Surviving your comprehensive exams

Quite likely the comprehensive exam (aka qualifying exam) is the most feared moment in an academic’s life. It is on my mind right now because it is comprehensive exam season. I am sitting on two comprehensive exams (hereafter comps) this week, advised a 3rd student how to prepare for comps in June and did several more of both activities the last few weeks. I thought I’d share a few thoughts on surviving. I expect a majority of readers have passed their own, but this post should: a) be helpful to those who haven’t, b) provide a place for survivors to share their advice in one place on the web, and c) help those of us who have students and have to help them navigate it.

I’ve served on 3 faculties in the US and Canada, so I think I have a pretty good idea of the range of variation in North America, but to be honest, I have no idea of how things differ in other continents, so keep that caveat in mind. Comps usually have 3 parts (always in the same order):

  1. Presenting your dissertation research proposal to your committee, answering questions about it, defending and revising until your committee is happy. Some places this is a formal part of the comps and possibly public, some places it is more informal and just your committee. Many places you are supposed to write your research proposal like an NSF grant (15 pages) other places they say 10 pages, but students often go way over these limits. I don’t actually recommend exceeding the limits (but I confess mine was 32 pages).
  2. Written exam. Normally 3-5 of your committee members will come up with a question that you have to answer in writing (on a wordprocessor, not handwritten). Two places I’ve been each question was given 3-7 days, was open book and you were expected to do literature searches, find new papers and synthesize them (thus spanning 3-4 weeks). My current place you have 4-6 hours on each question (3-5 consecutive days), often closed book, and more of a “what do you know” question. There are pros and cons either way. Mostly I am partial to the longer questions since it really tests the ability to read literature and synthesize. On the other hand, students who are facing closed book knowledge questions have been in my experience more motivated to study and often transition better to the oral exam
  3. Oral exam – this is the part that really scares everybody. Just you facing 5 professors asking you questions. Most places I’ve been this is required to last at least 2 hours and no more than 3 hours. Some places have external examiners (outside your committee and department) to keep your committee from being too soft, some don’t.

Usually the written questions will center around topics related to your thesis. Technically in a qualifying exam the oral questions are also centered around your thesis, but in a comprehensive exam questions across your whole field (i.e. all of ecology and evolution) are fair game, hence the name. In practice most are strongly centered around your thesis topic with a few basic broader knowledge questions (if you’re an ecologists know at least the basics of evolution; also know the names and contributions of the 10-20 or so most famous people historically in your field).

There is no way around it – comps are one of the most fear inducing experiences you will ever have. I get an adrenaline rush but enjoy job talks and interviews, I was cool as a cucumber on my wedding day, and I was well prepared and fully expecting to do well on my comps. But it was still numbing. I found myself walking into walls culminating in slicing my finger while cooking and going to the emergency room. After this, I started keeping track and large numbers of people I knew had some major incident in the months leading up to comps (driving over their backpack containing their laptop, minor car accident, etc). It is truly a distracted, even-out of body time.

I don’t say this to scare you if you are in the miraculous 1% that isn’t stressed. But just to normalize it for the vast majority who may find this the most stressful thing in your life.So does everybody else. A great deal of this is self-induced. Take a type-A personality who has gotten good grades all their life and tell them they’re about to take an exam that could flunk them out of their life dream, and well, we put pressure on ourselves. This is not productive. And not even really rational. Your committee has already sunk time in you – they want you to succeed! Pretty much everywhere will give you a 2nd chance if you fail. And ask around to see how many people in your department even failed once (varies widely but usually in the 3%-10% range) and failed twice (i.e. flunked out – usually it is down in the 1%-3% range), and most of these people you could have predicted in advance by grades, prior negative feedback from the adviser, etc. Its not particularly likely you’re going to fail!

