Expunging Myths from The Classrooom

[anorlunda]

I am not an educator. However, I just read a fascinating article that just screamed "education issue" as I read it. The article was:

The Strangest, Most Spectacular Bridge Collapse (And How We Got It Wrong) WRITTEN BY ALEX PASTERNACK
http://motherboard.vice.com/read/the-myth-of-galloping-gertie

The point of the article is that I and almost all other students of engineering or science were taught that the Tacoma Bridge (known as Galloping Gertie) collapse was due to resonance. However, the peer reviewed science many years ago showed that this explanation is dead wrong. Read the article yourself. It's very clear and the language is simple, even if engineering is not your field.

The article cites examples, where the resonance myth continues to be taught today, and that it has negative real world consequences as new things are designed incorrectly. On one hand, it seems hard to blame the teachers. Probably none of my peers, none of my teachers, none of the senior faculty, and none on the certification or accreditation boards ever heard any explanation (other than the false one) for the Tacoma bridge failure. If they never heard it was false, how could they correct the teaching? So widespread is the myth, that it is generally accepted knowledge. Indeed, the article said,

How did the incorrect explanation persist for so long? In their paper about the event, Bilah and Scanlan cite 30 sources that mention resonance as a cause of the bridge's failure. Ultimately, they point their fingers at a mix of rough, semi-empirical guess work and the "telephone" effect. "The primary reason for all this, we believe, is that many post facto accounts or investigations were speculative or reviews of still other accounts," they write.

The educational issue that shouts out to me is this. How does the educational system systematically discover and eradicate factually incorrect and false lessons from being taught in the class room? Perhaps there is something I'm not aware of.

I don't mean subjects where the truth may be subject to legitimate differences of opinion or of political hot buttons. But I do mean cases like this one where no generalist teacher can hope to read all the peer reviewed literature to discover where common knowledge is wrong. The literature is too voluminous and too compartmentalized for that.

If there is no such feature in the educational system, then the Tacoma Narrows case sounds like a perfect basis for a research project in education theory. It could begin with an audit; how many classrooms at any level from K-12 through doctoral are perpetuating the resonance story? Step 2, how should the system discover and eradicate these teaching errors?

It could be described as a quality control issue for education. If we strive for six-sigma quality improvement in industry, why not in education?

I'm not trying to be insulting to educators. I'm just suggesting that here we have an ideal case up which to do research to improve the educational system.

 

[PietKuip]

There are things in textbooks that are much much worse than the Tacoma bridge. It is quite terrible what even thermodynamics texts can say about what temperature is.

 

[Andy Resnick]

<snip>How does the educational system systematically discover and eradicate factually incorrect and false lessons from being taught in the class room? <snip>

I think it's a slow process that involves not just the teachers, but also the reference textbooks. One related example that comes to mind is from biochemistry- the mechanism by which cells derive energy by hydrolyzing ATP. If you open any 'old' textbook, you are likely to read something about 'high energy bonds', even though that phrase isn't logical and the chemi-osmotic theory had been formulated back in the 1960s.

In these and other cases, while the teacher may have 'improved' or 'more correct' explanations at hand, they still have to contend with what is printed in a textbook- broadly telling students 'the book is wrong' is not the best course of action. Over time, textbooks improve.

 

[MidgetDwarf]

I find the only solutions for students is to read more rigorous textbooks and post there questions on physics forums.

Maybe give extra credit to your students that contribute on physics forums?

 

[anorlunda]

I think it's a slow process that involves not just the teachers, but also the reference textbooks.

I would think that it takes standards from professional societies or accrediting boards.

Let me rephrase the OP in a more provocative form to elicit comments. Can we imagine a lawsuit in which victims of an engineering disaster sue not only the designers, but also the designer's educators, and the designer's textbook publishers?

That sounds far-fetched in real life, but it would make a great plot for a Hollywood lawyer drama such as The Good Wife.

 

[Andy Resnick]

I would think that it takes standards from professional societies or accrediting boards.
<snip>

That's one difference between engineering programs and science programs- engineering programs often have some sort of professional accreditation requirements (ABET, for example), while science programs generally do not (professional programs such as clinical chemistry, medical physics, etc. exempted).

 

[phinds]

The Strangest, Most Spectacular Bridge Collapse (And How We Got It Wrong) WRITTEN BY ALEX PASTERNACK
http://motherboard.vice.com/read/the-myth-of-galloping-gertie
Fascinating. Thanks for posting.

 

[anorlunda]

That's one difference between engineering programs and science programs- engineering programs often have some sort of professional accreditation requirements (ABET, for example), while science programs generally do not (professional programs such as clinical chemistry, medical physics, etc. exempted).

