Good at Math

Some students who have excellent talent in math (I’ll call them math-heads) have little tolerance for the proofy aspects of mathematical analysis or for math as an object of study in itself. They’re much more interested in math as a tool. And they may be really really good at this. It may be called “applied math,” “statistics,” “computational mathematics,” or “data science.” There are even college majors being developed in last category.

My son’s long-term and career interests fall into the “applied math” or “data science” category. He majored in economics in college. But he got into applied math and statistics by taking up “sabermetrics” – statistical analysis of baseball – as a young kid (9 or 10 years old).

I think more students would take an active interest in applied math or statistics if the high school curriculum were switched up a bit. Teach statistics/applied math before teaching calculus. Apply math to everyday problems, use more applied math in all the science courses, apply it in social science courses. It used to be that only physics had a math emphasis among the high school science courses, nothing in biology or chemistry.

I absolutely think calculus is in its right place as the AP/freshman college course of math. It is in a very big way the culmination of all lower mathematics and the start point for college level mathematics. It’s also the starting point for advanced math, in the form of “real analysis.” Statistics is sort of a side topic in the grand scheme of mathematical knowledge that segways into sampling and experimentation rather than proof. I think the problem with introductory statistics is it doesn’t lead into more advanced study the way calculus does, and without more advanced math it ends up being a dead end.

Though I do agree that statistics and probability have been less than respected as mathematical disciplines in the academic community. As recently as 70 years ago, probability was only considered a serious discipline in the Soviet Union (with, to be fair, some notable exceptions), until the western nations found the applications of advanced probability models to business. It’s as legitimately a field as PDEs or algebra but tends not to have the same amount of focus as those by the mathematical academic community in general.

@NeoDymium Oh, I hate the way statistics is taught in HS and in 100 courses. I get grad students (in Fisheries) who’ve had the required statistics and basically none of it is useful. They haven’t been exposed any stochastic processes–easier stuff than calculus—and that’s what they need for modern data analysis. And they’ve only seen really rudimentary combinatorics.

S did a middle school Math Circle at our local university and their material is far more useful as a foundation.
https://www.math.washington.edu/~mathcircle/circle/2015-16/first_2015.html

He’s due to take stats next year in HS and I’m thinking of making him derive the tests they learn or at least write little programs to see if he can figure out how the tests might arise.

I don’t really know enough about math, but I was certainly under the impression that you could approach calculus and linear algebra through statistics. When my son took a graduate statistics course, it was entirely proof-based and involved a great deal of calculus.

Somewhere in the annals of CC is a loooong thread (actually, probably more than one long thread) about the role of calculus in mathematics and liberal arts curricula. (And by “liberal arts” here, I mean non-engineering arts and sciences, not “humanities”.) I think the title was “What’s So Great About Calculus?” Something I learned from it is that math people have a real reverence for calculus, and that’s in part because it’s worthy of reverence and in part because the way you become a math person in our system is you get the beauty of calculus and are excited by it. Because it constitutes the gateway between school math and actual math, people who don’t appreciate it don’t move past the gateway.

I note that the medical school world has explicitly abandoned the notion that a full-year calculus course is necessary to become a doctor, preferring instead “some familiarity” with calculus together with coursework involving calculation and statistics.

Statistics is becoming a bigger department than Math at a lot of major universities and in many ways is overshadowing math departments. Additionally Probablility and Statistics at the graduate level involves lots of proof. Further the beginning probability class at the graduate level is usually a measure theoretic class.

I was talking to a bio engineer the other day. He indicated to me that they were moving away from using ODEs and PDEs to analyze devices to using statistics and probability. There are people using probability to do number theory and there are also groups studying "topological data analysis " which is a mixture of many things including probability and statistics and topology to study real life data

It’s the other way around. Statistics at anything but a highly rudimentary level is closely tied to probability, which is strongly rooted in proof-based calculus. Linear algebra are or should be prerequisites for advanced statistics, not the other way around.

One of my professors (math prof with a medical background) once tried to show the class why doctors would do well to learn not to be so dismissive of math. He gave an example that is actually very simple to solve, but that you will, intuitively, most likely get horribly wrong:

*A city is performing breast cancer screenings for its inhabitants. It is estimated that 1 out of every 10000 residents has breast cancer. The screening machine has a 2% false positive rate (will screen positive for cancer when the person is healthy) and a 2% false negative rate (will screen negative when the person is infected).

If a person screens positive for breast cancer, what is the probability that they actually have breast cancer?*

Gave this simple problem (which doctors could very well run into in their actual job) to a few of my med school bound friends, and they were all shocked by how horribly wrong their intuition was.