There is a piece of this fear that many call hazing. And I won’t deny that there is some piece of this in some departments and some individuals. But for every professor who acts this way, there are three who do everything they can to quell it. We all had to pass comps too! And again its a waste of our time to fail somebody. Less appreciated is that comps are not just a way of “keeping up the standards” but are actually designed for the benefit of the student, believe it or not! I say this for two reasons:

  1. Comps are a chance where you can force a student to learn something. I can’t tell you how many times I’ve had a student where I told them you really should read such and such paper 3 times and they never do. Then comps come around and I tell them to read the paper, and they do. And they thank me for it afterwards. Students who really need to know genetics but stay away, can be forced to learn genetics. And etc. Some part of comps will be forcing you to fill in the holes you’ve avoided filling in.
  2. You will face these situations in the future. Job talks. Postdoc interview talks. Presenting reports if you work in the government, possibly to hostile audiences. Giving testimony to legislatures or even in courts.

The last is really the main point of comps, and indeed the standard by which comps are judged. If you present as somebody who will go on and do a credible job of sounding knowledgeable and defending your ideas in your defense and job talks, you will pass your comps. This is the real goal.

A brief word on format. Since the 2-2.5 hour goal is nearly universal, this means the schedule looks similar most places. The professors will set up a rotation, usually the furthest from your research first, your adviser last. The first professor is told they have about 20 minutes to ask you questions. When they finish you go to the next professor and on through all four or five people. Many but not all places then take a 5 minute break where you leave the room and they confer. Then usually the professors go around as second time for 5-10 minutes each. Five to eight questions would be typical for one professor in the first round and one to three in the second round.

So on to advice I give students (this is all about orals – writtens vary more in expectations from place to place and are not what most people want advice about anyway):

  1. Just do it – students always want to postpone their comps. I don’t let them. I insist on 4th or 5th semester. Once you’ve got the classes and reading under your belt, you have nothing to gain (indeed possibly more time to forget) by waiting. Comps hang over everything and make you less efficient in your research. GET THEM OUT OF THE WAY!
  2. Prepare. This may sound blindingly obvious. But increasingly I am seeing students who haven’t studied adequately. You should study 10-20 hours/week for 1-3 months. And this is after your two years of course work and independent reading beyond coursework (its rather late to start cramming in learning completely new things). I produced a 20 page cheat sheet of everything I thought it was important for me to know. I still refer to it today. If you’ve gotten this far you know how to study – just be sure you do it.
  3. Do the standard tricks. Namely ask each committee member for their recommended reading list. And hold a mock oral comp (and maybe a mock research proposal defense/discussion). Mock comps work best when they are like the real thing. Not 15 grad students shouting out questions. This honestly is a waste of time. Instead, get 5 students who have already passed and are close to graduating, tell each one which of your committee members they are representing (ideally their adviser). And emulate the format (i.e. 2 hours, 20 minutes each person, etc). You’d be surprised how many professors have a question they ask in every comp. Advanced graduate students know what these are and they know what style each professor has. If you let them have fun pretending to be their advisers grilling you, you will have fewer surprises.
  4. You will not know the answer to some questions. That’s OK. Although you are at this point more of an expert than you think, there are still five of us. And it is our job to find out the boundaries of your knowledge, which we can’t do if we don’t ask you a few things beyond your boundaries. I had questions that I didn’t know the answer to. Your adviser did. Everyone of your committee members did. You will too. The real key here is how you handle that. Namely don’t panic yourself into a death spiral just because you didn’t know something. It is totally normal. A majority of failed comps I’ve seen occur when a student gets a question they don’t know early on and they lose confidence and start spiraling down. If you don’t know answers to half the questions and you’re two professors in, then panic. Otherwise, hang in there!
  5. It’s OK to say “I don’t know”. This is a corollary to #4. Part of the comp is making sure you know what you know and what you don’t. Sometimes saying “I don’t know” will be the end (usually for specific factual questions). Other times it will lead to a response of “OK, let’s see if we can think this through” followed by some leading questions to help you. Either way, you’re better off than umming and ahing and making up answers.
  6. Play to your strengths. Spend a little more time on questions you know a lot about. From the schedule described above, you can tell that most questions should have answers in the 30 seconds to 2-4 minutes range. A five minute uninterrupted monologue is getting too long for most questions and ten is way too long. Sounding meandering and unable to know when you’ve answered the question is bad. But be sure that if you get a question you know the answer to really well, don’t give a one word answer! Draw some connections. Expand! Comps may seem eternally long, but they are finite and you have some choice what you fill those two hours with. I am NOT saying you should watch your watch during your oral – your focus should be answering the question. But you might want to ask somebody to time you a bit if you do a mock comp. Some questions will be interactive. A professor will ask you to do a task (e.g. draw a particular diagram on the board), and then another (show how it changes in scenario x), and then so on that build on each other. If you sense you’re in this scenario (being asked to go to the board is a good clue) don’t drag things out because the real question is five steps in and the professor wants to make sure they have time to get there. In all I’ve given kind of mixed messages here – but be aware how long your answers are and think how well that is serving you and whether they are frustrating the person asking you.
  7. Its about attitude not knowledge. Not totally – you definitely still need to study. But attitude is a big part of deciding the outcome. Confident is what you are aiming for. Arrogant is a rare problem (and it is usually a nervous tick or lack of preparation when it happens). Timid and unconfident is a much more common problem. As I’ve said, people want to see you comfortable putting your knowledge out there and standing up for your opinions. Don’t keep checking your questioner’s face to see if you’re getting it right or not. If you’re answering (and not going for #5 above), sound like you know you’re right. It is usually blindingly obvious to your committee when you’re losing your nerve and when you’re digging in and plowing ahead even if you’re on the ropes for a moment. You will have moments of both, but try to have more of the latter.
  8. Get some rest – as I hope I’ve emphasized, there is a significant psychological component to this. Cramming until 2:00AM the night before is the wrong thing. Get your exercise, take care of yourself, go out to a movie the night before and get a good nights sleep. If possible schedule the comp in the morning or afternoon depending on when your body rhythms have you most awake (although increasing scheduling the comp with 5 professors is such a challenge you may have to let this one go).
  9. Have fun! – This might sound impossible. But the comps that go the best are ones where people go in thinking that they’re looking forward to having an intellectual discourse and treat it as a bit of a game. And two of the last three comps I sat in on the student actually did say it was fun afterwards. And not coincidentally, they both did great.