So what mechanism exist to assure that science taught is valid? It makes it sound like advocating quack science is forbidden on PF by PF's rules, but that no such rules exist at science degree granting institutions.

Suppose we limit the discussion to the most prestigious universities. Do they have quality assurance procedures and/or six-sigma programs? Is the quality of their teaching audited by a third party? Is the galloping gertie myth still being taught in their classrooms today?

 

[phinds]

So what mechanism exist to assure that science taught is valid? It makes it sound like advocating quack science is forbidden on PF by PF's rules, but that no such rules exist at science degree granting institutions.

There are none. Religious colleges/universities are free, if they choose to do so, to teach creationism in lieu of actual science.

Suppose we limit the discussion to the most prestigious universities. Do they have quality assurance procedures and/or six-sigma programs? Is the quality of their teaching audited by a third party? Is the galloping gertie myth still being taught in their classrooms today?

I doubt if there are specific quality programs but profs who teach nonsense are usually called out on it, although I'm not sure that always happens and in any event it's almost impossible to fire a tenured prof no matter what he teaches.

 

[anorlunda]

I believe you're right Phinds. That was what I was trying to fish out in this thread. In today's world where we increasingly try to hold everyone accountable for all things, it should only be a matter of times before class action tort lawyers smell this blood in the water.

 

[phinds]

I believe you're right Phinds. That was what I was trying to fish out in this thread. In today's world where we increasingly try to hold everyone accountable for all things, it should only be a matter of times before class action tort lawyers smell this blood in the water.

Not sure they could get far w/ it. Academic freedom is pretty much sacrosanct and probably should be. When people start telling profs what they can teach, we're headed down a slippery slope.

 

[256bits]

So what mechanism exist to assure that science taught is valid? It makes it sound like advocating quack science is forbidden on PF by PF's rules, but that no such rules exist at science degree granting institutions.

Suppose we limit the discussion to the most prestigious universities. Do they have quality assurance procedures and/or six-sigma programs? Is the quality of their teaching audited by a third party? Is the galloping gertie myth still being taught in their classrooms today?

Galloping Gertie would be brought up in a classroom to show that dynamic instability is a reality, and has to be accounted for in a design. The main focus would be that positive feedback can have dire consequences. If a student fresh out of university at least has and keeps that in his head from seeing the pictures, videos and print about the incident, he/ she is that farther ahead, even if actually they don't know all that much in total depth. At least they will be asking themselves " Is my design subject to dynamic forces?"

I would presume that anyone company building an aircraft today, just as an example ( or even more than several decades past ) would have taken the necessary steps in their calculations, manufacturing, modeling , testing to ensure best practices and current knowledge are used, and employ people knowledgeable in the subject matter. If it can be shown there was a possible willfull negligence, then a tort might be forthcoming. Sadly though, it does come about where financial considerations ( as evident in the article about the chosen design ) do lead to a poor choice of design - not always, but it does happen.

 

[Andy Resnick]

So what mechanism exist to assure that science taught is valid? It makes it sound like advocating quack science is forbidden on PF by PF's rules, but that no such rules exist at science degree granting institutions.

As phinds mentions, there aren't any explicit mechanisms in place. However, there are at least two informal mechanisms that work to maintain standards. First, all syllabi must be on file with the university- variances from accepted practices could be identified at that point. Second, the tenure-seeking process involves peer evaluation of teaching, that could also identify instructors who consistently fail to meet professional standards.

Suppose we limit the discussion to the most prestigious universities. Do they have quality assurance procedures and/or six-sigma programs? Is the quality of their teaching audited by a third party? Is the galloping gertie myth still being taught in their classrooms today?

Let's not confuse the two issues: that an instructor may be communicating old or outdated knowledge as opposed to an instructor intentionally communicating false knowledge. And further, those must be distinguished from notions of 'quality of teaching'.

As it happens, academic freedom does not overlap very will with first amendment protections for free speech. "Speech by professors in the classroom at public institutions is generally protected under the First Amendment and under the professional concept of academic freedom if the speech is relevant to the subject matter of the course. [...] At private institutions, of course, the First Amendment does not apply, but professors at many institutions are protected by a tapestry of sources that could include employment contracts, institutional practice, and state court decisions."

http://www.aaup.org/our-work/protecting-academic-freedom/academic-freedom-and-first-amendment-2007

Personally, I resent the notion of third-party auditing of my classroom. Broadly speaking, managerial control and oversight of professional employees is a complex problem that involves not just teachers, but also doctors, lawyers, police, etc. and the process appears to rapidly become highly politicized and divisive rather than produce any sort of meaningful professional standard-bearing.

