Review of the Graduate Program in Physics
Submitted by the Graduate Studies Committee
D. Dahlberg (Chair), C. Campbell, A. Grosberg, B. Masimore, S. Rudaz, R. Rusack, J. Zhang, and Y. Kubota (ex officio)
PREAMBLE:
The Graduate Studies Committee was charged by the Head of the School to complete a critical review of our graduate physics program and to report back to faculty our findings and make recommendations for changes that would improve our program.
As a starting point for this review, the committee broadly defined the goal of our graduate program as the development of each graduate student to their fullest potential. It is important for them and the department that when they leave they have level of physics expertise commensurate with their degree. The courses that we offer, and especially those that we require, must provide the base knowledge for their research at Minnesota and for their careers; the tests that we administer to the graduate students must allow the student and the department to track their progress and finally, we must provide an environment that is conducive to attain these goals and to enhance their graduate experience while in our department.
In this report we discuss all of the main aspects of the graduate student preparation: the examining process, the curriculum, the advising process, the TA training and work and the environment for graduate students in the school. These are treated in separate sections where we present our observations or findings and make recommendations for changes. The committee did not consider or evaluate the research side of the students’ education.
In preparing this report we have made extensive use of surveys of our students and faculty as well as the Report of the Joint APS-AAPT Task Force on Graduate Education in Physics (Draft). In addition, we have had many discussions with faculty both at Minnesota and at other Universities and with our students.
I. ENTERING PREPARATION3.
- BACKGROUND AND OBSERVATIONS
- RECOMMENDATIONS
II. GRADUATE STUDENT ASSESSMENT4.
A. BACKGROUND AND OBSERVATIONS
B. RECOMMENDATIONS
1. GWE RECOMMENDATIONS
2. ORAL EXAM RECOMMENDATION
3. GPA REQUIREMENT RECOMMENDATION
III. CURRICULUM6.
A. BACKGROUND AND OBSERVATIONS
B. RECOMMENDATIONS
1. STATISTICAL PHYSICS RECOMMENDATION
2. CLASSICAL PHYSICS RECOMMENDATION
3. PHYSICS IN THE 21ST CENTURY
4. COURSE RELEVANCY RECOMMENDATION
5. ADVANCED COURSESRECOMMENDATION
IV. ADVISING8.
V. TA DUTIES9.
A. BACKGROUND AND OBSERVATIONS
B. RECOMMENDATIONS
VI. ENVIRONMENT10.
A. BACKGROUND AND OBSERVATIONS
B. RECOMMENDATIONS
VII. OMBUSDMAN11.
APPENDIX A.Excerpts from the Report of the Joint 12.
APS-AAPT Task Force on Graduate
Education in Physics
APPENDIX B. First year courses, course content, special topics courses.14.
I. ENTERING PREPARATION
- BACKGROUND AND OBSERVATIONS
Defining the best course structure for incoming graduate students is complicated by the fact that they arrive with different levels of preparation. This is partially due to our policy of recruiting students based on our assessment of their capabilities to master physics, rather than on their past accomplishments. This can lead to difficulties for the less well-prepared students and, while the responsibility to solve this rests with the student, some of them do need to take one or more of our 4000-level undergraduate courses. Students coming from another institution will need help deciding what courses match their level of preparation and we need to help them with this decision.
Just as we recognize that many students come to our program with gaps in their physics background, the same is true of their mathematical skills. For example, some students are unfamiliar with the applications of Fourier transforms, asymptotics, and Green functions – all crucial elements of graduate E&M and quantum mechanics courses. Given the limitations on their time, it is impractical to require all students to take a mathematical methods of physics course before taking E&M and quantum, and yet this very real problem must be addressed.
- RECOMMENDATIONS
We recommend that once an incoming graduate student confirms their acceptance to our graduate program we send to them copies of the final exams in our undergraduate QM, E&M, Analytical Mechanics, and Stat. Mech. & Thermo courses. They can, in consultation with the DGS, opt to take one or more undergraduate course in their first year.
We recommend that if they take two or more of our undergraduate courses they should be guaranteed up to a third year of full TA support and extra time to complete the oral exam requirements. Two-thirds of the graduate students’ responses in the survey were in favor of this and some with great enthusiasm.
Students, who have only minor holes in their background knowledge should be encouraged to take the 5000-level courses and the instructors should provide them with references and guidance on how they might master the missing material.