(answer is slightly lower than 0.5% by the way. [Similar problem](An Intuitive (and Short) Explanation of Bayes’ Theorem – BetterExplained))

That’s curious. I find it hard to imagine how those two could be interchangeable in any meaningful way. Though the general applicability of probability to the more traditional fields of mathematical research is no surprise to me since probability theory as I learned it definitely had the potential to have that usage. It really seems like it has caught on much faster in engineering and finance than in math academia, from what I’ve seen. Not really sure why.

@NeoDymium : I am puzzled by your example. It requires maybe third grade arithmetic and decent middle-school reading ability. If I didn’t have to get more precise than “a little lower than 0.5%,” it would take about 15 seconds to solve. (Roughly .98 out of every 201 positive screens will in fact be positive.) It certainly doesn’t explain why doctors should have to know calculus. No one is arguing that adults should be innumerate.

As for what you said in #85 about statistics: that’s what I meant. If you go beyond the most basic level in statistics, you encounter the need for calculus and other highish level math. Your statement in #81 seemed contrary, but I don’t think you meant it that way.

" I find it hard to imagine how those two could be interchangeable in any meaningful way"

They are not really interchangeable. The way it was explained to me is they were using ODEs and PDEs to try and describe how things work and why they work to come up with a good solution. They decided I guess it is much easier to find something that works and then repeatedly test it and then say with a 99 per cent (e.g.) certainty that it works and not really care why.

I think they use that approach in pharmacology. A lot of depression meds work and they can show statistically that they work. They are just not sure of the mechanisms of how they work (For example you could look at SSRI drugs)

Sure, that one example does involve only basic manipulation. I wouldn’t say that it’s a problem for a third grader, as while it does involve only basic arithmetic mechanically, it does involve an insight into mathematical relations that a third (or sixth) grader would not generally have. Nor is it exactly a fundamental tenet medicine that doctors must know lest they be terrible doctors, though it is something that can come up that is contrary to intuition (a very important thing for a doctor) that can very well be relevant to medicine. The problem is, instead, something a little more subtle and a lot more significant.

Calculus is the place where college level mathematics begins. Without it, you will not be able to understand any mathematically-focused topic in college: physics, statistics past the first course, linear algebra, computer science past a very basic level, probability, engineering, and anything that has any of these topics as a prerequisite. When they say “some calculus” they basically mean, “we don’t intend for you to use math but we feel that we should have this requirement anyways.” It’s safe to say that someone who only takes only “some calculus” is expressing their intent to never use math again, even in the case of very simple problems like the previous cancer screening one. And if you don’t finish calculus, then you go no further and your math education stops there. It’s the same reason why engineers go further, to linear algebra and differential equations, when most of them won’t ever even touch that stuff: because those basics open the door to further, useful studies in the field.

And as for what doctors actually need math for? Strictly speaking, you can be a doctor without ever using any advanced mathematics for the rest of your life (many are, to the point that it’s a stereotype). It will, however, mean that you lose the ability to have any advanced understanding of physics or data, which are both pretty useful things to understand for a doctor. Those who practice medicine are in a powerful position to be able to make advancements within the medical field, often being capable of doing far more than just researchers who don’t have a medical background. An understanding of statistics and probability at a decent level, and an understanding of computation and software, goes a long way to being able to make sense of information obtained about patients, diseases, and the like (more advanced conclusions than that screening example, but the same kind of information can be obtained from statistical analysis). An understanding of physics helps understand the physics of the human body, and of medical devices which exist or could be created that would be helpful in medicine. All of that comes from topics that are dependent on calculus.

Can you still be successful, perhaps very successful, as a doctor who doesn’t do calculus and does none of that? Absolutely, and many are. Does it mean you shouldn’t learn calculus? Absolutely not. You could be a doctor without knowing history or without having good writing skills either, but I’m sure no one advocates that.

Let me put it another way: you can do higher math without statistics, but you most certainly cannot do higher math without calculus.

Interesting, even if a bit less fancy than I had hoped. I definitely would prefer an accurate PDE to a blind probability 15 times out of 10, although I know enough about biological compounds to know how tough it can be to get a physical model that actually works properly.

I sure hope the biomedical crowd is more careful with statistics than soft sciences, e.g. psychology, tend to be with them.

Calculus’ place in a Math curriculum interests me, but I don’t claim to know enough about Math to really have an informed opinion.

Here’s my experience in Math though:

I felt that Calculus was more plug and chug, all the way through multivariate, than everything before it. I’m not saying that I thought it was easy. I did well, but I put in the hours. But it was the first time that I felt like they were skipping the why and getting straight to the how of doing the problems. That left me feeling unsatisfied, and honestly, kind of confused, even though I got good grades.

For me, Linear Algebra was the beautiful gateway that someone mentioned above, not Calculus. I took a combined Linear Algebra/Differential Equations course, which I regret - I think pure Linear Algebra would have been better. But the upside was that the teacher I had was great, and the Linear Algebra part was done in the theorem/proof way, not the plug and chug matrix manipulation way. Unfortunately, I was taking a lot of other demanding courses, and I didn’t spend enough time on Linear Algebra.

