Posted to Guides  /  Tags: ,

A Survival Guide to Starting and Finishing a PhD

Tips on making it through, what I would tell my previous self going in, and advice on taking advantage of the unique opportunity that is graduate school.

Disclaimer: Everyone’s graduate school experience is different. Mine wasn’t a typical one, mainly because I spent so much time away from campus (in a different state), but hey, most of your PhD experience is independent learning anyways. That’s the best part.

Before you begin (or apply)

You should really like the field you’re thinking about pursuing a PhD in. You don’t have to have this, but you kind of do. A doctorate is a commitment of several years (for me it was 7), and if you’re not fascinated by your work, it feels like an impossible chore. There are a lot of things that are actual chores — administration, research results that go against your expectations, challenging collaborations, etc — and the interest in your work pulls you through.

I don’t know anyone who finished their PhD who wasn’t excited about the field in some way.

On that note, do your research before you apply to programs, and try to find faculty whose interests align with yours. Of course this is easier said than done. I entered graduate school with statistics education in mind and came out the other end with a focus in visualization. The size of my department probably allowed for some of that flexibility. Luck was also involved.

So what I actually did was apply to more than one program and then wait to hear if I got in or not. If I only got into one place (or none), then the decision was easy. In the end, I compared department interests and then went with the one I thought sounded better.

Consider it a red flag if it’s hard to find faculty information because there’s little to nothing online. There’s really no excuse these days not to have updated faculty pages.

Absorb information


Okay, you’re in graduate school now. The undergrads suddenly look really young and all of them expect that you know everything there is to know about statistics (or whatever field you’re in). This becomes especially obvious if you’re a teaching assistant, which can feel weird at first because you’re not that far out of undergrad yourself. Use the opportunity to brush up on your core statistics knowledge.

I had coursework for the first two years, but it varies by department I’m sure.You also take classes yourself. Don’t freak out if the lectures are confusing and everyone seems to ask smart questions that you don’t understand. In reality, it’s probably only a handful of people who dominate the discussion, and well, there’s just always some people who are ahead of the curve. Maybe you’re one of them.

Tough early goings has a lot to do with learning the language of statistics. There’s jargon that makes it easier to describe concepts (once you know them already), and there’s a flow of logic that you pick up over time.

There’s usually a qualifying exam after the first two years to make sure you learned in class.Don’t hesitate to ask questions and make use of office hours (but don’t be the person who waits until the week before an exam or project to get advice, because that’s just so undergrad). Once you finish your coursework, it’s going to be a lot of independent learning, so take advantage of the strong guidance while you can.

The key here is to absorb as much information as you can and try to find the area of statistics that excites you the most. Pursue and dig deeper when you do find that thing.

I remember the day I discovered visualization. My future adviser gave a guest lecture on visualization from a mostly media arts perspective. He talked about it, I grew really interested, and then I went home and googled away.

Oh, and read a lot of papers. I didn’t do nearly enough of this early on, and you need proper literature review for your dissertation. Background information also informs your own work.

Find an adviser


Actually, I don’t think I ever officially asked my adviser to be my adviser. It was just assumed when I became a student researcher in his group.I kind of had an adviser from the start of graduate school, because I was lucky to get a research assistant position that had to do with statistics education. However, as my interests changed, I switched my adviser around the two-year mark.

This is important and goes back to the application process. After a couple years, you should have a sense of what the faculty in your department work on and their teaching styles, and you should go for the best match.

I think a lot of people expect an adviser to have all the answers and give you specific directions during each meeting. That’s kind of what it’s like early on, but it eventually develops into a partnership. It’s not your adviser’s job to teach you everything. A good adviser points you in the right direction when you’re lost.

Jump at opportunities


Statistics is a collaborative field, and there are a lot of opportunities to work with others within the department and outside of it. A lot of companies are often in search of interns, so they might send fliers and listings that end up posting to the grad email list. Jump at these opportunities if you can.

Graduate school doesn’t have to be expensive.Opportunities within the department or university should be of extra interest, because it usually means that your tuition could be reduced a lot, if not completely.

If something sounded interesting, I’d respond to it right away, and it usually resulted in something good. A lot of people pass up opportunities, because they see the requirements of an ideal candidate and feel like they’re not qualified. Instead, apply and let someone else decide if you’re qualified. There’s usually a lot of learning on the job, and it’s usually more important that you’ll be able to pick up the necessary skills.