We recommend that instructors in 5XXX level physics classes devote an estimated 10% or so of classroom time (and to include a corresponding number of appropriate exercises in homework) to introduce their students to the use of the many important and recurringly useful mathematical methods which arise in the solution of physical problems (an ideal context providing students with a real motivation for the mathematics).
II. GRADUATE STUDENT ASSESSMENT
A. BACKGROUND AND OBSERVATIONS
The Graduate School requires that all University of Minnesota PhD students take a written and an oral exam. In our department these requirements have been implemented in the form of the GWE and the preliminary oral exam. The department further requires that students achieve a minimum GPA in their first year courses (which has degraded to an overall GPA requirement). In our survey both the faculty and the students have expressed concerns that there are large variations both in the format of these tests and the level of the material covered in them. We find that many of these concerns are legitimate and the committee considers it necessary for the school to have a clear set of guidelines and procedures for the students taking the exams and the faculty who are setting them.
In their answers to survey questions about the GWE, the students appear to be mostly happy with the format of the exam including both when the exam is given and that it is given twice a year. Several students, however, pointed out in the survey the lack of the balance in the core areas covered by the GWE and that graduate level material was sometimes included on the exam, which is contrary to the department’s policy. The committee concurred in both these views. This should be remedied as it is important for our students’ morale and the credibility of the whole process.
For the oral exam, the committee agreed that it is difficult to test graduate-level material in an oral exam setting. This is due to time constraints, the broad range of material to be covered, and the general anxiety of the examinees. It is also clear from the student survey and discussions with faculty, that the level of physics expertise required to pass an oral exam was not consistent between subfields or even within subfields with different committees. For example, some oral exams involve questions in all areas of physics, which may be at a graduate- or undergraduate-level, while other exams only have questions related to the projected research topics. It is our view that while it does serve a useful purpose in testing the student’s readiness to proceed to their research, it is not a useful forum for testing the level of knowledge of graduate-level physics; another method is needed for this. It is also important for the oral exam to be at a more uniform level.
B. RECOMMENDATIONS
We propose that the assessment of a graduate student be made in three distinct tiers. In the first tier their mastery of undergraduate material is assessed with the GWE, in the second their mastery of graduate level material is assessed by their performance in four of the core graduate-level courses, and in the third their readiness to conduct research in their chosen subject is assessed in the oral exam. In what follows we make recommendations for each of these three tiers.
1. GWE
The GWE must be both balanced and at an appropriate level. Great care must be taken in preparing this exam, concomitant with the level of seriousness with which is placed on the result by both the faculty and students.
We also recommend that the recent or current instructors of our upper division core courses conduct a final review of the exam to ensure that the problems are at the level appropriate to our upper division courses. They should receive the exam no less than one week prior to the date of the exam.
2. ORAL EXAM
We recommend that the oral exam include a 20 minutes presentation of their proposed research followed by questions from the committee. The overall length of the exam should not exceed 90 minutes. Also, as allowed by Graduate School rules, the number of members of the examining committee should be limited to four. As the questions should test the student’s readiness to continue or start their research, the committee should be composed of three faculty members conversant with the student’s research area and, mandated by the graduate school, a faculty member from outside of the department. These guidelines must be respected so that the exams will be given in a more uniform manner both in format and level of difficulty.
3. GPA REQUIREMENT
To ensure an adequate level of knowledge in graduate level physics in the areas common to all subfields we recommend that students be required to have a minimum GPA of 3.3 in the 5000-level core courses Classical Physics II (E&M), QM I, QM II, and Stat Mech, the syllabus for this last is discussed in the next section. This will limit the number of assessment courses normally taken by a student to two per semester.
We also recommend that there be an annual meeting of the graduate faculty to review the progress of all first and second year students. At this meeting an in-depth discussion can occur of each student’s progress. Most of this meeting would likely focus on those students making only marginal progress in course work with input from the faculty teaching our 5000- and 8000-level courses and their faculty advisors allowing a more thorough assessment of their progress. At this meeting the faculty can decide how best to counsel the students. The committee recognizes that this meeting would be rather long but we feel that this is would improve the quality of our graduate program.
III. CURRICULUM
A. BACKGROUND AND OBSERVATIONS
A major factor in deciding a graduate curriculum is to limit formal course work to the first two years of graduate school (in order to maintain an average time in graduate school of about six years). This places tight constraints on the number of courses that a student can take, and requires that the first year courses provide the necessary background for the 8000-level courses to be taken in their second year.