Years later, at one of my jobs, we had a study group in Linear Algebra, and we went all the way through the theorem/proof way. Very cool, and very satisfying.

I still think about studying the upper division Math courses, even though it wouldn’t help me one bit in my job.

Maybe in retirement :slight_smile:

If you want to understand calculus, then the course sequence you want is “discrete mathematics” followed by “real analysis.” It certainly does give you a much more rigorous way of reasoning about math and of understanding why calculus is the way it is. In the US, unlike in some other countries, it is not just integrated into the standard calculus curriculum. But it will answer the “why” of calculus in a solid 1-1.5 years.

Since most people don’t really care about why and limit themselves to relatively simple results of calculus, calculus as taught is pretty mechanical. Still, mechanical as it is, it does open the path to advanced math education.

@NeoDymium, I realized many years later that Real Analysis was the course that probably would have been more satisfying, but it was not required for my major, and honestly, I probably didn’t have the chops for it.

I did take Discrete Math, but never took any other upper division courses in the Math department.

Like I said, in retirement, or if one of my kids starts to struggle in college :slight_smile:

I sometimes brush up on some of the math I didn’t get a chance to learn as well as I liked in my spare time (ha!). So far it has always managed to be of practical use at some point in the future.

In a sort of follow-up to the slightly preceding discussion, I will acknowledge that the further you go in math, the more intermittent your use of any of that knowledge will be, so I can see why people say that they see why arithmetic and algebra matter but see no necessity for learning calculus. But it is those intermittent uses of that knowledge that make all the difference.

Since statistics is commonly cited as the math of practical value, let me use it as an example. If all you know is the normal distribution, everything looks like a nail. If you know significantly more than that, you will still probably rely very strongly on the normal distribution (because it is, in fact, a very powerful model) but you will know enough about probability to understand when something else is a better model (financial speculation has shown very convincingly how disastrous blindly following a normal distribution can be). That intermittent knowledge is what makes all the difference, even if it tends to come up very sparingly.

http://care.diabetesjournals.org/content/17/2/152.abstract is a medical research paper where the author apparently did not recognize a well known mathematical idea commonly learned in first year of college (or in high school by advanced students) and went about rediscovering it and writing about it.

Good old trapezoid rule.

In a slightly more forgivable turn of events, I have heard that early QM physicists studying symmetry inadvertently rediscovered group theory. Seems to happen from time to time.

Following up on posts about career paths for people ‘good at math’:
http://talk.collegeconfidential.com/discussion/comment/19581445/#Comment_19581445
http://talk.collegeconfidential.com/discussion/comment/19582178/#Comment_19582178

I recently gave a talk in a applied math/stats department and had opportunity to spend a couple days talking with faculty and students about their research on protein folding with applications in vaccine development and treatment for prion diseases. This is a bit old but explains a bit how math is involved https://www.princeton.edu/pr/pwb/99/0927/math.shtml That is just a small slice of how mathematical and computational biology is used in medicine. The fields of molecular and cellular biology have very quantitative aspects and genetics research is obviously very mathematically and computationally intensive. The people (faculty and grad students) all came from a math, stats, physics or engineering undergrad background.

And for those thinking about engineering, this is a good article:
http://www.huffingtonpost.com/marjorie-hansen-shaevitz/everything-you-should-kno_2_b_5548052.html
It references this talk on engineering in the UC system
https://www.youtube.com/watch?v=I8sCjTdSaBk

My son loves math, aced the AP Calculus BC exam, even has a bumper sticker of I “heart” calculus on our truck. We live in Hawaii which is very, very expensive. He will be going away to college. It was a huge struggle for him as he really wanted to attend our state flagship but realized that they did not have strong enough programs to satisfy him. His desire is to return to Hawaii to live. Due to our high cost of living I have stressed the need to find a career that will enable him to live comfortably here. I considered engineering to be a good fit for him as he did well in physics as well.

He did a summer internship in engineering at our state flagship. Although he enjoyed the internship it didn’t provide the “aha” moment for him that engineering was the right field. He is still as conflicted about the major as he was prior to doing the internship. He states that he wants to take the “introductory to engineering” course at college, but I have doubts that taking that course will help him make a decision as I’d think the summer internship was probably more useful in that regards. I also don’t want him to waste taking a class in college that could be substituted with something applicable to another major.

He expresses the desire to teach as well. I believe he’d be an excellent math teacher and support him 100%. But I have also told him that he’d struggle to make a decent living here in Hawaii. I told him he could major in a field other than secondary education and if he still feels the call to teach he could later attend our state flagship to earn his teaching certificate.

He is now looking at the BS in Math/MS in Biostatistics program at the college he will be attending in the Fall. I know relatively nothing about Biostatistics and I’d like to know if this field would be a good option for someone within such a narrow job market.