At the very least, you’ll pick up interview experience, which comes in handy later on if you want one of those job things after you graduate.

Learn to say no


As you progress in your academic career, you’ll look more and more like a PhD (hopefully). You have more skills, more knowledge, and more experience, which means you become more of an asset to potential collaborators, researchers, and departments. A lot of my best experiences come from working with others, but eventually, you have to focus on your own work so that you can write your dissertation. Hopefully, you’ll have a lot of writing routes to take after you’ve jumped at all the opportunities that crossed your desk.

So it’s a whole lot of yes in the beginning, but you have to be more stingy with your time as you progress.

There are probably going to be potential employers knocking at your door at some point, too. If you really want to finish your PhD, you must make them wait. I know this is much easier said than done, but when you start a full-time job, it’s hard to muster the energy at the end of a day to work on a dissertation. I mean, it’s already hard to work on a dissertation with normal levels of energy.

All the times I wanted to quit, I justified it by telling myself that I would probably have the same job with or without a doctorate. I also know a lot of people who quit and are plenty successful, so finding a job didn’t work for me as a motivator. But it might be different for you, depending on what work you’re interested in.



This might’ve been the toughest part for me. During my first two years in school, I hung out with my classmates a lot and we’d discuss our work or just grab some drinks, but I had to study from a distance from my third year on. I’ve always been an independent learner, so I thought I’d be okay, but my first year away, it was hard to focus, and it was lonely in the apartment by myself. I didn’t want to do much of anything.

I eventually made friends, and pets provided nice company during the day. It’s important to have a life outside of dissertation work. Give your brain a rest.

Separation from the academic bubble wasn’t all bad though. FlowingData came out of my moving away, and my dissertation topic came out of a personal project. So there are definitely pros and cons, but it’s mostly what you make out of what you have in front of you.

I found Twitter useful to connect with other work-at-homers and PhD Comics proved to be a great resource for feeling less isolated.Anyways, my situation is kind of specific, but it’s good to have a support system rather than go at it alone. I mean, you still have to do all the work, but there will be times of frustration when you need to vent or talk your way through a problem.

Write the dissertation and defend


Despite what you might’ve heard, a dissertation does not write itself. Believe me. I’ve tried. Many times. And it never ever writes itself.

I even (shamefully) bought a book that’s lying around somewhere on how to write your dissertation efficiently. That’s gotta be up there on my list of worst Amazon impulse buys. The book arrived, I started reading, and then realized that it’d be a lot more efficient to be writing instead of reading about how to write.

Procrastination comes in many forms.

The hardest part for me was getting started. Just deal with the fact that the writing is going to be bad at first. You come back and revise anyways. I’ve heard this advice a lot, but you really do just have to sit down and write (assuming you’ve worked on enough things by now that you can write about).

If you already have articles on hand, it doesn’t hurt to take notes so that it’s easier to clean up citing towards the end.Don’t worry about proper citing, what pronouns to use, and the tone of your writing. This stuff is easy to fix later. (It can be helpful to browse past dissertations in your department to learn what’s expected.) Focus on the framework and outline first.

Just google “successful PhD defense.”By the time you’re done writing, you know about your specific topic better than most people, which makes your defense less painful. There’s a lot of online advice on a successful defense already, but the two main points are (1) your committee wants you to succeed; and (2) think of it as an opportunity to talk about your work. In my experience and from what I’ve heard, these are totally true. That didn’t stop me from being really nervous though and probably won’t help your nerves either, but there you go.

I like this video by Ze Frank on public speaking.The best thing to do is prepare. Rehearse your talk until you can deliver it in your sleep. Your preparation depends on your style. Some like to write their talks out. I like to keep it more natural so it’s not like I’m reading a script. Go with what you’re comfortable with.

It’ll all be fine and not nearly as horrible as you imagine it will be.

Wrapping up

So there you go. A PhD at a glance. Work hard, try to relax, and embrace the uniqueness of graduate school. There are many challenges along the way, but try to learn from them rather than beat yourself up over them. A PhD can be fun if you let it.

Any graduate students — past or present — have more advice? Leave it in the comments.

About the Author

Nathan Yau is a statistician who works primarily with visualization. He earned his PhD in statistics from UCLA, is the author of two best-selling books — Data Points and Visualize This — and runs FlowingData. Introvert. Likes food. Likes beer. Follow him @flowingdata.

Become a member. Learn to visualize your data. Support FlowingData.