 

[atyy]

Hmmm, very interesting article in the OP. Thanks! But what about
http://asivr.ump.edu.my/index.php?o...damping-&catid=50:list-of-articles&Itemid=97?
Resonance, Aeroelastic Flutter, Vortex Shedding or Negative Damping?
By Prof Dr Abdul Ghaffar Abdul Rahman

Or

https://books.google.com.sg/books?id=1J-4CQAAQBAJ&dq=flutter+resonance&source=gbs_navlinks_s
Mathematical Models for Suspension Bridges: Nonlinear Structural Instability
By Filippo Gazzola
p32: "Como ... writes that ... 'Flutter occurs when a resonance is established between non stationary aerodynamic forces ...'"

http://www1.mate.polimi.it/~gazzola/aimeta.pdf
Old and new explanations of the Tacoma Narrows Bridge collapse
By Gianni Arioli, Filippo Gazzola
"This gave rise to an internal resonance which started the destructive torsional oscillation."

 

[anorlunda]

Hmmm, very interesting article in the OP. Thanks! But what about

Ah, more detailed investigations are welcome.

The first source defends resonance as an important factor in some structures.

http://asivr.ump.edu.my/index.php?option=com_content&view=article&id=121:resonance-aeroelastic-flutter-vortex-shedding-or-negative-damping-&catid=50:list-of-articles&Itemid=97?]Hence,[/PLAIN] if the Tacoma Bridge is analyzed with wind as an internal force, then the bridge collapse due to negative damping.

In short, certain dynamics can be explained from different perspectives, and all maybe correct. It depends on how you define a system.

I'm not fond of the idea that safety is a function of how you define a system.

http://www1.mate.polimi.it/~gazzola/aimeta.pdf]An[/PLAIN] external resonance, the phenomenon which matches the frequency of an external forcing with a natural frequency of the structure, cannot be the reason of the TNB collapse

That is my point, the resonance explanation is still being taught.

The article linked in the OP, says that the resonant frequency of the bridge was about 1 hertz, whereas the periods of the oscillations was 11-19 seconds. That is the danger of teaching the resonance explanation in so many classrooms. If an engineer is taught to check for resonance, and he calculates the resonance peak at 1 hertz, then he is led to believe that there is no problem. He checked the wrong thing because that is what he was taught.

One could argue that all possible physical risks to structures can and should be taught in classrooms, and all of them checked in every design project. But we know that budgets and schedules are finite, and that double checking a new design against the causes of past disasters takes priority. That is why it is most important to correctly identify the true causes of past disasters.

But we are getting away from the point. The issue is not what the most thorough engineering tells us, it is what is actually being taught in classrooms.
I stand by what I said in the OP, that the Tacoma Bridge example would be a great test case to use for investigations of quality control in classrooms all over the world.

 

[phinds]

One could argue that all possible physical risks to structures can and should be taught in classrooms, and all of them checked in every design project. But we know that budgets and schedules are finite, and that double checking a new design against the causes of past disasters takes priority. That is why it is most important to correctly identify the true causes of past disasters.

Excellent point.

 

[gleem]

Do we know if the lesson of the TNB has not been learned or for that matter is there any other "disaster" being repeated because of lessons not learned? Or proper remediation not taught/

 

[atyy]

Ah, more detailed investigations are welcome.

The first source defends resonance as an important factor in some structures.

I'm not fond of the idea that safety is a function of how you define a system.

That is not what the article says. The safety is the same no matter what language you use.

That is my point, the resonance explanation is still being taught.

The article linked in the OP, says that the resonant frequency of the bridge was about 1 hertz, whereas the periods of the oscillations was 11-19 seconds. That is the danger of teaching the resonance explanation in so many classrooms. If an engineer is taught to check for resonance, and he calculates the resonance peak at 1 hertz, then he is led to believe that there is no problem. He checked the wrong thing because that is what he was taught.

One could argue that all possible physical risks to structures can and should be taught in classrooms, and all of them checked in every design project. But we know that budgets and schedules are finite, and that double checking a new design against the causes of past disasters takes priority. That is why it is most important to correctly identify the true causes of past disasters.

But we are getting away from the point. The issue is not what the most thorough engineering tells us, it is what is actually being taught in classrooms.
I stand by what I said in the OP, that the Tacoma Bridge example would be a great test case to use for investigations of quality control in classrooms all over the world.

The article suggests an internal resonance. Can you point to any textbook which claims it is a linear external resonance? Or do they just qualitatively say it is a resonance phenomenon?

 

[anorlunda]

Or do they just qualitatively say it is a resonance phenomenon?

