Current students‎ > ‎

The PhD Track

Making the Transition from Coursework to Research

Most graduate students will have spent nearly 12 years of their lives taking classes and trying to earn good grades by the time they arrive in the Ph.D. program.  These 12 years of training can reinforce a mindset that is not necessarily the best for conducting research.  Courses rarely leave charted territory, and leave us unprepared for the uncertainty and unknown waters of research.  To contrast the two, consider the following aspects of coursework and research.

Coursework:

  • The beginning and end are well-defined.
  • The path from beginning to end is well-defined.
  • Everything that is needed to get from beginning to end is outlined for us in a syllabus.
  • Homework and exam problems are chosen to be doable within a fixed amount of time.
  • Homework and exam problems can be solved, and this is expected.

Research:

  • Research problems are rarely well-defined.
  • It will inevitably require techniques you don't know or have forgotten.
  • Your particular problem might be unsolvable.
  • There may be no time limit, affecting motivation.

Getting started

For some people it's as simple as going to a faculty member and asking them if they have any interesting problems to work on. The initial effort will probably consist of several weeks of reading the literature. At this early stage there's no commitment by either party to turn this into a student-advisor relationship.

To find out what your professors actually do, click on the research profiles on the faculty page on the department site. It's also a good idea to talk to their current students, or the professors themselves!

Choosing an advisor

Each student is responsible for selecting a research advisor to work with.  Students usually choose an advisor during their first or second year.  The student should try to find an advisor that is a good match in terms of personality, work style, interests, and expertise  

The role of the advisor

In general, the role of the advisor is to guide the student in research and career development.  Advisors often provide the student with a research project to work on, either by suggesting an idea for the student to pursue, or by bringing the student into a collaboration with other people.  In other cases, the research project might originate from an idea that the student has.  

Some advisors are very hands off, meaning that they only provide some general suggestions for the student to consider.  Other professors are somewhat authoritative, in that they provide directions and expect the student to mostly follow these directions.  Other professors work closely with the student and actually do some of the work.  

When you are close to graduating, the professor will have to sign your thesis and write recommendation letters for you.  

Finding your interests

When you begin your search for an advisor, it really does help to know what your interests are.  If you are not sure what your interests are, then you should go to seminars, talk to people, take classes, read books, read articles, and soul search.  

Looking around

Learn about a prospective advisor's research.  Try to learn about the field and read some of the professors publications.  Make an appointment with the professor to discuss his or her work.  Find out what projects the professor is currently working on and what projects the professor is planning to pursue  Tell the professor about your skills and interests, and ask if he or she would be willing let you join the group.  Thank the professor for his or her time.  
Choosing an advisor is an important decision, so you should probably meet with many professors before you make a decision.  

Making a choice

Here are some things you should consider when making a choice:
  • Talk to all of the students in the professors group.  Do they like being in the group?
  • Are the students in the group productive?  Are they getting papers out?  Do they graduate in a reasonable amount of time?
  • What do the students from the group do when they graduate?
  • Is the professor a good match for you in terms of personality, work style, and expectations?
  • Is the professor famous?
  • Do the research projects interest you?

Working with two advisors

Some students end up working with two advisors.  Sometimes the other advisor is from an outside department.  
A student might choose two advisors if the student wants to do interdisciplinary work or be part of some collaboration.

Once you have an advisor, it's time to prepare for your qualifying exam.

Designated Emphasis

There are currently two options for Designated Emphasis (DE) for the Ph.D. program in statistics.  One needs to file the application for DE before he/she advances to candidacy, namely before the qualifying exam. 

The requirements are listed in the following pages.  Besides a certain number of courses (most of which are cross-listed with the Statistics Department anyways), a recommendation letter from your advisor and a personal statement are usually needed.

Preparing for the qualifying exam

Background

The oral qualifying examination is meant to determine whether the student is ready to enter the research phase of graduate studies. Students usually take the qualifying exam during the spring semester of their second year, or sometime after that.  It consists of a 50-minute lecture by the student on a topic selected jointly by the student and the thesis advisor.  The topic usually involves the student's research, especially some part of the student's research that seems to be going somewhere.  The examination committee consists of four faculty members approved by the Graduate Division, three from the Statistics Department and one from outside the Department (the outside member cannot include faculty from other Universities).  The student is responsible for selecting the committee and the date and time of the exam.  The student's thesis advisor cannot chair the committee.

Paperwork and committee selection

The student must complete an application for the qualifying exam at least three weeks prior to the exam.   The student must provide a copy of the application to the Student Affairs Officer and to the Graduate Degrees Office.  If the student is in a Designated Emphasis (DE) program then there may be some additional paperwork.  

Note that the paperwork must be turned in ''3 weeks in advance'' of your qualifying exam date.  The time requirement may be greater if you are also petitioning to be admitted to a Designated Emphasis.

You need 2 inside members, 1 outside member, and your adviser on your qualifying committee (see http://grad.berkeley.edu/policies/faq.shtml#4).  The application itself has more slots but don't worry about that.  One person who is NOT your adviser should be the chair of the committee (note that if you have 2 advisers, neither of them can be the chair, and the outside member is rare the chair).   

A student will usually select a committee by considering:
  • Which professors do I like?
  • Which professors like me?
  • Which professors does my advisor like?
  • Which professors are knowledgeable about my field?
  • Which professors can give me meaningful feedback on my research?
Note that finding a time when 5 busy people (including yourself) can meet is difficult sometimes.

Preparation

The student is responsible for reserving a room and a projector.  Although the exam is supposed to be 50 minutes, it might last as long as 2 hours if the committee has a lot of questions.  This is something to consider when scheduling the exam and reserving facilities.  The student should prepare 50 minutes worth of material and possibly some extra material to help with questions.  A student will usually practice the presentation in front of his or her thesis advisor, friends, colleagues, stuffed animals, etc (practices in front of students can be arranged by the Student Seminar Committee).  Moreover, always practice with the same projector you will use for your qualifying exam because some colors do not turn out as clearly on the projector relative to your screen.

On the day of the exam

Arrive early to set up the projector.  Some students bring food to the exam (e.g. cake, fruit, and/or nuts, be sensitive to dietary restrictions if you decide to bring food) to feed the committee.  
Some committees prefer that your presentation stay within 50 minutes, including questions.  If so, you will probably have to drop some slides.  After the exam, the committee will tell you to leave the room and wait outside.  The committee will then make their decision.

After the exam

If the student passes the exam, he or she can then officially advance to candidacy for the Ph.D. by completing the appropriate form and providing copies to the Student Affairs Officer and to the Graduate Degrees Office.  A $90 processing fee will have to be paid from your own pocket using a personal check and you will have to declare your final dissertation committee.  The committee members are usually the same as your qualifying committee but may be different.  You can also change your dissertation committee members in the future.

If the student fails the exam, which in the Statistics Department is very rare, the committee may vote to allow a second attempt. Regulations of the Graduate Division permit at most two attempts to pass the oral qualifying exam.

After your quals, what could possibly be left to do but to file your dissertation?

Filing your dissertation

Dissertation formatting instructions are annoying.  At a time when there is so much else to think about, pagination and margins and line spacing and font sizes and page layouts and ... are the last things that you want on your mind.  The Berkeley grad division has a page with helpful guidelines.

Some General Tips

  • Use latex
  • There are several .sty files available for formatting theses. As of 2011, [http://math.berkeley.edu/~vojta/tex/ucbthesis-phd.html ucbthesis.sty] is better than ucthesis.sty.
    • If you have a '''designated emphasis''', you must make some changes to ucbthesis.sty to include mention of the DE in your title page and abstract..
    • With ucbthesis.sty,  \approvalpage  creates a signature page.  This should not be in the .pdf that you submit online.... make sure to remove \approvalpage.
  • For various reasons, it is often helpful to look at past dissertations on similar topics or with similar advisors.  You can find the title of dissertations from Berkeley Statistics alumni here [http://www.stat.berkeley.edu/people/alumni]. The Berkeley library has electronic copies of dissertations submitted more than 2 years ago at http://oskicat.berkeley.edu/.
  • Before submitting your dissertation, go through the thesis.log file and search for "warning"  (of course, if you named you thesis.tex file  xyz.tex, you should go through the xyz.log file).  The warnings in the log file will show you things like missing references that would appear as ?.  Also, it will reveal if you have defined two things with the same \label.  For example, if you label the conclusion in Chapter 2 and Chapter 3 \label{conclusion}, then \ref{conclusions} will do funny things.  Catch these problems (and others!) by going through the log file.