The Grad School Experience
What to expect when you're trying to become a doctor (but not the kind that can help you)
I spent a while on a Zoom call with one of my research students from last summer, who’s doing an REU in a lab at a major research university this year. He basically wanted to know how comparable his experience there was to the experience of going to graduate school in physics. While we’re coming up on thirty years since I started grad school, I suspect there’s still some relevant experience there, and it might be worth sharing a bit more widely for those who haven’t gone through it yet.
Obligatory caveats: My experience with grad school is, obviously, in the STEM fields, specifically physics (well, my degree is in Chemical Physics, but that was mostly the physics track with a few substitutions). What I can say will be slightly less applicable to theoretical physics students, significantly less applicable to folks in life sciences, and probably of minimal use to folks outside of STEM. If you have a background in one of those, you’ll be welcome to add your experiences in the comments, but please don’t berate me for not accurately describing the experience of an MFA program or some such: I’m telling you up front what this is going to be, and it’s not that.
Summary/ Timeline: For those not familiar, here’s a rough timeline of what the typical graduate school trajectory looks like:
— Year 0: Depending on how you get admitted, there may be an option for doing some summer research before you arrive on campus. This can be a useful way to get your foot in the door with a research group, so it’s worth looking into if that option is available. On the other hand, it’s also perfectly reasonable to want to spend that summer celebrating and relaxing after the completion of your undergrad degree.
— Year 1: Primarily class work, with a fairly set program: Take these three courses for the Fall semester, and those three for the Spring. There will very likely be a qualifying exam of some sort based on the material in these classes at some point. You will likely also be doing some TA-ing to earn pocket money. It might be possible to do some part-time research with a future lab group in this window, but it’s a little tricky to squeeze in. Definitely explore the possible research groups, though, and try to make a connection with those you might like to join.
— Summer 1: This is the point where people start to get hooked up with research groups. You’ll typically get some sort of trial-basis position for the summer, working on whatever they need done by a low-level student. If you’re facing a qualifying exam early in Year 2, this is when you study for it.
— Year 2: Slightly less class work— I think there were two courses per semester on the recommended program instead of three when I went through. These tend to be more focused on your eventual research speciality, which means there may be a bit more of them for theorists than experimentalists (on the experimental side, you dive into lab work a little earlier). If there’s a qualifying exam of some sort, it usually happens here. Probably also a bunch of TA work.
— Summer 2: By this point, you’re hopefully connected with a research group for real; if the previous summer trial didn’t work out, you might be switching to a new group here, if not, you’ll be taking on more responsibility with the group you already started with.
— Years 3 to (N-1): You’re primarily a full-time researcher in these years, ideally supported by a research assistantship paid by the PI of your research group. You might pick up a class or two if an elective that’s offered only occasionally comes up, you might TA or adjunct a course if you’re aiming for a teaching position down the road and want to bolster that section of your CV, but the primary concern here is to get deeply involved in a research project to the level of generating publishable work that you can claim as your own.
Countable metrics here vary a lot from field to field, but you should probably expect to be a contributing author on a few papers and the lead author on a couple more. This doesn’t necessarily mean that you conceive of a wholly original idea and do all the work yourself, mind— it’s very normal for these to be somewhat incremental advances in an ongoing program— but that you’re the one who knows the most about the details of what was done and why, so it makes sense for you to be the one who writes up the first draft of the eventual paper.
(There are fields where authorship of major works is allocated differently enough that no grad student would be the lead author on a major journal paper; in those, I believe it’s more typical to have some smaller more process-oriented papers or technical notes. That’s far enough out of my experience, though, that I can’t give much detail.)
— Year N (where N should be a single-digit number, preferably as small as one can make it): Thesis writing. This is where you pull together everything that you’ve worked on and put it all in a single document, ideally fleshed out with some additional detail that isn’t necessarily appropriate for a published journal article. The exact length of this varies from field to field, and institution to institution— in some places, it can be basically an introduction to a collection of papers, in others it’s a whole new document. Mine ran to 181 double-spaced pages (he says, looking at the PDF), not counting the front matter; that was arguably excessive, but also, a lot of the text for it came from papers that had already been published, with a bit of re-writing to suit the change in format.
(Funny story: Since I started each experimental chapter from the text of one of the published papers, none of the papers I was an author on was actually cited in the final draft that went to the committee. This led to an awkward moment when the out-of-program person showed up for the defense and asked “How can you possibly plan to give this clown a degree when he hasn’t published anything?!?” Which was very confusing to my actual advisors (“What are you talking about, he’s got three Phys. Rev. Letters?”) until they realized what had happened. Thankfully, they had that conversation behind closed doors…)
— Year N, very end: The final stage of the process is the thesis defense, where you give a presentation to a committee of faculty including the PI of your research group, at least a couple people in related fields, and generally somebody from a significantly different area. This is basically a seminar talk followed by some pointed questioning about the work you did, and also ranging a bit afield from that. This is a nerve-wracking experience in a lot of ways, but also if you’re doing things right, something of a formality— they wouldn’t let you schedule the defense if they weren’t willing to give you the degree. Play your cards right, and the snake will be small and docile, and at the end everybody shakes your hand and addresses you as “Dr. Lastname.”
