I am a senior maths major (computer science minor) who is pretty worried about the next step in my academic career. First, let me state that I'm about as sure as I can be that I want to get a PhD in mathematics. Unfortunately, I didn't realize what the field entailed, or my passion for it, until I had made many really poor decisions - mostly in the form of bad attendance. For example, I basically just showed up for tests in calc 2, 3, linear algebra, and differential equations, and consequently, my field of potential letters of recommendation is quite small. To make things worse, I come from a party school - I need letters!

I've had one professor (abstract algebra) offer to write me a letter, and I've taken my advanced calc sequence under a professor who I think could write me a good recommendation (adv calc 2 was a graduate course; had [i think] the highest grade out of about 15 students). I'm also taking topology (graduate level) this semester, and am hoping to impress my way to a third letter.

My GPA is okay - cumulative about 3.61; math is all A's and one C in linear algebra. I've also been working through a few books (Spivak's "Calculus" and "Calculus on Manifolds," and am about to start Birkhoff and Maclane's "Survey of Modern Algebra." Although I love the material, and enjoy learning it, the independent studying probably stems from some feeling of inadequacy due to my past immaturity.

I got a 169/170 Q, 165/170 V on the general GRE. Also, I think I can crack 80% on the subject test, but am not overly confident about this. One glaring hole is that I have done zero research, and have done nothing extracurricular - I literally have nothing "extra" going for me.

My concern is that I've seen the resumes of many people accepted to top universities (PhD track), and I just don't stack up. But if my goal is to become a professor one day, it seems that where I go to school is extremely important. So should I just hope that I can get accepted into a top 30-50 school, or would it be beneficial to consider improving my resume in a solid Masters program so that better schools become available?

And if a masters is a viable option, what caliber of school would I need to excel at in order to be a competitive applicant for a top 10 PhD program?

  • 11
    How can you be sure you want a PhD if you've never done research? I don't know anything about math programs so I can't comment on your specific question, but a masters degree seems like a smart way of trying out research in a relatively low-commitment way. It sounds to me like you've gotten to the end of college, don't know what to do, and grad school just seems like the "next thing". That's usually a bad reason to go for the doctorate.
    – jurassic
    Jul 17, 2012 at 3:48
  • 4
    If you really love learning math and want to make sure you are up to par in your general background, it might be worth applying to sit for part III of the Math Tripos. However, I agree with @m0untain that if you haven't done research there is no real way to know if research is for you. Jul 17, 2012 at 9:26
  • 10
    Research. Research. Research. Research. Research.
    – JeffE
    Jul 17, 2012 at 17:37

3 Answers 3


If you have a thin B.S. background in math, it is definitely very helpful to get a M.S. in math before (re-) applying to Ph.D. programs. (I say this having been on the Grad Admissions cte, and having been Dir Grad Studies in two different incarnations...) In the U.S., many undergrad programs are really very thin, due to the requirement (otherwise wholesome) of "breadth".

The "undergrad research" episodes in summers, and during the academic year, are good for generating enthusiasm and camaraderie, and especially for getting outside the rigid classroom/textbook atmosphere, but (apart from very broad features) are not at all good indicators of what serious research is or will be like. Those programs are designed to be fun (pizza parties, etc), so "doing research" in that sense is fun for nearly everyone.

So, apply to MS/PhD programs at the top 30-50 schools, and do the best job you can in the standard/required PhD curriculum (usually, there is no distinction between PhD curriculum and MS, except that the latter is designed to accommodate, if necessary, weaker students, perhaps weak enough so that the MS will be their last degree in Mathematics... don't be misled into taking an "easy route"). And, in the course of doing the coursework, don't be a stranger to the instructors of those courses, who will be your letter-writers for either re-application to "better" schools in a year or two, or will be your letter writers if you need to re-apply to that institution itself for the PhD program, for bureaucratic reasons.

Grades in non-math courses, and grades in calculus and lower-division courses don't matter much, although obviously good grades are a not a bad thing. Admissions committees are well-acquainted with the changes people go through around age 20 and so on. The question is not so much what silly things one has done a few years back, but where one is heading now, and what documentable evidence there is for this.

In particular, although self-learning is the most significant long-term way to develop scholarship, it is obviously hard to document. Perhaps the best way is to sign up for courses that appear to re-iterate (serious) content you've already studied. Presumably, you absolutely ace the material and draw the instructor's attention... since self-study beyond "requirements" is, strangely-or-not, extremely rare. Evidence of non-passivity is excellent, if it really proves to be what you feel it is (rather than, say, mere obsessiveness).

So, again, yes, think about "proving yourself" during a year or two of "MS work". No downside, really.