Join Today


This is for people who want to learn to make and design data graphics. Your support goes directly to FlowingData, an independently run site.

Benefits of Membership

  • Instant access to tutorials on how to make and design data graphics
  • Source code and files to use with your own data
  • Four-week course on visualization in R
  • Hand-picked links and resources from around the web


  • Excellent post Nathan!

    A major part of grad school for me has simply been learning how to be a grad student. Sounds obvious at first, but navigating a PhD program and dealing with problems that arise is easier said than done. By the time one earns a PhD, they are both an expert in their field and an expert at grad school.

    This is especially true when it comes to saying ‘no.’ That, above all else, is so important! It’s tempting to work on every project that gets mentioned, collaborate on every paper idea that comes your way, and volunteer your time helping peers neighboring departments. While those all have the potential to lead to great additions to your CV, the truth is that if you agree to them all, you won’t finish a single one.

  • Great piece! It is important to follow your passion and make a contribution to your profession. However, I would add “learn to play the political game as well.” As a preamble, I was thrown out of my PhD program last year for “taking to long.” This happened to me twice!

    My original PhD was to be in Geology (1984), but I got caught up in an internal battle between two faculty and became collateral damage. My advisor (a real mentor) was on sabbatical and was unavailable to help (it was the pre-internet days). I got to finish my PhD for publication while supervising his lab, but without the benefit of getting a PhD. My second PhD attempt (2002-2012) was in Public Health/Epidemiology while working as a full-time professional and raising a family. This time, I ended up in a box checking exercise without having the right boxes checked. My original advisor was forced out of the institution (politics again) and I was left without a politcal supporter.

    I want to be clear that I am not bitter, just disappointed. I thoroughly enjoyed most of my academic training and it has benefited me greatly on the professional side. It would be nice to have the PhD to go behind my name, but it isn’t an ego requirement to make me feel positive about myself. I do make a contribution to my community and profession without it. But ego is a big part of academia, and it is important to learn that lesson. It is a necessary evil, just like taking required courses that aren’t that interesting.

    The real value that anyone gets out of graduate school is in your education, not the degree. So get your education, make a contribution to your profession and community, and express your love and appreciation for the people around you. That is what makes you a success . . . the extra letters are just a bonus!

    Best regards and best of luck!

  • Here’s my grad school advice (based on my personal experience, so your mileage may vary):

    1. A PhD program represents a huge investment of time, energy, and opportunity cost. If getting a doctorate is not the single most important goal in your life it’s probably not going to be worth it in the end.
    2. The program will consume your life. Before you even begin, you should abandon any thought of having a personal life outside of the program. Just about every one of my grad school colleagues who was already married when they started the program either (a) got divorced before they finished their doctorate, or (b) dropped out of the program because it was ruining their marriage.
    3. Estimate the maximum amount of time you think it will take you to complete your dissertation, then triple it. It’s going to take a lot longer than you expect, trust me.
    4. The 80-20 rule applies: You’ll complete the first 80% of the work on your dissertation (i.e. conducting your original research and writing up your findings) in about the first 20% of the total time you’ll spend working on it. But the final 20% of the work (i.e. “housekeeping” tasks such as completing the lit review, making changes and additions suggested by your advisor and other members of your committee, proofreading and editing for style, preparing charts and graphs, doing rewrites, doing more rewrites, doing still more rewrites, etc.) will take up the final 80% of your time. Basically, once you think you’re almost finished, your work has only just begun.
    5. If you’re not sick and tired of your dissertation topic by the time you’ve finished, you didn’t do a thorough enough job on it. Similarly, if you’re not annoyed by the sound of your own writing style by the time you’ve completed your final draft, either your dissertation is too short, or you haven’t re-read it enough times to catch all of the errors that are certain to be there. (Note, however, that no matter how many times you’ve re-read it looking for errors, the final draft you submit to your committee will almost inevitably contain at least one error that you somehow managed to miss.)
    6. This should be obvious, but don’t go to grad school if you don’t like to read … a LOT. I’m not sure about other fields, but if you get your doctorate in one of the social sciences, you should be prepared to read the equivalent of at least one or two good-sized books every week, sometimes even more than that.
    7. Qualifying exams are NOT merely pro forma. Your department takes them seriously, so you should, too. Prepare diligently for them. No matter how much work you’ve put into your studies, you CAN flunk out of a doctoral program if you do badly on your qualifying exams. I know you’re eager to get past them and start working on your dissertation; but don’t make the mistake of skimping on preparation for exams. They’re more important than you think they ought to be; and the professors who grade them are not going to accept second rate work.
    8. The dissertation defense, on the other hand, IS pretty much just a formality. I guess it’s possible to blow a defense if you somehow come across as a complete idiot in front of the committee; but it’s pretty rare for that to happen. Your advisor is not going to let you defend unless he or she is confident that your dissertation is worthy of a doctorate. And your written work should pretty much speak for itself. The purpose of the defense is twofold: First, it demonstrates that you have researched your dissertation topic thoroughly enough to answer probing questions about it; and, second, it gives your professors the opportunity to make you sweat for a while before shaking your hand and saying, “Congratulations, Doctor So-and-So”, and then taking you out to lunch. They love doing that.
    9. While you’re in grad school, it will dominate your life. So, once you’re done, you may be left with a feeling of emptiness, ennui, or even depression. It’s as if your entire reason for existing has suddenly disappeared, and you’re left without a purpose in life. So it’s important to find something else to monopolize your time, attention, and passion as quickly as possible after graduation. It’s okay to take a short time off to catch your breath and reward yourself for a job well done; but don’t wait too long to throw yourself fully into some other big project.
    10. In spite of everything I have written above, I absolutely loved grad school. Getting my doctorate was the greatest personal accomplishment of my life, and gave me the greatest satisfaction of anything I’ve ever done. If I had to do it all over again, I wouldn’t change a thing. So, if you’re as passionate about wanting to get your doctorate as I was, ignore all of the warnings I just gave you, and don’t hesitate to throw yourself into grad school wholeheartedly. It’s an amazing experience.

