and for them to mesh so that students learn the way they learn best. In the meantime, I’m all for making students occasionally unhappy if it pushes them out of their comfort zone and into learning something.
Interactive learning is useful. Its used at OSU also and the classes that have it get much better attendance. I’d say it’s because it makes the class more fun, kinda like a quiz show (A great example is Legends of the Hidden Temple). The professor says alot of seemingly boring stuff, but you stay in tune so you can answer the questions correctly and give yourself a pat on the back. Also, this system does usually force students to communicate with one another which makes class more stimulating because in other lectures you sit there and listen to the professor talk the entire time. In all the classes I’ve been in that employs this system, I haven’t met a single person who doesn’t like it. Also, its alot harder to skip class without the professor noticing.
The report linked in John Belcher‘s comment contains the sort of comparisons which would actually be useful in discussing these different teaching techniques, the numbers which I am sad (but not surprised) to see the NY Times avoid. However, in looking at the “Conclusions” section, I find the following:
“Unlike the experimental group students, who responded to both conceptual and analytical problems as part of their weekly assignments, the control group students had to solve only analytical problems in their weekly assignments. The only times the control group students were presented with multiple choice conceptual questions were in the pre- and the posttests. The conceptual pre- and posttests administered to the two experimental groups were mandatory, whereas the control group students volunteered to take the pre- and posttests and were compensated for their time. Students in the experimental group were credited for attendance as well as for active participation in desktop experiments and visualizations. Consequently, experimental students’ attendance was over 80% while that of control students was about 50%.”
This makes the differences. . . less illuminating than they could be.
On a different note, Julian remarks, “My word, so the old lecture style teaching was creating people not willing to interact when solving problems and absorbing information? How far is that going to get a person in the working world?”
To which one might reply that freshman physics is not the only place one receives training for the working world.