From the surveys, many discussions with faculty, and input from the Report of the Joint APS-AAPT Task Force on Graduate Education in Physics (see appendix for relevant excerpts) we find our first year courses mostly are appropriate for their purpose.
One place where we are very much out of line with most other graduate programs is in the area of statistical physics. From the APS-AAPT report, of the 137 PhD-granting physics departments 85% require a course in statistical physics; of the 67 that reported the text used, more than 2/3 used a text of a significantly higher level than Reif (with Pathria and Huang being the most frequently used texts - see Appendix II for details).
Another question that needed to be addressed was the relative emphasis of E&M and Classical Mechanics. In the APS-AAPT report 89% of the departments require a course in E&M; 75% of these were two semesters long. Also the report showed 77% of the departments require a course in Classical Mechanics and 93% of these being a single semester in length.
The survey of our graduate students revealed much discontent with our advanced courses. Specifically they feel, and the committee concurs, that the courses must have a content that is specified and published by the department and that the content should not be left to the vagaries of the instructor. The committee firmly believes that the purpose of the 8000-level courses is to provide the knowledge needed by our students to advance in their chosen research areas; courses specific to a faculty members research interests have their proper venue in a special-topics course. Consultations that we had with graduate students, who have completed their course work, strongly confirmed this view.
Another issued that was raised was the lack of some courses in the newer areas of research. For example, there were a number of complaints from students there were insufficient 8-level courses in biophysics.
Graduate students would like to see a broader view of physics than we currently offer. They find it hard to make an informed choice of their area of research. The school relies at present on the Introduction to Research seminar series and on attendance at the weekly Colloquium. The former is generally limited to the immediate research interests of the faculty member giving the seminar, while the breadth of content of the latter is variable. Furthermore, the students in their responses to the questionnaire, and in discussions, have made the observation that in our core courses few, if any, connections with modern areas of study are made. This makes the material seem dry and obsolete, with the interesting and exciting physics deferred until later.
B. RECOMMENDATIONS
1. STATISTICAL PHYSICS
It is our recommendation that a graduate-level course in statistical mechanics be required and offered in the first year of graduate work. An example of such a course based on the textbook by Pathria is given in Appendix II.
2. CLASSICAL PHYSICS
We recommend that we make a modification of our two-semester classical physics sequence, which currently is 1 semester of Classical Mechanics followed by a semester of electromagnetism. The first semester should be divided into ten weeks of classical mechanics and five weeks of electromagnetism. This in fact is the division that we had under the quarter system. An example of the material to be covered in these ten weeks is given in Appendix II.
3. PHYSICS IN THE 21ST CENTURY
We propose that that we replace the research seminar with a new required course “Physics in the 21st Century: An Overview.” The purpose of this course would be to better imbue our incoming students with a broad culture of the successes and challenges of physics today. (For more detail see Appendix B.)
4. COURSE RELEVANCY
In each of the first year courses, we recommend that faculty should include a few lectures which indicate the relevance of the material in which the students are gaining expertise to relatively recent discoveries or current frontiers in physics. Examples include the recent discovery of negative index of refraction materials in E&M, the quantum Hall effect in QM, the confirmation of Bose-Einstein condensation in elements other than helium in Stat Mech. Since any single individual is unlikely to have the expertise needed to develop these “stimulating” classes, we recommend the department commission the development of a dozen or so lectures for this purpose.
5. ADVANCED COURSES
The committee strongly recommends that the faculty in each physics subfield agree upon and provide specific course content for the 8-level core courses. This information should be given in the graduate student handbook so that students will know they are covering the material the sub-field faculty deem important for their success. There should also be an appropriate text book selected to at least serve as a reference.
New 8-level courses should be created as field advance. In addition, we need to ensure there are sufficient 8-level courses in any new subfields we are developing within the department such as biophysics and cosmology.
There should be a selection of special topics courses taught every year. As seen in Appendix B in the last three academic years we have offered only one advanced topics course.
IV. ADVISING
A. BACKGROUND AND OBSERVATIONS
Although the faculty survey ranked our advising as poor, the graduate student survey gave the process better scores. These conflicting scores were the result of the students’ positive ranking of our newly installed senior mentor advising. However, many comments from the students did agree with the faculty assessment of the faculty advising. One disadvantage of the senior mentor advising program is that the advisors change every year and there is no viable mechanism to ensure continued quality. Thus although senior mentor advising may be a positive addition in some years it may not be in others. The responsibility for advising rests clearly with the faculty.
B. RECOMMENDATIONS
We recommend that the department assign a committee of faculty advisors to each entering class.