So if you haven’t yet had your comps yet, they’re not as bad as you think, have some fun, and good luck! If you have passed (or have been advising your students how to pass for 30 years), what advice would you add?

About these ads
This entry was posted in Advice by Brian McGill. Bookmark the permalink.

About Brian McGill

I am a macroecologist at the University of Maine. I study how human-caused global change (especially global warming and land cover change) affect communities, biodiversity and our global ecology.

19 thoughts on “Surviving your comprehensive exams

  1. Excellent advice, Brian.

    I would expand on #3 and say, more than just ask for reading lists, ask your committee members what they’re going to ask you! One of my committee members was pretty descriptive: “I’m going to ask you about your study system; know your natural history. And I’m going to ask about your design; know your statistics.” Of course, another just said, “whatever comes to mind,” so results may be hit-or-miss, but it doesn’t hurt to ask.

    Also, on #3, I didn’t bother staging a mock exam — getting several upper class students together at one time can be almost as difficult as scheduling your committee. But do go and ask students in your lab and in the labs of your committee members what the hardest questions were that they got and who asked them.

    I’d also add: bring some food and/or drink for your committee. Not as a bribe, but as a gesture. Two to three hours is a long time for anyone to sit in a meeting, and you want your committee members to be happy. I brought some fresh fruit and some homemade cookies; it started everyone off in a good mood.

    Finally, if you can can possibly take notes, do so. I got some very good advice during my exam and wished I’d had paper to write it down on. I tried to remember what I could the next day, but the exam was intense and I was rather tired by the end, so I’m sure I forgot some things

    • All good advice. The bringing food thing is definitely changing. When I did my comps, a majority of people did not bring food (we debated how it would be perceived). But these days it seems like everybody brings food. Sitting on the other side I can say it doesn’t matter much one way or the other, although as you note, one can never know the subconscious effects of low blood sugar levels (something that is sure to happen even to the questioners after 2 hours).

      I confess I didn’t do a mock comp either. But that was mostly because I was already really comfortable being in the hot seat. I followed your path of just asking people what they got asked. A lot of students swear by it though. And I think if you’re not comfortable being questioned, it is good to practice.

  2. Hi Brian,

    Good advice, though like Margaret I didn’t do a mock exam and indeed I’ve never heard of anyone doing one. Doing a mock exam seems like kind of a big favor to ask of your friends, especially if they’re going to put enough effort into it to do it well.