With that out of the way, some things to be aware of and miscellaneous advice:
Point 1) It is basically inevitable that you will initially feel out of your depth, particularly in the research context. This is mostly structural: when you’re joining a research group, you’re coming into a context filled with people who have been working on this exact problem for months or years before you got that. This is in stark contrast to most undergrad experiences, where you move through in a fairly linear way with a cohort of people who are at the same stage you are, and are just as confused. (The class component of graduate school will feel much more “normal” in this regard— there will be one or two crazy brilliant people who aren’t visibly exerting themselves to solve the homework sets, but most of the group will be at a roughly comparable level of confusion, and you should find them and work with them.)
Generally speaking, people in the group, particularly senior students, are not going to be deliberately trying to confuse you, they’re just so immersed in what they’re doing that they’ve forgotten what it’s like to not know the basics. It’s perfectly reasonable to ask questions, even very basic ones, and outside of a few total assholes, people in the group will be happy to help you get up to speed.
Point 2) Research will involve a surprisingly wide range of tasks, which can shift with very little notice. At various stages of my graduate career in experimental physics, I was doing stuff that was clearly identifiable as physics of the sort I knew I was signing up for (tweaking up lasers and shining them on atoms), but also a lot of other stuff: making small parts in the machine sharp, doing unlicensed plumbing and electrical work, writing C++ code to do data analysis, debugging badly documented LabView programs and writing new badly documented LabView programs. I spent one whole afternoon hanging whiteboards in the hallway outside the lab, which I remember because the person who was (I think) third in charge of the Physics lab at NIST walked by while I was standing on a chair with a cordless drill and said “Some day, I would love to read your job description…”
The basic rule is that you’re going to have to do whatever needs doing to advance the research project, and that can change dramatically with very little notice. You might walk in on Monday thinking you’re going to zap some atoms with some lasers, and by Wednesday you’re doing second-rate software engineering, and the next Monday you’re repairing leaky pipes.
You’re going to have to pick up some odd skills on the fly, which again will be very disorienting. And, again, almost nobody is deliberately trying to confuse you, so if you need help getting up to speed on a thing you haven’t done before, it is perfectly okay to ask for help.
Point 3) You will probably never be around this many people you can learn from again in your life, so take advantage of that. This one’s more advice than observation. Every academic department will have some sort of seminar or colloquium series where they bring in the smartest and most interesting people they can find to talk about their research. You should go to those, and not just because they usually involve some level of free food. You’ll get the chance to hear first-hand about an incredible range of work, and should make the most of that opportunity.
Point 3a) Be aware that many of these talks will be baffling, at least at first. This, again, is largely structural— most seminar talks are by people who are deeply immersed in the subject, and will at some point veer off into technical matters that very few people in the room will really understand. Also, some of them will also just be bad at public speaking; that sucks, but hey, there’s free food. As with previous points, if you find something baffling, it’s perfectly okay to ask other students after the talk “Hey, what was the deal with that?” That’s a great way to sort out what was confusing because you’re missing some key background, and what was confusing because the speaker was terrible.
Point 4) Get to know as many people as you can. This is the area where I come closest to regretting any of my grad school education. I did my lab work at NIST, which was a fabulous experience in a great many ways, but also meant that I was generally the only grad student in a group full of post-docs. This means that I don’t have much in the way of contacts or connections with people in my grad school cohort, and that’s kind of a shame. It made some homework sets harder than they probably needed to be— I would’ve been better off studying in a group— and reduces the network of people I know in physics, particularly outside the world of cold atoms.
(I do think that, on net, I came out ahead. But there are a lot of classic grad school bonding experiences that I never had, and that’s a bit of a shame.)
Point 5) Everybody around you was once where you were, and someday you’ll be where they are. This might seem repetitive, but it’s important enough to hammer on: all the senior people in a research group were once junior people, and have at least dim memories of what it was like to try to get up to speed on a new project in a hurry. They’ll generally be helpful if you ask for help, and the few assholes who won’t are good to identify early.
And if you embark on this course, keep in mind that someday you’ll be the senior student who seems incomprehensibly smart to the new student, who will be terrified to ask you for help. So when they do ask you for help, give it. And maybe even consider reaching out to them first when they’re clearly trying to work up the courage to ask about something they don’t understand.
And that’s the condensed version of my advice about the general experience of graduate school (again, primarily relevant to physics, and probably minimally relevant outside of STEM). If you like this sort of thing, here’s a button to get more of it:
And if you’d like to fill in any gaps I left, the comments will be open:
Excellent job Chad. It parallels my particle physics PhD experience essentially exactly.
Chemistry graduate school is very similar.
I would say that, depending on your supervisor's grant rate, you are likely to TA (teaching assistant) for 1 or more years...