I think a mathematician should somewhat know both, with the ability to be versatile enough to step into almost any situation and not feel that they weren't exposed to something. For jobs that require heavy statistical analysis, a stats major would probably be the best option...pure mathematicians aren't even necessarily exposed to stats classes unless they seek them out, other than Probability. We get so deep into calculus, topology, abstract algebra, etc, that the real-world skills that we should be exposed to are overlooked. And let's be real for a second, an engineer may do some calculus sometimes, but for the most part, everything's automated nowadays. So unless it's to check your work, all you have to understand is how to use a program, and be able to interpret what the results tell you.
It definitely seems to be the case that pure mathematics students are not exposed to anything other than pure math. I'll talk more about that below, but at my school, since we don't differentiate between "pure" and "applied" (all we have are BA, BS, and Actuarial Sciences), I'd argue that the BA is more "pure" and the BS is more "applied" since the BS student has to take more science and engineering credits that the BA does not.
And I'm curious to know what that means for students getting either BA's or BS's in math. I was reading another forum where someone noted that at their school, "pure math" or "mathematical sciences" was a liberal arts degree, but "applied math" was actually an engineering degree. I've noticed that job listings looking for math majors or listings considering mathematics to be a sufficient degree, they are usually looking for a BS, but do they look at BA students as well provided they have some type of coursework in science/CS/engineering or finance/business/accounting? I would love to know the distinctions.
Can I ask, was your undergrad degree a BA or a BS?
I don't agree there. I wouldn't have done math if I had to take a lot of applied classes. ODE was the first (and definitely not the last) time I questioned whether this math shyt was for me. I liked pure math, enjoyed learning the concepts, but in retrospect I feel that much more could've been done in order to prepare us for the kind of work people with our degrees pursue. Also, after going back to school and experiencing the contrast with systems engineering/OR, I was kind of miffed about it.
This puts our previous discussion in somewhat of a different light. I have to ask (if you're still reading this, I know I'm several days late with this response), what were your expectations going in to pure math and where you aware at the that it was more tailored toward academia, and that applied math and other subfields were (at least somewhat) more tailored toward industry? It seems like from your previous point that your point of contention is that you were not aware of this, is that fair to say? And what do you think should have been done to make you more aware of these points as a young mathematics student?
And I feel like that speaks to a larger problem that people get into when deciding to become a math major. I was reading an article a few days ago about how math majors don't have a "survey" course about what college mathematics is and what they can do with it (and if they are required to take one, it's usually under the generic science course that's probably more tailored toward biology students since they make up the majority of STEM majors). I'll talk more about that below.
I'm not even sure how that gap can be bridged. Maybe more theory should've been incorporated into prior calculus classes...but then again, that would've been detrimental to the majority of people taking the classes that never go on to take real analysis.
It's "detrimental" in the sense that those students may not need it directly, and it might come at the expense of teaching more practical material (as far as time limitations are concerned), but considering how many students in other majors complain about upper division mathematics courses being "too theoretical," I can't see how an incorporation of theory would be a problem beyond that.
Even the math courses I've taken that were specifically for engineering majors (PDEs, ODEs, and computational linear algebra) still had a good dose of theory in them, especially linear algebra, and that's why a lot of the CS majors in that course failed. Though that may be a different case entirely since many of them had either already taken discrete or were taking it concurrently, so they should have already been exposed to it.
But there's still a problem of needing prereqs in order to get to it. You need Discrete, Set Theory, Calc I, II, III, Linear Algebra, DE, and I'm probably missing more. That takes you to at least your junior year, if you load up. There's no way to learn that material without the majority of those courses as base knowledge.
At my school, discrete and set theory are the same class (though there is a more advanced set theory course that only CS majors are required to take). So all we need is discrete (since it's the introductory proofs course) and all three semesters of calculus (since real analysis is essentially learning calculus the hard way). Linear algebra is also a prereq, but I don't think we've ever done anything that required it. Diff Eq. and any other upper division math course can be taken concurrently, but there are issues with that that I'll discuss below.
I feel the ideal mathematics courseload (assuming you didn't switch majors/have to take any remedial courses like precalc/your school is set up on the semester system with similar prerequisites as mine) would be something like this:
Freshman Year:
Semester 1:
Calculus I
A MATH 100 "survey" course where students learn basics about the history of mathematics, the different types of subfields of math (pure, applied, computational, stats, OR, actuary), what kinds of jobs are available for people in those subfields and what degrees they'd need for them, and how to make themselves the most marketable to get those types of jobs (what to minor/dual major in, which electives to take, etc.). And also give them a basic intro to proofs, but nothing more rigorous than anything learned in high school.
A lot of people in other majors (like ME or EE) complain about the 100-level survey course because they feel it covers stuff they already know, but I feel it's important for math majors since a lot of them have no idea what they're getting into or what they plan to do with that degree. So it's important to tell them that, no, they don't have to be teachers, but they also need transferable skills if they want to find a job in industry.