    • Greg,

      Just to support your second bullet point, I got pinged on both counts! My marriage fell apart and I got thrown out of the program with 5 months left to go before defending my dissertation! (Not the same as dropping out, but the same net effect). It is tough to teach your kids (High School and Jr. High) to “work hard and you will be rewarded for your efforts” when academic institutions cut the legs out under you so close to the finish line.

    • Nice article. I don’t agree with Greg in the comments above though. It IS possible to have successful relationships while in a PhD program. I was married before I went into the PhD program, and now a few months before I defend, I am still happily married. It is all just a balancing act.
      Grad school isn’t the cause of ones divorce…it just maybe speeds up what would have already happen.

      • This. It can be especially helpful if you’re both PhD students. My husband is a week away from defending. If I hadn’t already gone through the process myself, I’d be tempted to move out of the house right now. As it is, I know how difficult I must have been to be around in the months before my defense so I don’t take it personally.

        I also disagree with the idea that a PhD program and subsequent career in research have to take over your life. I love what I do, but it doesn’t define my existence. If you want to be the very top of your field, it might be a different story and you’ll have to decide on what work/life balance you find acceptable, but that’s true of many careers. It’s possible to do your research, do it well, and still have a healthy marriage, kids, or other outside interests.

  • Great post. There are so many things to say about a PhD. I would have appreciated a particular focus on how to manage the professional relations with other collaborators (especially when they are not so satisfying) and how to manage the relations with supervisors.

    • @Michele — I thought about including a section on working with others, but I think it varies so much by individual that anything I said would be overgeneralizing. You have to take into account your personality and the personality of others’. It’s definitely an important thing to learn though.

      If I were to offer one piece of advice, it would be to keep things positive as much as you can. When you have harsh criticism, it can be good to sandwich it between something good to soften the blow. If someones screws something up, use the experience to learn how to avoid that mistake later on.

      As for relationships with supervisors: Good work goes a long way.

  • “(but don’t be the person who waits until the week before an exam or project to get advice, because that’s just so undergrad).”

    bwahahahahaha I’m going to totally use that with my grad students.

  • If you are considering working in industry, pick a research topic (or two) of industrial relevance, otherwise you may not be able to find a job to apply it and end up doing something totally unrelated, with your education more or less going to waste. Even more demoralizing, people with less education may outperform you because they learned relevant skills on the job. It’s hard to know what skills will be in demand by the time you graduate, but make an effort, and don’t be too surprised if your niche is of no interest to anyone.

    • Not really – that is the fundamental difference between education and training. PhDs should be about learning to learn and that makes you employable unless you have tunnel vision – “I am a world expert in X so I need to get a job doing exactly X and nothing else”. I did a PhD in the chemistry of the rare earth elements and I work in the energy industry – the only stuff I have ever used directly is some basic thermodynamics. One of the best people I ever employed had a PhD in the physics of the processes that make bread rise and he became a top modeller.

