CoC Graduate Student Council

Surviving the Qualifying Exam at the College of Computing

Last Modified: January 5, 2004

Table of Contents

Introduction
The Basic Process
Preparing for the Qualifiers
Taking the Written Exam
Taking for the Oral Exam
Resources
Author Information

1. Introduction
Getting a PhD has been compared to applying for membership to a very picky club that has a lot of hurdles for the applicants. The qualifying exam is the first of two very strenuous hoops that you must jump through to be recognized for PhD candidacy (Georgia Tech doesn't officially recognize you as a candidate until you pass your Proposal). Many students have been daunted by the prospect of the exam and with good reason. You shouldn't fear the process but should view it as a golden opportunity to become an expert in your particular area of knowledge. Once you graduate to either industry or academia, the demands of the work often hinders your ability to keep up with the field of research. The qualifying exam not only affords you the opportunity to really learn your field but to gain some valuable knowledge during the process that may contribute to your dissertation. The process (the stress, the pressure, and the intensity) of the written and the orals may lead you down some intellectual paths that you haven't yet explored.

The qualifiers are not designed to be easy (and if they were, no one would take them seriously). The faculty in each area design the exams to test basic knowledge of the area, creativity in thought, and thoroughness in problem-solving. Remember this: The ultimate purpose of the qualifiers is to assess your ability to do research. They are not testing your ability to regurgitate what you've read or how well you can crunch numbers in your head. Your advisor and committee members want to know whether you have the right skills and mental tools to be able to take a difficult problem, attack it correctly, and produce a range of solution directions. Whether you pass is contingent on your ability to convince them of this through your answers on the written exam, your research portfolio, and how you answer their questions during the oral exam.

The following document contains some tips for surviving the qualifier process with a minimal amount of stress and sleep loss. It explains what you need to do at each stage of the qualifier and one method for attacking it. Don't forget to check with the senior students in your area for more information as each area has their own quirks and peculiarities about how qualifiers are taken. There is also an older version of this document which you can check for additional information and perspective: http://www.cc.gatech.edu/gsc/old/PSH/#chapter9 (courtesy of former GT CoC grad student and current GT CoC Professor Tucker Balch).

2. The Basic Process

When should I take Qualifiers? - You should probably think about taking qualifiers towards the end of your 2nd year or no later than the end of your 3rd year. This is a general guideline. It really depends on your focus, your knowledge of the area, and most important, whether your advisor thinks you should take qualifiers or not.

When are they offered? - Qualifiers are typically offered twice per year. Currently, they are offered Fall and Spring semesters.

When do they take place? - The written exams are scheduled depending on when your area committee decides they should and can be held. Typically, they will be scheduled towards the end of a semester. The oral exam should be scheduled by you about a week to a week and a half after the written. This gives your committee and the exam writers a chance to look over the exam and your research portfolio.

What's involved?- The requirements may change from year to year so you should definitely check the Official College of Computing Qualifier Requirements. The requirements stated as of December 11, 2002 are as follows.
1. A one-day written examination covering the pertinent research area(s). It will typically be near the tenth week of the semester. All students taking the exam at the same time in an area take the same exam. This component evaluates the student's depth of knowledge in his or her chosen areas.
2. The submission of a high-quality research deliverable, as evidenced by a portfolio consisting of an exam-committee-reviewed and publishable article, and possibly other work products as approved by the exam committee. This component evaluates the student's creative and research abilities, along with the potential to do dissertation research.
3. An oral presentation and examination which also allows for follow-up on the first two parts.
What happens if I don't pass? - This happens every now and then. Usually students are given conditional passes that require them to write a publishable research paper because their portfolio lacked sufficient depth. Once this deficiency is made up, the student has passed the qualifier. Students can fail because they demonstrated a deficiency of knowledge or lack of clear thinking during the written or during the oral exam. In rare instances, students have failed because of a basic incompatibility with their advisor that was not revealed until the qualifier. If you don't pass the first time for whatever reason, you have a second chance. If you don't pass the second time, then your committee will recommend that you finish a Master's degree and try your luck in industry.

