Career Paths for PhD Students in Biostatistics

As a biostatistics PhD alumnus who just graduated earlier this year, I thought I would share some of the things I learned regarding potential career paths. If you are a current PhD student, I would encourage you to seriously think about your future career plans in your third or fourth year, with some level of detail beyond “I want to stay in academia” or “I want to go into industry.” This gives you more time to prepare and plan (e.g. if you want to do an internship1). Hopefully, this post will be a useful starting point for your research into different career opportunities.

In the sections below, I summarize several different career options. It will inevitably be a high-level overview, since I have not actually pursued most of these paths myself. Additionally, the range of options I’m discussing is strongly influenced by what I’ve seen biostatistics alumni (specifically from my department at JHU) pursue as careers. This doesn’t necessarily mean you can’t do something others haven’t done before; what I discuss below is just indicative of where people with a similar background have successfully gone before.

Academia

There are a couple different options within academia that you can pursue.

Tenure-track research professor

Your advisor is likely in such a position as well as many other professors in your department, so you probably have some familiarity already with what such a position entails (i.e. research, mentoring, teaching, service), but they would be good people to talk to in order to learn more about this career path.

Depending on the institution, your funding as a professor may be “hard” (your salary comes from the school/tuition) or “soft” (your salary comes from your own grants) or a mix of both. For example, professors at JHU talk a lot about writing grants because theirs is mostly a soft money position. More on this is covered by SimplyStatistics’ Rafael Irizarry here and here.

Because such positions tend to be highly competitive, there is a growing trend to do a postdoc before applying in order to craft a more competitive CV. Likely due to this trend, I believe that the most common2 first role after graduation for alumni from my department is to do a postdoc. However, judging from the faculty candidates I’ve seen interview at JHU over the years, there are still plenty of candidates coming straight from a PhD – they may just be somewhat disadvantaged in comparison to a postdoc since they haven’t had as many years to do research and get published.

Research faculty member

At JHU’s biostats department, there are also research-track faculty (other institutions may have different names) who are not tenured that conduct independent research. Aside from the difference regarding tenure, there may be other additional limitations, though what these are will vary from institution to institution. For an in-depth perspective on what such a role looks like at JHU, I recommend reading this post by John Muschelli.

Teaching professor

Some of the alumni in my department who love teaching go on to pursue more teaching-focused positions. One option is to join a tenure-track position at a liberal arts college. For example, Leslie Myint, an alum, is now a professor at Macalester College. In comparison to research professors at R1 universities, such positions emphasize teaching more, though they may still be expected to do some research in order to obtain tenure.

At many research universities, there are also teaching faculty positions, whose primary responsibility, unsurprisingly, is to teach. This may be another good fit for those who enjoy teaching. Shannon Ellis, a former postdoc who is now a teaching professor at UCSD, wrote about her career path here.

Government

There are research positions in government that are common for our alumni to enter as well, particularly at the NIH and FDA. For example, Haoyu Zhang is an alum who is now a tenure-track investigator at NCI. My sense is that there is considerable overlap in research and collaboration between academia and some of these institutes. Another alum, Therri Usher, now works as a statistician at the FDA and talks about her experience here.

Industry

I think many PhD students underestimate the sheer variety of positions in industry, since our perspective is coming from an academic environment. This is why, as I mentioned earlier, it’s not sufficient to say you want to go into industry since “industry” is such a broad term. I will describe below the main sectors I’ve seen biostatistics PhD students go into, but keep in mind that I won’t be able to describe everything and the job landscape is always evolving.

Statistician in Pharma

This is probably the most common industry sector for biostats PhD alumni, as there is a long-standing pipeline going from biostatistics to pharmaceutical companies. The job title is usually something like “research scientist” or “statistician.” Regardless of title though and depending on what team, division, and company you’re going to, the job responsibilities can lean more into the clinical side, where you are a statistician involved with running a clinical trial, or more heavily into the research side, where you work on research projects similar to academia.