      • Nobody teaches you to learn in graduate school; that’s something you are expected to know before you get in, so why spend years and money on a niche unless you are going to work on something related? That’s precious time you could have used to learn things that you actually need to know for your work. If you’re smart enough to get a PhD, you’re smart enough to learn on your own.

      • The difference for me has always been learning what others need me to know versus learning what I need to know.

  • Nice post Nathan! I would stress what you started with, that having a strong interest is crucial for engaging with a research topic for several years.

    When it came to writing up my thesis, I found this advice ‘paper’ surprisingly useful:

    The lightheartedness really helped me overcome obsessive perfectionism and actually finish it.

  • It’s very different in the UK…but one suggestion:
    work flat out on your research WHEN IT’S GOING WELL.
    A lot of people work harder and harder when it’s not.
    That’s the time to take it easy so you can be imaginative.

  • Engelbert Hanzal April 2, 2013 at 7:57 am

    This is really for ANYONE in academia. Thanks a lot!

  • This is really good – advice from the front line. If I had to give one bit of advice it would be that you have to be self-motivated to do a PhD – unless you are part of a big research group that is in effect a PhD factory the reality is going to be it is all down to you, your supervisor may help a bit but really 99%+ of it you have to do by yourself. This is actually one of the best things about it – but you have to keep going when things are going badly and to do this you need to have a passion for what you are studying.
    My second takeaway was that it is never to soon to start to write – the only way you will improve is by doing it so the sooner you start the better. The writing fairy isn’t going to appear I am afraid – you can start by doing the literature review on the fly as you read stuff, it is easy to rearrange it into a coherent story later once you have it straight in your head and it breaks the “activation energy” barrier too. I was fortunate as I had to write a “proto-thesis” quite early in order to stay in the PhD programme which was basically a literature review and a research design exercise (as it took 3 more years before the experiments actually worked – the joys of pushing the envelope, though at times I did wonder what would have happened if they never worked!) and that made a big difference when I came to write the real thing – still took 6 months solid writing though!

  • adwoa amankwah April 9, 2013 at 2:51 am

    The tone is lighthearted and reduces the seriousness around having a PhD which is my immediate goal. I am a journalism lecturer at a private university and would be happy to receive useful flowdata information and self tuition formats in data journalism applications to boost my teaching and writing skills.

  • Thanks for sharing your experience.
    I’ve been considering about applying graduate programs abroad recently…
    at first,this is just one of the goals in my life
    but I got really worried as I start thinking about the problems I might be encountered…such as,the language barrier,the qualification,just like what you wrote.

    Now,I think I know about grad program more.

  • Great post Nathan — I think many of us could write books on this topic.

    I’ll add three things for those seeking doctorate degrees:

    1. Your Ph.D. topic will not make you a Nobel Laureate. Find a topic that you and your advisor can live with AND has clearly defined, time-bounded completion criteria. Ignore either of these at your own Ph.D. peril.

    2. Avoid departmental politics as much as you can. Stay out of space wars, TA battles, or any other “grown folks” business. A Kikuyu (Kenyan) proverb states “when elephants fight, it is the grass that suffers.” Even though you may be brilliant, hardworking, and/or right, the faculty are elephants and you are grass.

    3. Be a good peer. Your fellow Ph.D. students will become research colleagues, grant reviewers, and journal editors. Don’t feed the rumor mill. Apply common sense and and the golden rule in your relationships. I finished in 20004 and still remember the folks that were helpful, even in small ways. I also remember the people who were jerks.

  • Having supervised a few graduate students, I have come to say — only partly joking — that when a Ph.D. candidate starts saying “no” to their advisor, they are ready to graduate.

  • Or maybe ready to call it quits :)

More Tutorials See All →

Transitioning Map, Part 3: Animate Change Over Time

How to make a bunch of maps and string them together to show change.

Moving Past Default R Charts

Customizing your charts doesn’t have to be a time-intensive process. With just a teeny bit more effort, you can get something that fits your needs.

Using Color Scales and Palettes in R

Color can drastically change how a chart reads and what you see in your data, so don’t leave it up to chance with defaults.

How to Make Square Pie Charts in R

Instead of traditional pie charts that rely on angles and arc lengths to show parts of a whole, try this easier-to-read version.