Are many physicists not fluent in the language of physics?

[Gort]

Most physicists would agree, I believe, that mathematics is the language of physics. Mathematical models are used to describe the physical world.
I therefore found it somewhat amusing but disconcerting that a recent paper found a statistically significant negative correlation between mathematical equations in papers and the number of citations those papers receive.
See https://zenodo.org/record/58792/files/eq_physics_2pager_2016-07_27_Analysis.pdf
If that correlation is not a fluke (always a possibility), do many physicists simply not fully understand the mathematics published by others? Do many physicists suffer from the same math anxiety as many in the general population? Is there a more benign reason?

 

[Dale]

Is there a more benign reason?

Sure. Increasing the size of a paper reduces its quality, whether you are measuring size in number of equations or some other metric.

 

[Gort]

Not sure I follow. By analogy, a short story is higher quality than a novel because it has fewer words. Surely the number of equations can't be the metric by which quality is judged - can it?

 

[boneh3ad]

I am not sure why you think that the only reason a very "mathy" paper wouldn't get cited as much is because physicists are intimidated by them. While it's true that equation-dense papers may have less broad appeal, there are myriad other factors. For example:

Perhaps the correlation between writers who use many equations and writers who don't understand the virtue of brevity is strong and positive.

Perhaps papers with fewer equations get more citations because they get more traction in other branches of physics that are not as familiar with the particular mathematical nuances of the paper's subfield.

Generally speaking, the more work that the reader has to do to understand a paper, the less likely he or she is to cite that work. Sometimes an overly mathy paper may lead to that. Sometimes an overly verbose paper may lead to that. The two are also not necessarily independent phenomena.

 

[Dale]

Surely the number of equations can't be the metric by which quality is judged - can it?

Not "the" metric, but certainly "a" metric. I suspect that the number of equations simply correlates with the length of the paper and the length of the paper correlates negatively with the quality.

 

[Gort]

I think the "citations outside the paper's subfield" is an interesting and valid argument. I'm less inclined to accept the brevity argument (if it's a good source, it's still a good source). But maybe that's just me.

 

[Dale]

I'm less inclined to accept the brevity argument

It is not so much an argument as a (testable) hypothesis. I think that it is a much more plausible hypothesis than the idea that physicists are uncomfortable with math.

 

[ZapperZ]

There is already a thread on this :

https://www.physicsforums.com/threads/physics-essays-heavy-in-math-bad-or-hard.893930/

and my comment posted in there still applies.

Zz.

 

[Dale]

Frankly, I find the whole analysis rather silly. Sure, they found a correlation, but I wouldn't read anything more into it.

 

[Gort]

I think the "uncomfortable with math" hypothesis was thrown out there simply to generate alternative hypotheses. They're all testable, though - I'm not a statistician, but I'm sure it would require a much more extensive study to develop any real conclusions.
Disclaimer - I don't have any evidence. But if I were voting on the "better" hypothesis, I'd put "citations outside the paper's subfield" as the main influencer, followed by brevity, followed by a lack of mathematical dexterity. I don't think "math anxiety" would even make my list.

 

[Gort]

Thank you for pointing out the existing thread. Consider it closed.

 

[Dale]

I think the "uncomfortable with math" hypothesis was thrown out there simply to generate alternative hypotheses.

I think that it was thrown out there simply to generate public interest (and thereby additional funding)

I'm sure it would require a much more extensive study to develop any real conclusions.

Precisely.

 

[russ_watters]

Not "the" metric, but certainly "a" metric. I suspect that the number of equations simply correlates with the length of the paper and the length of the paper correlates negatively with the quality.

It may not just be quality -- it is also possible that physicists, like humans, have limited attention spans.

 

[kuruman]

“Words are like leaves; and where they most abound, much fruit of sense beneath is rarely found.”

Is there a correlation between words, leaves and equations?

 

[lychette]

Is there a correlation between words, leaves and equations?

At the end of every paragraph type 20 words of nonsense,...change the text colour to white and print it...does wonders for your word count.

 

[Khashishi]

There are many reasons why there might be a correlation. I don't think math anxiety is the cause.

 

[gleem]

As an experimental physicist I would think that experimental publications would cite more experimental papers to start which would probably have less math. These papers could be linked to a theory or not for example if it is a new experimental technique. If a paper is tied to a particular aspect of a theory there might not be plethora of relevant papers . I any event if a paper is important and relevant it should be cited even if one cannot fully appreciate the math.

 

[f95toli]

Note also that articles published in high-impact journals (Nature, Science, PRL etc) are supposed to be written for a non-specialists (not that it always is). This in itself limits the amount of math you can include in a short paper. This will obviously affect the result since papers published in these journals are generally highly cited.

Note also that the math we use in physics tend to be very specialized; the problem is not understanding what the symbols mean (we can probably all -mostly- do that) but that the formulas rarely mean much unless you have the necessary background to put them into context; you need to understand WHY a certain formula is used and (ideally) also have built up enough familiarity with similar results that you can see what e.g. the form of a specific Hamiltonian is telling you.