1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Is a physics PhD becoming unemployable

  1. Oct 3, 2014 #1
    It seems to me, correct me if I'm wrong, the vast majority of physics PhD's do not work in physics, science, or even related disciplines (applied science, engineering, etc). From what I understand, the majority seem to be working in finance, software development, and data mining.

    At one point in time in the past, it wasn't uncommon to see physics PhDs and even those with only a bachelors working engineering type jobs. These days this is almost never possible and a physics PhD is deemed as qualified for a related engineering job by HR as is someone with a PhD in history. If this shift to credentials rather than hard skills occurred in engineering, what's to stop this from eventually happening in all of the common industries most physics PhDs tend to work in? In software development, I myself am seeing the beginning of this and the strict requirement of a computer science degree is non-negotiable for a majority of the job postings I've seen. What exactly would stop the same from happening in data mining and finance jobs, where HR and the hiring manager deem only those with relevant credentials to be qualified for an interview?

    Are the days of a physics PhD using the skills they learned to retool for a different but tangentially related career numbered? Will there perhaps soon be a large number of unemployable PhD graduates who need to restart the training process for a different career in a formal setting (i.e college/university)?
    Last edited: Oct 3, 2014
  2. jcsd
  3. Oct 3, 2014 #2


    User Avatar
    Gold Member

    Any PhD candidate ought to be planning on a college teaching career; otherwise, don't do it.

    This is just the reality of the situation. In a vast number of cases, you will be over qualified and immediately discarded for a job. I'm sure that this has happened to me more times than I like to think about.
  4. Oct 3, 2014 #3
    I would agree, except the chances of getting a college teaching career may not be that great, unless it's as an adjunct, which is an unenviable position to be in. So, I would tend to say just don't do it, period, in most cases. However, there are exceptions. There are people who are so far ahead of the game (to an extent bordering on the absurd, rather than just a run of the mill top student) that they can be assured of success and people who have off the charts abilities, as far as selling themselves and getting a job. There are also people who just want to study the subject for the sake of studying it and are willing to accept the risks and trade offs. Rather than planning on teaching at the college level, there should be a back-up plan.

    Right now, data science is the next big thing, but the window of opportunity may have already closed for really flexible hiring in that area because more people are developing that skill set. I was just reading an article which mentioned how LinkedIn originally just hired people who they thought could learn to do it a few years ago, but already, they expect you to know more and more to be considered. To me, it seems easier to qualify for software developer jobs. I see quite a few postings that are flexible on whether they want a CS degree, still.

    There may be other "next-big-thing's" that pop up. I suppose that's where there is hope.
  5. Oct 3, 2014 #4

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    http://www.aip.org/sites/default/files/statistics/employment/phdinitemp-p-10-fig3.jpg [Broken]
    Last edited by a moderator: May 7, 2017
  6. Oct 4, 2014 #5


    User Avatar
    Education Advisor

    homeomorphic, you keep claiming in all of your posts to not do a PhD, but don't you think your advice is based strictly on your own experience as a math PhD, and may not apply to other people's experiences?

    As far as whether a physics PhD is becoming unemployable or not, this varies tremendously between what specific sub-field within physics the PhD had worked/research on, as well as any other set of skills that person brings to the employer. Also, it is worth mentioning that, at least in my experience here in Canada, it was never that common for physics graduates at the undergraduate level to be working as engineers, because most physics undergraduate programs that aren't engineering physics simply don't teach the various skills that are required for most engineering jobs. And employers have never looked at physics undergrads as potential employees.

    As for data science being the next big thing, it is worth keeping in mind that data science is a particular industry that doesn't really have a built-in degree or credentialed program associated with it. Data science, IMO, is a cross between statistics and computer science, so anyone with quantitative skills (obviously statistics and computer science, but also applied math, physics, industrial engineering/operations research, or electrical engineering) can break into this, and I don't really see this changing any time soon.
  7. Oct 4, 2014 #6
    I have admitted as much. However, it is FAR, FAR from being only my own personal experiences. If you notice, if anyone asks about PhDs in engineering, operations research, computer science, or statistics, I don't object to those, although it still should be taken on only with great caution because it's a difficult process and takes a lot of time without getting paid well. Math and physics can also become much more viable if you choose the right subfield. When you get into the details, my opinion is more nuanced than you think and if you read everything I write, I think that comes across. Even if you study something ridiculous, like topology (for which I am guilty of studying), you can be in much better shape if you have a good exit plan, but you have to be VERY serious about your exit plan and not treat it as just an afterthought.

    Whether or not anyone should do a PhD is a subjective decision. I don't presume to make up anyone's mind for them, just to give my take on it. In the majority of cases, I think I am right. It's a bad deal for the majority in math and physics, specifically. I don't think it's a stretch to say that at all. There are some people for whom it will work out well and some for which it will sort of work out, but at a great cost.

    It's true that there are people within math and physics who can break into data science, but, for example, for me, it's fairly non-trivial because I don't have a strong statistics background, which will be true for a lot of physics people, as well. If you look at the job postings, you'll see that they generally don't want someone who just knows math or physics and can learn other things. They want someone who ALREADY has at least some of the skills.
  8. Oct 4, 2014 #7


    User Avatar
    Science Advisor
    Education Advisor

    One thing that's important to keep in mind is that academia is not job training in the first place. It's an education.

    I'm never sure if I articulate this point all that well or not. But job training is something that happens mostly at the community college level. There, you enroll in a program that's designed to give you a set of skills and the accompanying certifications so that you can perform the duties that come with a specific career. There are professional programs at the university level such as medicine, law, or engineering that perform a similar function. Academic programs such as most of the sciences and humanities give you an education in a specific field, but then leave you as an educated graduate to figure out how to apply that education in a career.

    In my opinion, there will always be options open for well-educated and well-motivated people (at least in physics), but I think you're correct in so far as a trend of increasing credentialing and certification in fields that are traditionally non-certified. Movement in this direction often comes in fields where there are drastic consequences to professional errors. In medical physics, about a decade or so ago, a PhD from any other branch of physics could easily step into a residency or even a junior physicist position where they would learn what was needed on the job. But since then the field has adopted a more rigid training and certification sequence, for a number of reasons. One of those is demonstrable diligence on behalf of the organizations that hire medical physicists - if you're facing a lawsuit because you over-irradiated a group of patients, you want to be able to demonstrate that the people you've hired to keep your machines calibrated and functioning properly know what they're doing.

    It's not too difficult to see how such a trend might apply in other fields. In financial circles, organizations that have traditionally employed physics PhDs may move towards certification in an effort to avoid catastrophic investment errors.

    In engineering - well, I think certification has always been the case (at least since the last few decades). When physics PhDs work as "engineers" they are often not performing the professional aspects of the job, such as signing off on designs, rather, they're applying very specific skill sets that they've developed to further advance the development of industrial interests, often in cases where your typical engineering graduate can't. I think, in such cases you'll continue to see physics PhDs getting work because the specific nature of the work overrides the need for certification. But, over time, as the new development becomes more common, again you'll see movement towards increased certification.
  9. Oct 4, 2014 #8

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Homeomorphic, your experience as detailed here and in your other posts, suggests a sense of surprise that there is not a booming market in commercial topologists and also that industry doesn't feel that topologists make better engineers than engineers. Do you think it might be possible that your experience is more personal and less universal?
  10. Oct 4, 2014 #9

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Not everyone here will agree with that.
  11. Oct 4, 2014 #10
    I think you know that that isn't exactly a fair characterization. Topology has absolutely zero to do with my surprise. I always knew no one wanted topology and never even suggested otherwise. When someone gets a PhD in math, they get a PhD in math. Not topology. They just did their research on that, but it's not the only thing they know. Also, I studied EE for 3 years, so we're not talking about topologists making better engineers than engineers, but rather a topologist with an engineering background, and I'm also comparing myself with other entry level people. I know what people learn when they get an EE degree and I know almost as much as someone who just graduated with a fresh degree. I wouldn't initially be any good at it, but I think I could compare to some of the recent grads in EE, who also have no experience and wouldn't be particularly great at it at first. I actually remember what I learned in my undergrad to some degree. What I thought is that maybe there would be someone willing to hire someone who could learn. Besides, I haven't even tried to get engineering jobs. I just looked and concluded it would be too hard, so then I turned towards actuarial work or programming.