I thing you are missing the point of this thread. The "they" in question are not researchers, or authors of peer reviewed papers, but rather classroom teachers and textbook authors.

Suppose for the sake of argument, that the article you like really nailed the ultimate truth about the TNB disaster. What is the mechanism by which that truth gets spread to every classroom in the world, and what are the quality control mechanisms to verify it?

 

[Andy Resnick]

<snip> What is the mechanism by which that truth gets spread to every classroom in the world, and what are the quality control mechanisms to verify it?

This is the essential flaw in your reasoning. In any science course (and likely, any engineering course), statements that are universally and permanently true cannot ever be made. Conceptual models of reality are *by definition* low-fidelity models, created to emphasize a particular subset of features over others. As new facts are discovered, models can be modified to reflect the new information. Do you propose to ban any course that uses classical mechanics?

Let me be clear- I am not calling for non-critical use of authoritative textbooks. I'm simply asking you to provide a coherent method to teach elementary concepts that relies on your notion of 'truth'. So let's start with your OP: please explain how you would incorporate videos of Galloping Gertie into a junior-high or high-school science class (student ages 13-18).

 

[anorlunda]

Conceptual models of reality are *by definition* low-fidelity models

Using that argument, teaching 2+2=5, or $F=ma^{1.1}$ could be explained as acceptable low-fidelity models.

By the way, when I was taught about Gertie in engineering school, the professor just said that it was an externally forced resonance. He showed no calculations to support the assertion. The article in the OP suggests that exactly that classroom experience is still common.

Re Your challenge about how to teach Gertie: I would cite all the suspected causes of the failure. That does not mean that I must teach how to calculate all of them, but engineers should be aware of their existence. That would lead to a eventual follow-up discussion important to any kind of engineering; how to prune an overly-long list of potential factors to consider. Real life engineers must do such pruning, but rather than ignore the pruned factors as if they don't exist, they should mention their assumptions and their justifications in their report. In cases such as professional engineers and structural design, they must be taught to refer to codes and to do what the codes require of them.

I fear that too much education is designed for multiple-choice-question exams, and so-called facts. Important topics such as risk list pruning, how to conduct studies, and how and when to consult codes, never find an explicit place in the engineering curriculum.

 

[atyy]

I thing you are missing the point of this thread. The "they" in question are not researchers, or authors of peer reviewed papers, but rather classroom teachers and textbook authors.

Suppose for the sake of argument, that the article you like really nailed the ultimate truth about the TNB disaster. What is the mechanism by which that truth gets spread to every classroom in the world, and what are the quality control mechanisms to verify it?

Maybe you are spreading a myth of a myth?

 

[micromass]

Using that argument, teaching 2+2=5, or $F=ma^{1.1}$ could be explained as acceptable low-fidelity models.

Yeah, they could. The only issue is that in most cases there are models which are as simple but work better. Those models are then prefered. The example in your OP is not such a model.

Here's an example that might be useful: $1024 = 1000$ is an acceptable low-fidelity model. And this one is actually useful! Ever wondered what $2^{1020}$ looks like? Well, since $10^3 \sim 2^{10}$, we have that $2^{1020} = (2^{10})^{102} \sim 10^{3*102} = 10^{306}$
So the number will have approximately $306$ decimal places. Cool huh?
I can imagine similar situations where $2+2=5$ could be seen as useful in making good approximations. On the other hand, I can't imagine $F = ma^{1.1}$ to be very useful, since the standard model $F=ma$ is always more accurate and simpler.

 

[anorlunda]

Some of these replies are getting absurd. A common flaw in public debate is falsely claiming that the opponents argue extreme all or none positions.

One extreme would be that the one and only ultimate truth can be mentioned in classrooms. That is not my position. The other extreme is that anything and everything may be taught in classrooms. I don't believe that those opposing me seriously take that position either.

My motivation for this thread is the simple fact that crackpot science can get you thrown off of PF, but it can not get you thrown out of a classroom if you're the teacher.

My position is to advocate for quality control and for process improvement akin to six-sigma, that have become common in industry, should be applied to education. Dispute me on that, not on false all-or-none exaggerations.

 

[micromass]

Some of these replies are getting absurd. A common flaw in public debate is falsely claiming that the opponents argue extreme all or none positions.

But this is not a public formal debate with opponents. It's educators talking about education. Nobody is misrepresenting your point. It is actually what I think you argued. If you didn't want to argue this, then you should be more clear.

 

[jim hardy]

Interestng article
but...
Isn't it splitting hairs to suggest torsional oscillation is not resonance?
It's harmonic motion.