3. Preparing for the Qualifiers

Registering

Determine your area - Hopefully this won't be a problem by the time the qualifiers are an issue. However, some research areas are ambiguous or not established and you may feel the need to declare a secondary area. If you do, you will have to read two sets of reading lists. For the written, you will be given two exams and can choose questions from both. You may answer up to 50% of the questions from your secondary area.
Declare intent - Once you and your advisor have decided that it's time to take qualifiers, you need to email The PhD Program Coordinator, The Graduate Coordinator, and possibly the Area Coordinator. As of January 5, 2004, the PhD Coordinator is Barbara Binder (binder@cc) and the Graduate Coordinator is Dr. Leo Mark (leomark@cc). Do this at the beginning of the term that you will take qualifiers.
Attend the Orientation Meeting - There will be an orientation meeting at the beginning of the semester for all students who intend to take qualifiers. It will be posted to the phd-list mailing list so pay attention. It is very important that you attend this meeting! Even if you're undecided whether you're going to take qualifiers that semester, you should still attend. In this meeting, you will agree with other students in your area on the date you want to take the written exam.
Register to study for the qualifiers - Students may register for one of two Qualifier prep courses during the term they plan to take the Qualifier. Both of these courses are permit only; contact the PhD Program Coordinator for a permit, and make sure to indicate which class you will be taking.
- CS 7999PHD - Register for between 3-6 audit hours. Note that these hours may not count toward the 12 required for your assistantship.
- CS 8902PHD - Register for between 3-6 pass/fail hours.

Planning

Start Early!! - If you haven't started reading from your area's reading list from Day 1 of graduate school, you should start as early as the term before the term of the qualifying exam.
Make sure your area advisors update and finalize the reading list by the beginning of the semester. - Reading lists can fall out of date really fast and you want to ensure that you're spending your time well. Try and get your area coordinator to update the reading list quickly. If you're really organized, you did this the semester before!! With any luck, the faculty have finally responded to your request. You should follow up on this constantly until the area advisor has formally confirmed that the reading list is finalized. The last thing you want is to be surprised in the 11th hour with a new set of readings.
Find a study group - Try and coordinate with other students in your area so that you're all taking the exam at the same time. There's nothing like group studying when it comes to sharing ideas, finding readings, and reinforcing morale.
Develop a plan of study - Don't concentrate solely on getting all the readings out of the way like a checklist of chores. This isn't an exam for a class, this is a qualifier. You need to be able to speak intelligently about your area, not just regurgitate names, dates, and facts. Develop a schedule for presenting papers, emphasizing or de-emphasizing topics as needed, depending on your strengths and weaknesses in the area. Make sure to allow for schedule snafus and to determine your studying resources, including library and faculty access.
Check the reading list - The reading lists for all the areas can be found on the CoC body of knowledge section.
Gather the readings - If your fellow students haven't already done so, designate a filing cabinet or some sort of library space and begin storing readings there. The Software Engineering Group has one such library that has proved to be extremely valuable. During one qualifier, someone had misplaced a critical reading and was able to go to the SE library and get the reading that he needed. It not only provides a central location for everyone to store and collect readings, it also saves future generations the hard and pointless labor of having to find these readings the hard way. Also, some of the seminal papers can be extremely difficult to find, if not impossible.