    Re: written questions, if you’re asked to write some sort of review paper, the examiners often will be looking for it to be more than just a list of “here’s what so-and-so found, here’s what paper such-and-such says…” They’ll be looking for some evidence that you’ve synthesized and thought about what you’ve read (including critical thought). This is why, at Calgary, we give students three weeks to answer the written questions, and we recently cut back the written portion of our candidacy exams from three questions to two (one 20 page, AREES-style review of some broad topic, the other a 10 page TREE-style review of some narrower, more focused topic). We wanted the students to have enough time to put a bit of thought into them.

    A few more small pieces of advice for the oral exam, some of which reinforce things Brian said, some of which are additions:

    As Brian’s post notes, the rules and format vary somewhat from place to place. Know the rules at your university (apparently, we’re softies at Calgary-orals here last a *maximum* of 2 hours!) In particular, at some places the proposal defense and the candidacy or comprehensive exam are the same thing, at other places (the places that do it right, in my view!), they’re completely separate. At places where the proposal defense and candidacy exam are the same thing, the focus will often be much narrower, on topics more directly related to your proposal.

    A favorite professorial trick, after you’ve answered a question, is to respond “Are you sure?” Profs like asking this *whether your answer was correct or not*. As Brian says, candidacy exams are in part a test of your confidence, and there are good reasons for that. Someone who really knows their stuff isn’t going to be easily swayed by any little challenge–and that’s as it should be!

    Some questions will force you to think on your feet, and may not even have a clear-cut correct answer. Other questions check that you have a firm grasp of the basics. One sure way to risk failure is to be shaky on things like definitions of basic concepts.

    Ask post-candidacy students about the sorts of questions the profs on your committee tend to ask. For instance, my advisor, Peter Morin, always asks students to name five famous female ecologists or evolutionary biologists and briefly summarize their contributions to the field.

    If your university has open oral exams (many don’t, but some do), attend someone else’s before you take yours, so that you can see what they’re like.

    At Calgary, if students ask for a brief adjournment to go to the bathroom, it’s routinely granted and it doesn’t raise any eyebrows on the part of the examiners. Not that you shouldn’t go right before the exam, of course.

    If you don’t understand the question, it’s fine to ask for clarification, or to rephrase the question to make sure you understand.

    Yes, there are going to be some questions you don’t know the answer to, because the whole point is to find out the limits of your knowledge. If you don’t know, say so. Although I have to say that one of my proudest moments as a student was when Mike Pace told me at the end of my orals, “You know, the point of these things is to push the student until their knowledge is exhausted, and we never managed to get to that point with you.” #shamelessbragging #multanovitvulpes :-)

    And finally, this old post might help you relax a little. Or not. ;-)

    • Yeah the I know one prof who pulls the “are you sure” line out at least 5 times every exam. It really flusters students. But you should be sure.

  3. Hi Brian,

    I’m wondering if you can expand a little more on this “I produced a 20 page cheat sheet of everything I thought it was important for me to know.”

    Although I am only 1.5 years into my M.S., I am trying to read broadly as a general preparation for being a well-rounded and knowledgeable professor (1 step of which includes my comps). (Caveat: I realize I still have a long way to go, and will need to narrow my reading based on the sub-field of my PhD and my PhD committee). Keeping a cheat sheet seems like an impossible task given the breadth and depth of ecology and evolution. What types of ideas, or topics, or references are on your cheat sheet? Was this based on suggestions from your committee? Any advice for text-books or classic papers to read (my interests are very similar to yours, macroecology, community ecology, mathematical and statistical models for population and community dynamics)?


    • At the time I took my comps, my proposal was based on revisiting the increased diversity lead to increased stability question. So my cheat had all the major population dynamic formulas, all the more recent modifications for spatial/metapopulation/stochastic variations and who authored those. At the time I took comps pushing 15 years ago, I was able to fool myself into thinking I had a pretty comprehensive list, but I think it would be hopeless now.

      I had a committee member who studied evolution of sociality. So I had the 5 or so major milestones there on my cheat sheet. I had a committee member who was big on evolutionary history so I made sure I had a good review of the geologic eras and history of life (as you can see exams at my school were much more on the comprehensive all inclusive side).