Everything has a natural frequency , just damping keeps it stable.

What have i missed ?

 

[anorlunda]

Isn't it splitting hairs to suggest torsional oscillation is not resonance?
It's harmonic motion.

Everything has a natural frequency , just damping keeps it stable.

What have i missed ?

You just compare the frequency of oscillations with the resonant peak frequencies. If they are not close, then resonance is not a big contributor. In the article, they said that the resonant peak was at 1 hertz but the oscillations had periods of 11-19 seconds.

It is like the subsynchronous resonances in your steam turbines Jim. If you stay away from those frequencies, and you still get an oscillation, then it is caused by something else.

 

[Dr. Courtney]

I think the bigger issue with quality control is limiting the number of myths and bad approximations, rather than reducing them to zero.

Einstein had a few bad ideas, yet he did OK. Academic freedom necessarily includes the freedom to be wrong.

I also was taught that the Galloping Gertie phenomenon was due to resonance in high school physics. It didn't hold me back from graduating summa cum laude from LSU (BS in Physics) and earning a PhD from MIT. I was also taught the bit about ATP's high energy bonds in high school biology.

The training that served me the best was my high school teachers' insistence that science never proves anything, it only disproves things, which eventually grew in my mind to the notion that all models are approximate and limited in scope and that the importance in science is understanding the scope and accuracy of the models.

As a teacher, my view is that students who reach college without the Algebra and Trig skills needed to succeed in a 1st year Physics course have been harmed much more than students who believe Galloping Gertie was caused by resonance or that ATP has high energy bonds. The notion that they will pass college courses without mastering challenging material is most dangerous of all.

 

[anorlunda]

I think the bigger issue with quality control is limiting the number of myths and bad approximations, rather than reducing them to zero.

Einstein had a few bad ideas, yet he did OK. Academic freedom necessarily includes the freedom to be wrong.

Thanks for that. In my original post (#1 in this thread), I suggested an academic survey study of what was actually being taught world wide, and that Galloping Gertie could make a suitable test case for such a study. The object of the study would be to provide data to support quality improvement for education. Presumably, an improvement program would try to eliminate the worst practices while not impinging the best ones. But no such program can even begin without metrics and data.

I fear that standardized tests have given metrics a bad name, but that does not mean that we should abandon it. ISO 9000 and six-sigma cultures have genuinely improved industry. I believe that similar methods could improve education.

As a teacher, my view is that students who reach college without the Algebra and Trig skills needed to succeed in a 1st year Physics course have been harmed much more than

I certainly can't dispute that. I didn't mean to suggest that misinformation was the biggest problem in education. I see lack of objective data as the biggest obstacle to quality improvement. Data is the necessary prerequisite to enable thinking of how to improve things.

I confess to exaggerating a bit in the title of this thread to entice people to read it. The job of a headline writer it to entice people to read the article. Accurately reflecting the content is secondary to the headline writer. Here is an example from today's news. 'Deranged' nude nurse strolls through St Peter's shouting 'I'm from Brazil'...

 

[Andy Resnick]

Using that argument, teaching 2+2=5, or $F=ma^{1.1}$ could be explained as acceptable low-fidelity models.

Absolutely not- neither of those are conceptual models. The first is a mathematical error, the second a definitional error. In your second example, the actual conceptual model consists of 'force laws', i.e. F = -kx. And indeed, that is a low-fidelity model. Even F = Gmm'/r^2 is a low fidelity model because it assumes m and m' are mass points separated by a single distance r.

Re Your challenge about how to teach Gertie: I would cite all the suspected causes of the failure. That does not mean that I must teach how to calculate all of them, but engineers should be aware of their existence. That would lead to a eventual follow-up discussion important to any kind of engineering; how to prune an overly-long list of potential factors to consider. Real life engineers must do such pruning, but rather than ignore the pruned factors as if they don't exist, they should mention their assumptions and their justifications in their report. In cases such as professional engineers and structural design, they must be taught to refer to codes and to do what the codes require of them.

That's fine for a upper-division engineering course, but you did not answer my question- how would you discuss the observed phenomenon in the context of a primary-school science class? Or would you simply not permit a discussion because those students aren't ready for it? As a point of fact, the observed motion can be well described (kinematics) as periodic. Using low-fidelity models of forced oscillators is then justified in this case.

I fear that too much education is designed for multiple-choice-question exams, and so-called facts. Important topics such as risk list pruning, how to conduct studies, and how and when to consult codes, never find an explicit place in the engineering curriculum.

I tend to agree, and would include the sciences as well. Unfortunately, all professionals need to have a good command of a set of facts in order to perform the higher-level functions you list.