Gather the past exams - If you haven't noticed, the past exams are also located with the Body of Knowledge readings. Try and get in touch with students who have taken the exam before to get a sense of their answers. This is actually more useful right before the qualifiers than it will be when you're first starting.
Talk to old students - You're not the first person to ever take this exam. In fact, there's probably at least one student who has taken the exam within the last year. They will have in-depth advice about preparing for and taking the exam from the student perspective. If you are lucky, these students will also recommend additional papers to read that are not on the reading list.
Start developing your committee - The committee consists of your advisor and 3-4 other faculty members in your area. You're not required to but it helps to have a committee with people that may be at your proposal and eventual research defense. This way you begin developing a committee that will be familiar with your work by the time you're ready to defend your dissertation.
Preparing
1. Study, Study, Study - As someone pointed out, there's only one proven method for passing the qualifier - hard work. Be prepared to do lots and lots of studying. By the end, you should be dreaming about the exam (these are typically unpleasant but make for fun storytelling).
2. Build in time for reflection - Don't forget to schedule time to reflect on what you've learned. Remember that this isn't a memorization game. You need to understand the readings at a deep level. This means that you should be able to construct an example given a model, algorithm, or method that you've read and understand how the authors arrived at their findings. Your understanding of the research process used to generate those results is, in many ways, more important than the knowledge of the results themselves.
3. Ask the faculty in the area about the readings - The faculty assemble the reading lists and devise the exam questions. They have made some very specific decisions about what to include. It would be a good idea to meet with each of them and ask them what they wanted you to learn from the readings and what kind of questions do they expect you to be able to answer.
4. Hold weekly group meetings - If you're meeting in a group, split the reading list. Organize it around the study plan that you have developed. This allows a little bit of breathing room and also encourages some good discussions with one person as the primary "expert" in that area presenting and the rest contributing questions, contention, or answers.
5. Read outside of the reading list - Scour the journals, books, and associated research in your area. The reading list should be considered a bare minimum for knowing your field. Chances are, it's not as current as it needs to be and you need to fix that. But be picky! Your advisors will occasionally complain about the volume of reading that they have to do. Your problem won't be finding readings to add to the list but keeping the number at a manageable level. Make sure that you select good papers to present and share with your group.
6. Aim for frameworks and patterns - Remember that as PhD students, you will be expected to make a contribution to your field. By the end of the qualifiers, you should be able to articulate a detailed picture of your field and your opinion of where you think it's going. While you're doing the readings, ask yourself 'why', 'how', and 'what' questions:
  - Why is this reading important?
  - What did it contribute to the area?
  - How has it influenced other research (if it's a seminal paper)?
  - Why is research being done in this area?
  - In what directions is the field headed? Where will it be in a year, five years, or a decade?
7. Be skeptical - As a PhD student, you can't afford to be passive about your study and learning. You need to be critical about the papers that you read. Just because someone put it on the reading list doesn't mean that it's the best work in that area. Very few papers present a brilliant idea, support it in a solid manner, and describe these two things clearly. Most good papers have at least two out of three of those qualities. Ask yourselves:
  - Do you believe what you are reading?
  - Do you know who wrote the paper and where they are coming from?
  - Did they support it well?
  - How did the paper extend the knowledge of the field, or did it?
  - Has there been a more recent development in this area?