  • Hey Paul, quick question: Will doing a MS reduce the time you're allowed to complete a PhD? That is, if you enter a PhD program with a MS will you be provided less funding/time to finish your PhD? Feb 6, 2014 at 18:39
  • 3
    @MatthewTran, probably not, but it depends on the program. Years ago, our program tried to push people who already had Masters through faster, on some general principle, but since Masters' degrees vary so widely, as do undergrad degrees, it eventually seemed foolish to us to have such a rule, and we dropped it. But you'd want to check, if possible. Feb 6, 2014 at 18:46

I was in a similar situation 1.5 years ago: made a series of bad decisions, have 0 publications or research experience, nothing outstanding about my academic CV. I was in fact wavering between pure and applied mathematics, not really knowing much about research in either area.

After doing a 1 year mathematics MSc at a top-tier university, would I say that in situations similar to yours and mine, doing an MSc is a very wise choice. Doing well in a good MSc will certainly overshadow what you did in your undergrad.

The primary goals of the MSc are:

  1. Obtain good grades and recommendations letters.
  2. Do Research! It is absolutely vital that the MSc has a significant research component.

Things I wished I knew back then:

  1. Take courses to maximize your grades. This could mean taking courses you have already taken before. Its not strictly a waste of time: you can see this as a test to see if you can perform in the topic at a graduate level. Also, if you're interested to do research in it, this would seriously help reinforce your knowledge in that area and you can take the opportunity to know the professor teaching it, even doing your dissertation in it.

  2. DO RESEARCH. Grab any chance you have to do research. In fact, be prepared to stay behind for a few months after graduation to turn your dissertation into publishable material or for an internship in the department.

  3. Pin down your interests ASAP. Do your MSc dissertation in that area if you can. It is such an advantage to have a dissertation project in and a letter of recommendation from someone in the area you're applying to. Contact relevant professors about graduate applications asap.

Advice for applications:

IMHO you should aim as high as possible when applying for MSc. Top tier self funded MScs are a lot easier to get into if you don't have terrible grades.

Its not about the prestige of the department. Rather, more competitive places tend to attract highly motivated and competitive people. Being in that environment would seriously inspire you to push harder and accomplish more. Also, they tend to have more "intellectual resources" - brilliant professors, brilliant classmates etc for you to learn from.


I am a student still in the process of applying for a PhD in Applied Mathematics.


I've always thought that one good indicator of whether or not you even have the motivation to complete an entire PHD is whether or not you do the practice problems in text books. If you enthusiastically do those practice problems, like you solve them in the shower, then I would say that, barring talent, you at least have the requisite level of enthusiasm for the subject. In other words if your not a fan of practice problems, your prolly not gunna like the 300 page writing part of the PHD, nor the fact that not all 100% of the work you do will make it into that writeup (there's alot of tangential calculation and verification). In this way, personal interest and commitment to mathematical activity is absolutely requisite.

You should wait until you finish that course in topology. Math takes on a different character when you get into analysis, manifolds, algebra and beyond. For me, smooth manifolds was as far as I needed to go in the analysis route to satisfy my curiosity. Then I became interested in other things. If I had had that shift of interest midPHD then I don't think I would have been able to finish.

You should also just sit down, learn LaTex if you haven't already, and write about something that interests you, exploring it to the absolute highest level of detail while always leaving an obvious path for generalization and application. Make it lucid and interesting. Convince the reader you have an idea and entertain them with it. Put it on the Internets, have a proff edit 1 or 2 pages, show it to a friend or classmate, stick it in a library book, whatever. This is one defining characteristic of a mathematician, communicating your thoughts to paper so that they may survive.

You should read this The Best Writing on Mathematics 2011, it will give you a good idea of what doing math as a profession is like. Regardless, you may have an excellent academic record, but what makes a good mathematician is a commitment to doing math and that should be your primary focus, grades second. Although there's nothing wrong with being competitive academically, if that's your thing then go for it.

Personally I didn't like Spivak's manifold calculus text. I went Munkres' Topology, to Lee's Topological Manifolds, and have yet to finish Lee's Smooth Manifolds. If your looking for a reliable publisher, just stick with the yellow covers.

  • 2
    One of my classmate during my mathematics masters at a top tier university didn't have any motivation to do practice problems. We would discuss tests after we have taken them, and he would demonstrate the inability to solve even straight forward questions. However, his supervisor said this about him: "This man is a genius!". Apparently, although he was completely in-adept at coursework, he was very into research and is very good at it.
    – Legendre
    Aug 11, 2012 at 9:26
  • 1
    As someone with a PhD in applied math who never did more than the required homework in courses I'd have to disagree with your first paragraph. There's hardly anything in common between PhD level research and practice problems in an undergraduate textbook. Dec 18, 2015 at 3:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .