Read our COVID-19 research and news.

Master's and PhDs


Dear CareerDoctor,I used to be interested in doing a PhD, but as I read about the difficulties and hardships in pursuing one, let alone in actually finding a faculty position later on, I am beginning to wonder if a PhD is worth the considerable time and effort.I have a master's degree in applied mathematics and am considering going into industry, ideally in an area where I can combine maths with medicine. My professors think that I have what it takes to do a PhD, and that I shouldn't sell myself short by stopping at master's level. What career options are available within mathematical modelling with only my master's? Do you think a PhD would really make me more marketable? And once I've got a PhD, would I need to spend additional years in academia doing a postdoc to become competitive?Ideally I would like a 40-hour-a-week job that will be intellectually satisfying. Is that an oxymoron? My supervisors contend that if I just have a master's, I'll always be working on someone else's project, and feel stifled.I am, of course, considering doing a PhD, but one thing in particular that concerns me about academia is the constant stress--publish or perish! Does that type of mentality also exist outside the university, or is it more relaxed, so to speak?Thank you,Jon

Dear Jon,

You are right to ponder all the pros and cons of a PhD. I do not think anyone should ever embark on such a journey without being fully informed of what lies ahead, and motivated for all the right reasons. Your main concern is a more personal one though--will the investment of time, effort, and possibly money in a PhD pay off in terms of getting you the type of career you've described?

I've heard mixed views on whether or not a PhD is necessary for people considering industrial posts. However, assuming your interest is in a research role--directly applying your knowledge of maths to improve or develop knowledge, my personal view is that having a PhD would be to your advantage. I feel industry is often looking for qualities (advanced, abstract problem-solving for example) that are best developed by taking responsibility for a comprehensive, independent research project.

This is supported by the findings of a report from the Society for Industrial and Applied Mathematics (SIAM) on Mathematics in Industry, which is incredibly comprehensive and addresses many of your questions--master's versus PhD, opportunities in industry, personal reflections on the value of qualifications, and much more. Although the survey was carried out in the United States, and in the mid-1990s, it is still very relevant to you. Even though medically related applications are barely mentioned, please don't let this sway you as things have moved on since then! I'm not sure where you are hoping to work, but if it is in the US, the contents of the SIAM Web site and details of their conferences would convince me to join the society.

I'd recommend that you look at this report and consider the skills and knowledge that you have developed through your master's and the type of contribution that you hope to make to an organisation. I'm not suggesting that only having a master's will prevent you from carrying out research in industry as many case studies I've read refute this, but you need to be clear about your personal ambitions so you can find out from firsthand sources how the careers of master's and PhD graduates develop in relevant companies.

For more career resources for applied mathematicians with master's, I will refer you to the Mathematics Research Scientist, an article I have written for the Prospects Web site. In this profile I link to other sites such as Plus magazine, which features interviews with mathematicians in many different fields and offers case studies of three mathematicians--one of whom worked in industry and another in biomedical applications.

As for the differences between academia and industry, I'll point you towards one of my previous columns. Briefly, you will probably have less freedom in industry, but this may be countered by knowing you are making a contribution to medical understanding or drug development. The pressures are also different--you will still need to achieve success, but it won't be measured in the same terms as in academia, that is by your publication record. Many companies also seem to recognise that a happy, well-balanced workforce is more effective, particularly when the work depends on solving complex problems, so your dream of a 40-hour working week is not unreasonable!

An important point is that your options are not "either or". In fact you can do a PhD which is industrially sponsored, like the CASE studentships offered by UK Research Councils, or even based in industry. Even a PhD student working on an academic project without industrial sponsorship or interest can build links with industrial researchers and get their opinions on ensuring employability. Again, SIAM makes some excellent suggestions.

After a PhD with an industrial sponsor (or time spent building your profile amongst industrial researchers at conferences and events) I don't think that a postdoc would be essential, but by then you may have caught the academic bug. Academic research is a competitive, underappreciated, underpaid, demanding job--but also the dream of many researchers! If you would secretly prefer to remain on the academic track, but are afraid to admit to such an implausible career aim, take heart from David Earn's article which suggested that "there is every reason to believe that embarking on a career in mathematical biology is likely to be exciting and very rewarding." Applying maths in medicine is a fantastic growth area at the moment as last month's Next Wave feature on Careers at the Interface between Maths and Biology pointed out.

All the best in your career,

The CareerDoctor

PS: I am soon to speak about the value of a PhD at a conference (now that I've passed the initiation test for Next Wave they are letting me loose on an unsuspecting public), so keep an eye on this column for details of future events.