4. Taking the Written Exam

Basic Questions
1. How long is the written exam? - The written exam is typically scheduled from 9 am to 5 pm.
2. How do I get the test? - The PhD Program Coordinator is responsible for handing out the exam. The current coordinator is Barbara Binder, and she is located in Room 122 of the CCB. The test should include an instruction sheet for reurning the exam.
3. Where do I take the test? - You can take the test anywhere you want. It's open note and open book. You should use some kind of word processor to take and print the exam. This not only makes it easier on the graders but gives you better throughput for answering the questions. Equations and diagrams can be hand drawn if needed.
4. What is the format of the exam? - You will probably be given N questions and be required to answer some fraction of those questions (e.g. 6 out of 9, 8 out of 12).
5. What about diagrams, references, and tables? - You may need to write equations or cite papers during your exam to support your answer. For this reason, make sure you're using either a good word processor or a reliable text editor.
6. I can't type. Now what? - If you have a disability such as Repetitive Stress Injury or something similar that requires you to write the exam by hand, you should make arrangements to either get some extra time or, preferably, to split the test across two days. Eight hours of straight handwriting is not very entertaining and may even exacerbate your injury. Your committee will generally be sympathetic to these kinds of problems.
Before the Written Exam
1. Organize your work area - You'll want a quiet place with the computer that you will be writing the test on, all your readings and books, and with the basic necessities nearby (food, water, restrooms). Because of the immense volume of readings that most of you will have, you'll often need to work out of your office/cubicle for practical reasons. It's okay to have music (with headphones) at the test if that's what makes you happy. Some students have even made arrangements to have the test emailed to them at home with on-line submission at 5 pm.
2. Plan Ahead!! - For example, all tests are picked up from the CCB. If you are in a distant building, you can ask for additional time to travel between your home building and the CCB. Typical allowances are 15 minutes (allowing a start of 8:45 and an end time of 5:15).
3. Organize your computer - You can bring in your home computer to your office for the purposes of taking the exam. This leaves you free of potential system hang-ups and access problems.
4. Organize your readings - I organized and alphabetized my readings and papers by area groupings. This proved to be incredibly helpful for finding something in a hurry. Avoid the stack method of organizing readings.
5. Type in your reference list before the exam - If you are already using some kind of bibliography tool like EndNote, then you're all set. Otherwise, you should copy the reading list and add all the supplementary readings to a file using whatever word processor you will be using. Because the questions tend to be organized around areas, I recommend that you organize the reading list the same way. This way, during the exam, you can simply copy and paste the references you cite in your answers into your exam, rather than having to type it in on the fly.
6. Build Templates - Prior to the exam, I developed blank test templates in my word processor that had page numbers, questions, styles, and fonts. This is a minor time saver but it was very helpful for getting started on the day of the qualifier. The Software Engineering Group had also developed templates for typing in equations and had some PowerPoint slides from Sommerville with diagrams showing various software engineering processes that could be copied into the exam. Basically, make any preparations you can before the exam to make your life easier. Also, get familiar with your word processor (for example, shortcut keys) and if you use LaTeX, have a LaTeX manual handy.
7. Place signs up around the area that you're working - Before my qualifiers, I posted 'Do Not Disturb' signs all around the College. You may want to coordinate with the other people taking the exam to do something similar.
8. Plan your printing process - Some students bring their computers into the test - printers and all. Consider where you will be printing out your document. Make sure you have a backup floppy disk or something that will allow you to transfer the test to another computer in case something bad happens.
9. Make sure you have food nearby - You won't have time to get food. Make sure you're well supplied before you start or that you have friends willing to supply you with snacks and stuff.
10. Sleep - There's nothing else you can do at this point except get a good night's sleep.
During the Exam
1. Use good test-taking techniques - Read all the questions before you begin, answer the easy ones first, etc.
2. Don't panic - The questions may look intimidating but they're supposed to be tractable. If you find yourself in a bind and are forced to answer a question you're not comfortable with, recast the question into something that you understand but without deviating from the original statement. "I believe that the question is saying X. The answer to X is..."
3. Watch your time - The average test is 6 questions over about 8 hours. This leaves roughly 10-15 minutes to sketch out an outline and 1 hour to write the question. Leave time for proofreading.
4. What if I turn the test in late? - You need to make one copy of your answers and hand them in to the Graduate Coordinator (currently Dr. H. Venkateswaran). Some allowance is made for copier and printer snafus. In general, be prepared to stop writing around 5 pm and concentrate on getting your test and copies turned in. At this level, they won't say something like "You're 15 minutes late, we're not accepting the test. Better luck next time."
After the Exam
1. Go decompress somewhere - You can't do anything about the test now. Go relax and party with your fellow victims.
2. Wait - The faculty grade the questions that they designed (assuming anyone answered them). This takes about a week.

5. Taking the Oral Exam

Basic Questions

When should I take my Orals? - Oral exams should take place a week or more after the written exam. My advice is to get it out of the way as soon as possible so that you can get on with your life. Also, the sooner you give your orals, the fresher the knowledge from the written.
How long are they?- Again, this varies by area and committee but you should schedule out a 2 hour block of time, just in case.
What do they involve? - You will have to give a 15-20 minute presentation on the area of research that you've been working on and where your thesis might be headed. The committee will spend the rest of the time asking you questions about your written exam, your research area, and anything else they feel like.
How can I study for them? - You can't really study for your orals except to think a lot about your research and where you intend to go. You can meet with the other people who took the written with you and go over your answers. If you made any errors on the written, the committee will want you to address them. Otherwise, the orals are designed to test your creativity and your grasp of the field. In your career you may be asked to respond to questions or issues that you haven't thought of or have thought of but have never formally addressed. You'll need to show that you can be thoughtful, creative, and confidently articulate about your area of research.