Data Scientist in Tech

Data science is an infamously vague term and there is no real consistency in how it’s defined across different companies. However, in my view, there are roughly 3 different types of data scientists in tech today, any of which you may want to pursue depending on your skills and interests.

  1. Modeling and inference, where the goal is to deliver useful insights with some mixture of stats/ML (e.g. this is most similar to the role I’m currently in at Amazon)
  2. Analytics, where the goal is to design and run A/B tests; they may overlap with product analysts at some companies (e.g. Hilary Parker is an alum and talks about her job at Etsy here)
  3. Engineering, where the goal is to build ML products; they may overlap with machine learning engineers at some companies (e.g. Alyssa Frazee is an alum who previously worked as an MLE at Stripe)

For more on how different companies view the data science job family, you can read these posts (1, 2) by Lyft or this post by Airbnb as examples. At Amazon, where I currently work, there are 3 different science roles: data scientist, research scientist, and applied scientist, but these job titles don’t really map to role delineations at other companies. To understand what a specific company means by “data scientist,” you’ll have to read the job description and understand what your job responsibilities are and what skills they are asking for.

Some big tech companies also have separate research divisions. My impression is that they tend to be more focused on CS research, though there are some stats research roles as well.

Data Scientist in Biotech

For many biostats graduates, their knowledge and skills will transfer particularly well to data scientist positions in biotech companies. These can range from early-stage startups to more established companies like 23andme or Flatiron Health. Depending on the role, it can lean more heavily into the research side, where you’re expected to write papers and publish, or more on the applied side, where you work on the statistical analysis and coding.

Quant in Finance

Some biostats PhD students also go on to become quantitative researchers in finance, though I’m not familiar with the details of what this role or industry looks like. This seems to be a far more common path for stats PhD alumni than biostats PhD alumni, however, likely due to pre-existing differences in interests.

FAQ

The above summarizes the different career paths I see being pursued by biostatistics PhD graduates. Below, I will summarize some of my remaining thoughts related to career planning.

Master’s vs PhD?

Some of you reading this may not be a current PhD student, but a master’s student or college student, so I’ll share a little bit about what I think regarding a master’s versus a PhD regarding job opportunities. Very broadly speaking, the more research-y a job is, the more likely a PhD is required or will be helpful because a PhD is about training you to be a researcher. On one end of the spectrum, an applied data science role probably won’t require a PhD, though it may make you somewhat more competitive in the job market. On the other end of the spectrum, a tenure-track research professor position will require you to have a PhD.

Salaries

An important part of researching different jobs is also understanding what the compensation differences are. To help you with your research, here are a couple sites you might find useful:

  • The ASA conducts salary surveys and publishes their results here. You can look at the broad differences in compensation between academia and industry, for example.
  • https://www.levels.fyi/ collects anonymous self-reported compensation, primarily for employees in the tech industry.
  • Glassdoor also collects anonymous self-reported compensation, though I’ve personally found that it doesn’t do a good job of capturing the distinctions in base salary, bonus, and stocks (stocks make up a large part of many tech companies’ compensation package, so this is particularly relevant in that industry).
  • https://h1bdata.info/ is a database of the base salaries given to H-1B visa holders; it’s required by law to be public.

Additional Reading

Lastly, here are a few additional links that may be interesting to peruse.

  • 80,000 Hours is a career planning site with job profiles; the one on academic research and data science may be of particular interest.
  • Jeff Leek’s guide to career planning. Jeff is a former JHU biostats professor and currently works at Fred Hutchinson Cancer Center.
  • Jakob Fiksel’s thread on Twitter discussing his post-PhD job search. Jakob is a biostats PhD alum and currently works at Vertex Pharmaceuticals.

  1. I previously wrote about my experience applying to internships here, if you’re interested in what that is like.↩︎

  2. As in the mode, not the majority. You can view the PhD alumni placement for Johns Hopkins’ biostatistics program here.↩︎