    I know my experience itself is personal. But you are assuming that my opinion is based on my own personal experience, which is false. My own personal experience is just one piece of data out of many.
  12. Oct 4, 2014 #11
    Well, academia is a job. And a PhD is mainly training for a certain job--that of an academic. So, basically, we have tons of people getting trained for a job that they aren't going to get. It is a personal choice whether you think it's worth getting an "education" is worth it for its own sake. You can't impose your opinion on other people. I am not trying to impose my opinion on anyone. I am just letting them know that the education they are getting IS for its own sake, rather than for a job. Whether they think that is good or bad is up to them. But if they don't think it's worth doing for it's own sake, they need to be fully informed, so they can either train themselves for the job they are going to get on the side while they get their "education", OR just focus on something more relevant to the job, if they are more risk-averse in terms of employment prospects.
  13. Oct 4, 2014 #12
    That's interesting, as it runs completely counter to my experience. I have met quite a few physics Alumni at my school and even hired for an internship by a physics MS who all work as engineers. Thing is, most of them finished their education 15-20 years ago. My guess is that they were able to break in at a time when there were more emphasis on hard skills and knowledge rather than the relevant credential.
    Last edited: Oct 4, 2014
  14. Oct 4, 2014 #13
    What hard skills does a physics curriculum provide with respect to engineering? Nearly none in my experience.
  15. Oct 4, 2014 #14
    The thing that I think will be universal is the job postings for the generic quantitative career-changer type person. If you take even a cursory look at those postings, you will see the source of my pessimism about the job market. It's not my experiences as a topology PhD. It's about what I see employers saying they want. THAT was the shock for me. I knew they didn't want topology and even most graduate level math. What I didn't know was that they wanted x, y, and z, and that they aren't going to train you for that--you are just expected to know it. I thought that they would train me in x, y, and z because they know I'm smart, and I can learn it. THAT is where I was sorely mistaken, for the most part. There may be a few exceptions to this, and in fact, even as we speak, I may have found one of those few exceptions. It may be true that it took me surprise that industry doesn't want topologists, but I never even came within a hundred miles of the assumption that they would want to hire them AS topologists (with a few ultra-competitive exceptions, such as Microsoft research or something). Of course that is silly. You have to use some common sense and give me some credit when you interpret what I say.

    I also realize that there are people out there who will be more prepared for those job postings that were so discouraging to me. But I know most physics and math people aren't going to be a whole lot better off than I am. So, that's what I am trying to get across. Don't listen to me, go look at the job postings because that's the thing that's driving my opinion on this. I am just trying to make sure people don't get slapped in the face with this towards the end of their degree, like I did, when it's too late to be adequately prepared. All I'm really saying is don't count on the mythical career-changer type jobs that you've heard about, like all the physicists on Wall Street. It doesn't happen automatically. You have to prepare yourself for that or else be very lucky. There are people who know what they are doing and may know how it is that they will get a job, and that's fine. The point is to try to get some strong evidence that your career is safe before doing something stupid like getting a PhD in topology or string theory or whatever.

    So, my main point is not just "don't get a PhD". My main point is to make sure you are going to get a job, PhD or not. The don't get a PhD part is secondary and has a lot to do with things besides getting a job, as well.
  16. Oct 5, 2014 #15
    Depends on the curriculum. My school requires two semesters of computational physics and a semester of basic electronics. There's a second electronics course that goes a lot more in-depth into digital electronics and rudimentary transmission line and antenna theory. A time-series analysis course offered gives at least a good introduction to signal processing. There are also two advanced experimental/computational physics courses one could take if they are interested and it is even possible (with permission) to take up to two upper year electives in the EE department if one is interested.

    If all you did was E&M and quantum along with no programming, signal processing, and electronics, then you would be right about having no hard skills that are transferable to engineering. However, it seems none of that really matters anymore.
    Last edited: Oct 5, 2014
  17. Oct 5, 2014 #16
    I took classes like that of my own accord during and after my physics degrees. They were far more useful for me now on the job as an engineer than any of my physics curriculum classes were. I did have one semester of electronics required, but it was super basic. Just a little AC, DC, filters and op-amps.

    Interesting that two terms of computational physics is required for your physics BS. That's good, I can see how that is useful. Neither my undergrad nor my grad school required any programming or numerical analysis at all for a physics degree.
  18. Oct 5, 2014 #17
    Exactly. It happens that along the way to my topology PhD, I did E&M, I did signal processing, I did electronics, lots of programming, and much much more. The fact of the matter is that I know my stuff just as well as some new EE grad who doesn't remember every last thing from their classes, any more than I do, apart from specializing a little bit more in their senior year, plus all the math involved is trivial to me. I would like to think my extra general math skills and other skills could make up for that extra year that I missed (and it wasn't even a whole year, by the way). The truth is that the reason why my dad's lousy EE student that he was complaining about can get an EE job and I can't is more a matter of how the system works than who is good at what. Vanadium can disbelieve that all day, but that's not going to make it true. Has nothing to do with topologists being good at engineering. It has to do with someone studying engineering, but getting a piece of paper that says something different on it being good at engineering.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook