There is a serious housing shortage in Berkeley as a result of excessive demand and local government restrictions which reduce economic incentives to build housing (such as rent control and restrictions on development). Parking spaces are also in very short supply. (If you arrive on campus before 7:30am, you might get a space.) Try to find housing that is accessible to campus by public transportation, bicycle or walking.
Also known as I-House. Provides dormitory-style housing specifically for graduate students. Foreign students comprise 50% of the residents of I-House, and U.S. citizens make up the remainder. I-House is conveniently located within easy walking distance of Evans Hall, location of the Department. Several statistics graduate students currently reside there.
One-person and two-person rooms are available. You must make your request early to get a private room. Some students find that it is too expensive to stay in I-House over the long-term, but that it is very convenient to stay there for your first year in Berkeley. Sometimes there is a waiting list, usually for Fall semester. Foreign students note: if you apply early (say in April), you will probably get in.
Individuals interested in living at I-House should, at the earliest opportunity, request information from the Residence Office, International House, 2299 Piedmont Avenue, Berkeley CA 94720 U.S.A. The phone number is 510-642-9490.
Apartments are located in the cities of Albany and Berkeley, and residents must commute to the main campus. Rents in these pet-free apartments are generally lower than prevailing market rates. If you qualify and are interested, immediately contact Housing and Dining Services, University of California, Berkeley, 2401 Bowditch Street, Berkeley CA 94720 U.S.A. The phone number is 510-642-4109. Of course, there is a waiting list. Many statistics graduate students reside there.
This is where one person rents an apartment or house, then seeks roommates to spread the cost among additional persons. Individuals interested in this type of housing will find a large variety from which to choose. However, there is also considerable competition for this type of housing in the fall semester. Listings of this and other types of available housing can be consulted in the Community Living Office, 2405 Bowditch Street; phone 510-642-3642. You must present a valid university ID card or a letter of admission to gain entry to the housing listings.
This is the only category of housing in Berkeley for which there is a somewhat reasonable supply. However, you will need patience and persistence to find what you are looking for. Be careful in your choice of roommates.
It is rather difficult to find an apartment or house to rent in Berkeley. This is an unfortunate situation, but it is a fact of life here.
If you want an apartment in the city of Berkeley, you will need patience and inside connections: friends who are moving out and who will release their apartment to you. New students generally lack the inside connections, hence are less successful in their searches for an apartment in Berkeley. According to a survey of apartment residents in Berkeley, two-thirds of the people living near the University found their rental units through personal connections and word-of-mouth. Here are some suggestions for those who are searching for an apartment or house:
If you choose university housing, make sure you get your application materials and submit them by the deadlines. But what if you'd prefer to live out in the community? When should you look for a place to live? The word on the street is: the sooner the better. As indicated above, there is limited housing available, and as the summer wears on, more people will be competing for less space. How long should you give yourself? That's a hard call; folk wisdom suggests about two to three weeks, but it could take more or less time than that. As for expenses and conveniences, they depend on your choices: shared unit or by yourself? close to campus? bay view? ritzy part of town? The list goes on.
There are several places to begin your search, whether you want shared housing or vacant rentals.
Once you have moved in, you will want to orient yourself to your program. If you have never studied at the graduate level, the transition can be somewhat stressful. Here we offer some advice on hitting the ground running.
Regardless of your program, you will probably need to use the computing facilities extensively, including data analysis, simulations, and typesetting reports. This subsection gives some pointers on starting out; for more details on facilities, see section 5.2. Individuals with no computer experience and those with extensive experience in other environments will need to adjust to the available systems.
The SCF computers use the UNIX operating system. Most statistical computing is done using S, C, and Fortran; BMDP, GLIM, BLSS, and SAS are used on occasion. Biostatistics students have additional access to IBM 486's, which support SPSS and STATA.
Students in both programs may open personal accounts with the SCF; see James Blakly in 407 Evans. James also has a supply of introductory SUN UNIX documentation for sale. He has other manuals available for short-term loan. When you open your account, ask to be placed on the ``grads'' mailing list, and check your email frequently. Since announcements of seminars and social events are often made through electronic mail, these accounts are important even for biostatistics students who may not otherwise be closely affiliated with the Statistics Department.
In addition, accounts will be provided for classes in the Statistics Department which require computing.
A number of courses will require extensive computer use. You might familiarize yourself with the system by diving right in to one of these courses, or by familiarizing yourself with the computer system in the first weeks of the semester, or even before the semester starts. If you do this, you will avoid trying to learn the computer at the same time you are struggling to complete your lab assignments. The course Stat 243 is designed to introduce students to statistical computing and the department computing facilities, and many students find it a good way to develop computing skills.
First-year Ph.D. students generally take two of Statistics 205, 210, and 215. All of these courses are strenuous, with lengthy homeworks and readings. It's enough to take any two of these courses first year (although all three must be taken eventually). Most first-year students find that taking two of the above courses is a full-time job. Thus you will need to manage your time well. Also keep in mind that your thesis topic does not need to come from a general area covered by the courses you take in first year. For example, it could be possible to take 205 and 215 in your first year, but end up working on theoretical statistics for your dissertation. Which two should you take? That depends on your background and interests. Here's some information on each course:
Rigorous and theoretical; a suitable course if your mathematical background is strong. Some knowledge of measure theory is very helpful here. The homeworks given in this course are time consuming and relatively difficult. Course material includes laws of large numbers, characteristic functions, and some families of stochastic processes.
Some students find this course to be easier than 205, while other students consider it to be harder. Best advice: talk to students who have taken both. Homeworks are time consuming. Course material includes both small-sample and large-sample or asymptotic theory for estimation and hypothesis testing.
Requires up to 12 computer assignments per semester. Typically, during the first semester these assignments require extensive work while the second semester generally requires large amounts of reading in addition to the computer labs. Course material includes statistical methodology, regression, analysis of variance, and generalized linear models.
The effort required by the first-year courses will leave you with little spare time. Most students are able to do their coursework and still have time to serve as a TA. If you can afford to do without TA income consider whether it is wise to work as a TA during your first year. Although some students enjoy TA work, others do not. The lower division courses can be quite challenging, as they make minimal use of formal mathematics. Also, if you are a 50% TA, try to get two sections of the same class; being a TA for two different classes requires a considerable amount of extra time. The department might offer alternative funding for the first semester to relieve new students from TA work. See section 4.3 for more details on GSIships.
One function of the first-year courses is to prepare the student for the written preliminary exams. These exams are difficult and require extensive study and preparation to pass. The department policy is that students take the written preliminary exams at the beginning of the second year. Failure to do so is regarded as evidence of unsatisfactory progress and will have an adverse effect on your level of financial support. See section 2.1.3 for department policy and a discussion of pass rates and related factors.
You should plan to study during the summer preceding the exams. If you take time off, do it at the beginning of the summer. Be careful to avoid taking too much time off. In preparing for the exams, your worst enemy is procrastination. Be sure that you work lots of problems and examples. Old exam problems are particularly suited to test your skills. Copies of exams from previous years are available from the graduate assistant in room 373. Talk to students who have already taken the exams for hints. Keep in mind that students frequently organize study groups to meet over the summer to discuss the first year course material, work problems from past qualifying exams, and otherwise prepare for the prelims. These groups are organized on an informal basis: establishing and participating in such groups is purely a matter of individual student initiative. Students participating in such groups in the past have found them to be very useful.
Entering Masters students should meet with the M.A. advisor as soon as possible upon arrival at Berkeley. The choice of first-semester courses is very important, as many courses build upon each other. A solid background in mathematics and elementary probability and statistics early on can help ensure success in more advanced courses later, as well as enable you to get the most out of the program. You might consider enrolling in math courses if your background is weak. Two years is more than enough time to meet the 24 unit requirement, so there is plenty of time to take courses from other departments in the university.
Most graduate-level courses will have a mixture of statistics and biostatistics M.A. students, students from other disciplines, and statistics Ph.D. students. Upper division courses will have a mixture of graduate students and undergraduates.
Unlike some other departments, it is not automatic to move up from the M.A. to the Ph.D. program, as the admission standards for the two programs are quite different. Entering students who might seriously consider pursuing a Ph.D. should speak with the M.A. and Ph.D. advisors early on (see section 2.1.2).
The M.A. students are just as much part of the Statistics Department as are the Ph.D. students, so you should not feel shy about attending graduate student events.
The course work taken by a Ph.D. student depends largely on that student's background. A student with a Master's degree in statistics should broaden his or her background to include areas of biostatistical application and as well as the theoretical basis of biostatistics. In addition, classes from the statistics department such as Stat 215 or 210 and others are desirable to solidify one's education in statistics. Most biostatistics students take PH 240ABC, Biostatistical Methods, a sequence of courses designed to introduce the student to the primary components of biostatistics. PH 240A covers risk research methods; PH 240B, survival analysis; and PH 240C, clinical trials. While enrolled in 240ABC, students concurrently take PH 247ABC, the data analysis workshop for these courses. If the Master's degree came from another field the above will apply; however, one must follow the similar coursework as a Master's student would to acquire the necessary background in statistics.
The program also offers higher level courses, such as PH 243AB (topics vary by semester). PH 241 and PH 252, although not as mathematically demanding, help provide a better understanding of epidemiological statistics. The program is very flexible, and can be tailored to suit any student's needs and interests.
Since the students are required to submit a question list for taking the oral qualifying exam (see section 3.1.2), it is a good idea to keep this in mind as one proceeds through the courses, working on the question list and what professors one would like on one's committee.
Second year students have much flexibility in their scheduling. Strongly suggested courses would be Stat 230, 232, and 240. In addition, the advanced biostatistics courses and epidemiology courses may be desirable, depending on the student's future goals. The adventurous may try the Ph.D. statistics courses such as 210 and 215.
Entering Masters students are assigned an advisor upon arrival at Berkeley. You and your advisor will decide which courses are appropriate for your first semester. Students are expected to enroll in 12 units during each of the four semesters of the MA program. There are no ``required'' courses for the M.A. degree; however, students are expected to take PH 240ABC, Biostatistical Methods. This course is designed to introduce students to the primary components of biostatics. PH 240A covers risk research methods; PH 240B, survival analysis; and PH 240C, clinical trials. While enrolled in 240ABC, students concurrently take PH 247ABC, the data analysis workshop for these courses. Additionally, in the first year, most students will also take Stat 200AB, Introduction to Probability and Statistics at an advanced level.
The Biostatistics M.A. program is flexible; students can take a variety of course in statistics and other areas suitable to individual needs, interests, and backgrounds. You might consider enrolling in undergraduate math or statistics classes if your background is weak.
Masters students usually take their oral examinations in the spring of their second year. The first year, however, is not too early to begin assembling a question list (see section 3.1.2) and practicing questions with peers. A seminar class is sometimes organized for the purpose of enabling students to become more adept at the oral expression of statistical concepts.
Students are required to make a detailed question list for their Master's exam. Like the PhD student, it is a good idea to keep working on the question list as one proceeds through the courses. Remember, the committee is most interested that you have learned the kernel of statistics and biostatistics rather than fancy or advanced methods.
There are no graduate level classes taught in the summer. Since M.A. students do not write a thesis, many students spend the summer working. Some internships are available from outside agencies; see Bonnie Hutchings for details.
The Statistics Department has an active seminar program, and hosts many social events. Biostatistics students are always encouraged to attend and participate in Statistics Department functions.
In order to understand the factors involved in student retention and tenure (time to receipt of the desired degree) in the Department of Statistics, the SGSA undertook a study of students entering during the years 1980-1984. These results were summarized and updated in 1991.
There are three basic types of Masters students: (1) joint Ph.D. students, who are taking the degree as part of a Ph.D. in another department, (2) Ph.D. students in the Statistics Department receiving the Masters degree based on passing the preliminary exams, and (3) M.A. students for whom the degree is the terminal point of their studies here. We will only be looking at the tenure of students in the first and third categories.
The basic withdrawal rate for these students in recent years has been about 25%. This figure includes students who transferred to other universities, or just lost interest in the course. Students with a prior M.A./M.S. had a lower withdrawal rate, around about 15%. The median period (in full-time academic years) in attendance from entry to graduation with an M.A. was 1.50 years. The joint Ph.D. students were observed on average to take a slightly shorter time to complete the degree requirements.
Passing the preliminary exams at the end of the first year has been an important indicator of the ultimate success of the students in the Ph.D. program. The overall withdrawal rate was about 25%. Approximately 60% of the students pass the preliminary exam on the first attempt; for this group, the withdrawal rate was only 6%. Among the students who initially failed their preliminary exam, about 55% eventually passed the exam. The students who were not able to pass the exam on their second attempt all left the Ph.D. program. The students who passed the preliminary exams on the second attempt had a withdrawal rate of about 10%. If a student fails the exam on the first attempt, the departmental advice (whether or not to sit again) is a good predictor of eventual success.
Also, there was only a small difference in pass rates between groups with different academic preparation at entry, with the prior M.A./M.S. students having a slightly higher pass rate and lower withdrawal rate than the students with prior B.A./B.S. The median period (in full-time academic years) in attendance from entry to graduation with a Ph.D. was 4 to 5 years, due partly to department policy (see section 2.1.3).
There is no formal mechanism for matching up students with thesis advisors. However, the graduate advisor and the first-year advisor assigned to a student may help in this regard. Typically, students choose a broad area of interest and talk to potential thesis advisors in their second year, and try to find a faculty member who is willing to supervise their work after passing the written preliminary exam. Generally, a student who shows personal initiative has a better chance of finding a helpful research advisor. If a student is unable to find an advisor in reasonable time, the head advisor will step in and encourage the most suitable faculty member to take on the student.
Going to seminars is one way to find out about current research in statistics. Seminars are held nearly every day; they are listed in section 6.7. Although nobody can be expected to go to all the seminars, it is a good idea to go to seminars regularly, say at least once a week. As well as fulfilling a scientific function, seminars are also excellent places to meet faculty members and fellow students.
It is wise to become acquainted with the faculty members and their interests (see sections 2.2.1 and 3.2) as soon as possible after arriving here. Good ways to meet the faculty include participating in the department's social and academic functions and being active in the various projects that the SGSA sponsors. Another time-honored method is to be an outstanding student in a class that the professor is giving. Being enthusiastic and having a good reputation is very helpful when you begin looking for an advisor. If you have a problem that you have done some preliminary work on, you will generally find it easier to obtain an advisor than if you approach them with a What-do-you-want-me-to-work-on? attitude.
Thesis advisors must be on the UCB faculty, but not necessarily in your program. In exceptional circumstances, it is possible to have an advisor from outside the university. Assistant professors are generally not thesis advisors, but there is no rule against having one as an advisor. Although most students get an advisor after passing the qualifying exams, getting an advisor earlier has advantages. And it is easy to change your advisor or thesis topic if you want to.
Berkeley graduates find jobs at a broad range of institutions throughout the world. Some recent graduates of Berkeley's programs in statistics or biostatistics started as assistant professor or lecturer in statistics, mathematics, computer science, or public health. Others began as postdoctoral fellows or researchers at various research institutions. Finally, several graduates have taken positions as statistician, consultant, programmer, or researcher for employers in industry.
The lists below show a sample of jobs taken by students who have earned a Ph.D. in statistics or biostatistics within the past few years.
It is wise to begin looking for jobs well before the beginning of the semester in which you will graduate. The Career Planning and Placement Center (CPPC), located on 2111 Bancroft Way, is a good place to begin looking for a job. Visit the placement office at least one semester before you graduate and leave your resume with them. The advisor for statistics graduates, Kathleen Stanton, can provide information and counseling; call 642-3602 to arrange contact. The Center offers a four-part seminar for Ph.D.s on searching for academic and non-academic jobs; for information on job-search orientation workshops (given several times each semester), contact CPPC at 642-1716. To aid you in academic job-hunting, the CPPC offers a dossier service through which they maintain and forward as necessary confidential letters of recommendation which you solicit. Every few weeks CPPC puts out a list of companies that will be visiting the campus; this list is posted periodically outside 373 Evans. For on-campus interviews, you need to go through the placement office. If you are interested in any of the jobs listed, call 642-8148 for information. The CPPC also has a library which contains many companies' addresses.
U.S. citizens can be reclassified as California residents for administrative purposes if they satisfy certain requirements during their first year at U.C. Berkeley. Once reclassified, a student is not charged non-resident tuition, effectively reducing fees by several thousand dollars per semester.
Foreign students with F or J visas cannot qualify for residency. Foreign students with permanent resident status (``green cards'') may qualify under certain conditions.
The department expects all U.S. citizens to qualify for California residency. Accordingly, the department does not grant tuition waivers to U.S. citizens after the first year of attendance. It is in the interests of U.S. citizens to take the necessary steps to qualify for California residency. These students should begin collecting the necessary documents as soon as they move here, including the airplane ticket to get here (if applicable). Soon after arriving, new students should pick up the relevant information in 120 Sproul.
To be classified as a California resident for tuition purposes, a student must have relinquished prior residence and established residence in California for more than one year immediately preceding the residence determination date. When you apply for reclassification, you should be able to demonstrate physical presence, intent to reside permanently, and financial independence.
Physical presence in the state for educational purposes does not result in reclassification as a California resident! You must provide evidence of your intentions to reside permanently in the state. Documents must be dated to verify that the actions were taken at least one year before the semester for which resident status is requested. Primary indicators of intent include:
You may also be required to prove that you were financially independent from your parents during the three calendar years preceding the semester you wish to be reclassified; this requirement is not imposed on students appointed as GSIs or GSRs at 49% time or greater during the first semester of reclassification. Be sure to keep good records of your tax history.
For more information, contact the residency deputy in 120 Sproul Hall, or phone 510-642-1614.
The Statistics Graduate Student Association (SGSA) is the official organization for graduate students in the Department of Statistics and the Group in Biostatistics at the University of California, Berkeley. The SGSA exists to look after student needs, to represent student interests, and to function as the formal channel for communication between graduate students and faculty. All graduate students in the Department of Statistics or the Group in Biostatistics are automatically members of the SGSA.
The SGSA is in contact with the Graduate Assembly (GA) and the Association of Graduate Student Employees (AGSE), to discuss topics of concern to all graduate students. SGSA officers occasionally meet with the department chair and department committees to discuss topics of interest to students. The SGSA has representatives on some of the department committees.
The SGSA provides some academic and recreational services to the campus statistics community by sponsoring a picnic each semester for graduate students, faculty, and staff; providing refreshments for the daily social; organizing student seminars; and publishing Degrees of Freedom.