      My server is down at the moment because they just renetworked my building but when I get it back up (might be a couple of days), I’ll post a link to my cheat sheet.

      • Hi Brian,

        Thanks for the response. If you ever get the chance to post your cheat sheet, I’d love to see it. Thanks for offering!

        Sorry for the late reply

    • Hi Chris,

      this sounds impossible, but I once did (following the advice of a friend) a 1-page cheat sheet (back and front) on ecology and evolution as it is (was) presented in the textbooks by Begon/Harper/Townsend and Futuyma. That really got me focused and, moreover, the huge effort I had to put into it made the whole sheet obsolete. What was the point of the whole excercise, I guess.
      It’s fair to say that the writing (by hand, incl sketches) was really tiny. Shame I don’t have that paper anymore.

      good luck,

  4. Nice post. I could have used a post like this a year ago.

    One thing I would add – Know your committee members ‘favorite’ organism/pet theory, not necessarily in-depth if it is not your area, but at least some familiarity with it.

    And like Margaret writes, find out what the hardest questions were and practice answering different flavours of the same questions.

    And yup -GET IT OUT OF THE WAY, as early as you can.

  5. I just finished my quals on Monday — this advice is spot on.

    I would add one thing: Be the first in your cohort to take your quals. In my experience as a master’s student & this semester as a phd candidate (!!) committee members appreciate the bold students who step up to the plate early in the year. Get it done!

  6. Lets face it these exams are pointless just like graduate school. I have not learned one new piece of knowledge in the 18 credits I’ve taken so far and have done less work in a whole semester than I did in a month in an undergrad class but I’ll get a pay increase if I finish.

    But those points aside, I am a part-time online master’s student in biology (non-thesis). Last week I was randomly told that I need to take a comprehensive oral exam in the Fall, since they finally finished transferring 12 credits from my old university bumps me into the credit range in which I HAVE to take the exam. During my whole transfer process and looking into the program I was never told about this exam, my advisor didn’t even mention it when I first transferred. Do I really have to study for this exam or can I just wing it? I have good grades graduated with a 3.6 in undergrad and have a 3.8 currently. I work full time, train horses (to make extra money and keep my sanity), and do school and don’t want to add 10-20 hours of studying when I only have about three hours to sleep as it is.

    • Hi Sam,

      Thanks for your comments. The post and the other comments so far are all focused on the experience of students enrolled in full-time, on-campus graduate programs, especially thesis-based programs. That’s why your experience is very different.

      As to whether you have to study for your exam or just wing it, afraid I can’t offer any advice. I actually had no idea that comprehensive exams were ever required for part-time online MSc students, or that the requirement would have something to do with how many transfer credits you have. Perhaps there’s another reader in a program similar to yours who can offer some advice.

      • Thanks Jeremy, for the reply. I was worried no one would have any idea want a non-thesis comp exam for an MS would contain, since I’ve never heard of this being a requirement and I have friends going for their MS all across the U.S.

  7. What advice could you give a Masters student taking her comps in a month and having panic attacks in a daily basis. I’m taking my comps for the Masters level (which are constructed differently from that of the Doctoral level), and what scares me to be franc is remembering the names of the authors and their books. I can remember the conferences themselves (I’m in IR) very well but my sources? Now that’s another issue! Our MA comps are for one day (6 hours) and closed book, i.e. no notes, no books, no nothing. What advice can you give this panic-stricken Masters student?

    • Hi Phoenix – just remember your committee wants you to pass and doesn’t have unreasonable expectations. If you were smarted enough to get admitted and have done your studying you should do fine. If you feel like sources don’t come easily to you then trying making lists of the sources you need to remember – author – year – title and just spend time committing them to memory. I doubt you will have to be able to remember ever source in a closed book exam. But if you make a list of the top 20 and then nail those and then do the best you can on adding in another 20 you’re probably good (it sounds like we’re in a different field so the numbers might be too high for your field).

  8. This is the second time I’ve come to your blog, thoroughly enjoyed a blog post, and then realized that you gave a seminar at my school (UVM) recently! Thanks for the excellent advice.

    – A Gotelli Lab grad student

  9. Pingback: Don’t worry (too much) about whether you’ll get tenure, because you probably will | Dynamic Ecology

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s