Also, a brief intro to the history of mathematics might be interesting as well. I was listening to a podcast yesterday where they were discussing how a lot of times people don't get into math because they discoveries made in it are not put into any historical context (I know I had a newfound appreciation and respect for proofs just by learning about how long it took to prove Fermat's Last Theorem), so this class would be a good introduction without having to take a full course in it (though one would still be offered).
Semester 2:
Calculus II
Sophomore Year:
Semester 1:
Calculus III
Discrete Mathematics
Semester 2:
Linear Algebra
A "pre-analysis" course where students are introduced to more rigorous proofs than what they learned in discrete, but less rigor than what they will learn in real.
Junior Year:
Semester 1:
Real Analysis
Ordinary Differential Equations
Abstract Algebra (though this can be put off until the following semester/year if the student feels that courseload is too heavy)
Semester 2 through both semesters of Senior Year:
Any upper division mathematics courses the student needs
Any any electives/general eds/prereqs needed within these requirements.
My issue isn't with the difficulty of the real analysis (now that I'm done with it
),
it's that it really doesn't serve a functional purpose for an undergraduate. You spend all that time deciphering proofs, then unless you make a concerted effort, you forget it within a year.
The bold part isn't true, and I think that's a misconception that people make because 1) no one told them what it was used for, and 2) it's not a prerequisite for anything in most schools (other than the second semester of the sequence), so people think of it as just another upper division math course that they're required to take for some reason. But just because the material in real analysis isn't necessarily directly applicable in real life (though it does have uses in finance and physics), it doesn't mean it has no functional purpose. PDEs, complex analysis, and numerical methods, all of which are essential for an "applied" mathematics student or engineer, use a good amount of real analysis, which is why, as I said before, many schools want you to take a whole year in it. Don't get me wrong, I and many of my classmates think/thought that it had no purpose as well, but I've realized that I probably would have done a lot better in PDE had I actually done better in real analysis the first time.
I understand you have in mind what you want to do, and that's great, but would you be better served developing one of the other skills for what you want to do, or doing applied math? Your resume under education to employers is gonna look like: MS - "Blah, Blah, Blah" Math, BS - Math. The vast majority aren't going to understand the difference between what you did in undergrad and grad. So you're kind of painting yourself into the role of being a mathematician wherever you go. As I mentioned earlier, taking a few CS courses, taking a few physics courses, doesn't make you competitive with someone that has degrees in math & CS, or math & physics. And let me be clear, robust degrees like Econ, Finance, CS are understood to have many more applications, so it isn't the same as a CS undergrad staying on the same path for grad school.
The bold part is essentially what a "good" computational mathematics program is. It allows you to take the core of both applied math and computer science while trimming the fat of both (if very few jobs care about a math major's knowledge of real analysis or number theory, imagine how few care about a CS major's knowledge of automata theory or compiler design). The reason I wanted to go into it is because there are a lot of jobs that require more math than what the average "specialized" major would take, but still need a solid knowledge of computing. In fact, I looked at some of the course requirements for operations research at various schools, and they're not much different.
I definitely hear you on the part about being branded as a mathematician though. I'm curious to know, however, how much of that is simply the HR department only looking "buzzwords," though. That's my biggest concern getting that degree. I agree that very few jobs are going to have any idea what that degree is or what it entails, and apparently, if you're going into a tech-based field and your degree doesn't have "engineering" and/or "computer" in the title (like physics, math, etc.), you're basically looked at as a "second-rate engineer" and you'll have a harder time getting an interview.
Though the degree is called something different depending on what schools you go it. Generally it's called "Applied and Computational Mathematics," but at other schools it may be called "Computational Sciences" or even "Computational Engineering." Do you think that getting a degree from a school that calls it one of the latter two would minimize the "risk" of being branded as a mathematician?
And with that I have to ask this: you mention the increased job opportunities that your master's in operations research has afforded you, but you didn't know how much of that is simply because of having a master's degree at all. Do most employers know that, by most academic standards, that that's considered a "math" degree? Have you had trouble with people branding you as a mathematician because of that? Or has it not been a problem because the degree doesn't specifically have the words "mathematics" in it?
And if you get the opportunity, definitely find out what those companies want. More importantly, find out what those companies have in the people currently employed. Also take into account the outlook for the job market you want to enter, and the outlook for yourself. What's your backup plan to do if you don't land the job of your dreams? What's the career trajectory you're aspiring to achieve? Does the path you want to take maximize your potential while minimizing risks? If you're looking at getting another math degree for one particular purpose, it could very well be the case that simply having an undergrad math degree could be sufficient to get your foot in the door there.
This is what I was thinking, but I feel like my math degree is a bit too general for one specific thing (which could be good or bad, depending), which is why I felt a more specialized graduate degree would give me a more competitive edge. I talked about how numerical methods are so essential, and now I'm regretting "wasting" a semester in number theory when I could have been taking that. I hear you on all the other stuff too, though. I guess I'm just uneasy about leaving my future in any type of uncertainty.