Preparation

Schedule your oral exams. - You'll need to find a time that your entire committee can meet. Schedule your exam at least 2-3 weeks in advance!! Scheduling 4-5 faculty members to meet for two hours is like herding cats so you'll need to have scheduled this exam before you've even taken your written exam. Don't forget to schedule a room.
Distribute your portfolio- Your portfolio, assuming that you haven't been given some kind of creative question to answer in lieu of a publication portfolio, should consist on any conference papers, journal articles, and technical reports that you have published either alone or with your advisor or research group. Make copies of these and give them to your committee. If you don't have a portfolio, you may be asked to spend a week answering some kind of creative question that concerns your current research directions.
Review the written exam - Sure, you never want to see it again but you should look it over anyways. Make sure you can answer the questions that you avoided or messed up. You have a week to figure these out. It may not come up during the oral but you never know...
Prepare your talk - You will be asked to give a 10-15 minute talk that discusses your research history and your directions. You're not required to have research that's at the proposal stage but you should think of this talk as an articulation of a pre-pre-proposal. Remember to use good design heuristics in planning both the verbal and visual presentations of your talk.
Practice your talk - Get some of the friendly senior students in and around your area to listen to your talk. They will hopefully be very critical and this will be in your favor.
Remind your committee - Remind them what time and where the oral will take place a week before. This may seem silly but I did have the situation where the week before the orals, one faculty member had written down the wrong time and was starting to tell other faculty that time, generating lots of confusion. This was remedied with some frantic scampering on my part. (It had taken me three days of schedule consulting and negotiating to get everyone to agree on a time slot. I wasn't about to have to repeat the process.)
Print out slides- You should not only print out slides for your talk but backup copies to hand out to the committee members. If you're not doing a multimedia presentation, make sure that your committee members have an abstract and outline for your presentation.
Get food - Before your orals, get snacks and drinks for your committee members. This is not bribery, it's part of the culture. It also saves you time. (I didn't know about this for my oral and I had to wait an extra 5-10 minutes while various members went to get coffee, soft drinks, or whatever). Doughnuts, bagels, cookies, milk, soft drinks, etc. are nice. Remember that this is a two hour oral examination so don't forget something for your own throat.
Arrange the room- Make sure you have working AV equipment and that the room is set up to your satisfaction.

Taking the Orals

What will they ask? - It depends on the committee but some good questions that every graduate student should be prepared to answer include:
- What will your contribution be to the field?
- What was the best research paper, researcher, or research portfolio that you've seen and hope to emulate?
- Where do you think your academic area is going?
- How will you validate your research or findings?
- What do you consider to be good research?
- What do you think your weaknesses as a researcher are? What about your strengths?
Take notes - Your committee will ask you long and difficult questions. Make sure you understand the question completely and take notes on what they asked. I found myself forgetting the original question halfway through my answer. These notes will also be research directions that you will need to examine after the orals.
Whatever you do.... - Try to avoid saying that you absolutely don't know something. Ideally, they wouldn't be asking you unless they were sure that you could answer it. If you're not sure of the question, get clarification or get them to restate the problem. If you need to, you should recast the problem yourself and answer it that way. At the very worst, consider the relevance of the question to your research and your area. It may be an issue that you haven't considered yet and will pursue after the qualifiers and you should say it. Definitely make a note of the questions that you couldn't answer because they will haunt you later in your career.
Be confident! - You have to believe that you know your area. Granted, you don't have decades of working in the field but you know what you've been working on. If you don't believe your own work, who will? Project confidence and ability and you'll be fine.
Have questions for your committee - Did you think the test was fair? Why was this question asked? Did you have problems with the reading list? How about the courses that you had to take? Do you want to know what they did in your position?

After the Orals

Standard Operating ProcedureIn theory, here's what happens: you leave the room, your advisor and your committee discuss some things, your advisor leaves the room, your committee votes, your advisor finds you and tells you the results. If you've passed, you'll need to thank everyone and get their signatures on this piece of paper and turn it back in to the PhD Program Coordinator.
Leave a Record for those to follow - What extra readings were a big help? Which readings were useless or outdated? Help the faculty and future students out by making suggestions to the reading list. Make sure that the library for your area readings is brought up to date.
Help to Update this Document - If there are things that you learned during the process of preparing for and taking the qualifiers or there have been changes to the procedures, please help future generations of PhD students by updating this page.

6. Resources

7. Author Information
This page is maintained by the Graduate Student Council as a whole, and monitored by the PhD program Coordinator. The GSC acknowledges Idris Hsi as the original creator of the document, and thank all of those who have contributed to it's evolving nature.

If you have any comments, suggestions, or questions concerning this document, please contact the current GSC Liaison. For more information on GSC, please visit our website: http://www.cc.gatech.edu/gsc.