Department of Electrical & Computer Engineering
.
Hiring and Mentoring Faculty Members: A department chair's
perspective Chaouki Abdallah, Professor & Chair
Department of Electrical and Computer EngineeringUniversity of New Mexico
Albuquerque, NM 87131 [email protected]
http://www.ece.unm.edu/faculty/chaoukiOctober 15, 2010
Department of Electrical & Computer Engineering
.
DisclaimersThese tips are available from various sources but I am using
many from Professor Dennis Bernstein: http://aerospace.engin.umich.edu/people/faculty/bernstein/
Other resources: http://www.cs.virginia.edu/~robins/YouAndYourResearch.h
tml http://www.mccurley.org/advice/hamming_advice.html http://aerg.canberra.edu.au/pub/aerg/edu12min.htm http://www.physics.ohio-state.edu/~wilkins/writing/Supp/
dazzle.html
October 15, 2010
Department of Electrical & Computer Engineering
.
About This TalkThe most important aspect of chairing an Academic
department is to hire and promote productive faculty members. In this talk, I will describe the dos and don't for a junior faculty member and chart an assistant professor path into tenure and eventual full professorship. This will include a discussion on recruiting graduate students, building a research team, and publishing.
I will also discuss how teaching, research, and service are "actually" evaluated at one US university via a variety of case studies.
October 15, 2010
Department of Electrical & Computer Engineering
.
What do I Look For?
I try to hire people who will succeed anywhere (not just at my university) subject to constraints.
In the last 5 years I hired 11 assistant and associate professors.
Two were CAREER awardees and 3 have since obtained CAREER awards.
October 15, 2010
Department of Electrical & Computer Engineering
.
What Criteria?
Good research potential as evidenced by current papers, choice of topics, etc.
Good recommendation letters from reliable and respected references.
Good teaching potential, as evidenced by prior experience, seminars…
Passion!
October 15, 2010
Department of Electrical & Computer Engineering
.
You Will Be Judged (again & again)
Research: scholarly and funding Teaching: classroom and graduate
students Service: within the institution and
professional
October 15, 2010
Department of Electrical & Computer Engineering
.
What is Necessary for tenure? Good research record. Papers should be in solid
journals. Good h-factor. Good teaching record (no major
mistakes/problems, some undergraduate teaching, 1 or 2 new courses). At least 1 PhD graduate.
Good service record (professional as well as at the department/university).
Good external letters.
October 15, 2010
Department of Electrical & Computer Engineering
.
What is Sufficient? Nobel Prize, Fields Medal, etc. Case 1: NSF Career awardee with solid research but
no PhD graduates and weak teaching record. Case 2: NSF Career awardee with solid teaching
record, but weak publication record. Case 3: Average funding/research record, excellent
teaching record, good service record. Case 4: Weak funding record, few excellent
publications, good teaching and service.
October 15, 2010
Department of Electrical & Computer Engineering
.
From PK Kumar How can I constantly produce new results? Isnʼt this an impossible job? I am having a tough time getting done with my thesis, how can I think of doing this as a career? Will I be able to come up with problems to solve all by
myself? Am I in the right place? What job should I apply to? How can I succeed in an academic career? …
October 15, 2010
Department of Electrical & Computer Engineering
.
Now that you are a faculty member
Do:1. Seek a respected mentor.2. Work hard and produce quality research and teaching.3. Contribute to your department intellectual discussions.4. Speak your mind up in a respectful and professional
manner.Don’t:5. Take sides.6. Get too involved in university politics.7. Make excuses for shortcomings in one area of your
portfolio.
October 15, 2010
Department of Electrical & Computer Engineering
.
By The Numbers
Technical Area and department-dependent: Research: 2 Journal numbers/year, $100K per
year Teaching: Some undergraduate teaching, 1-2
PhD graduates by the time of tenure. Service: PhD/MS committees, departmental
service, AND reviewing, editorials, etc.Outside letters from acknowledged experts trump
everything!
October 15, 2010
Department of Electrical & Computer Engineering
.
Don’t Get Swamped by Teaching
If teaching takes up all your time and swamps you, that is not good!
You need to pay attention to your research, and lots of it!
At the same time you need to teach well!
October 15, 2010
Department of Electrical & Computer Engineering
.
How do we generate research problems?
Let me illustrate a (relatively) easy route: Start with a practical problem, and try to get to the
heart of it.– Practical does not necessarily mean you can apply it
tomorrow.– It means motivated by a real application.
The real world is very rich and admits a lot of new ideas
October 15, 2010
Department of Electrical & Computer Engineering
.
Attend Conferences About 2 or 3 (or more) a year This is where you find out how little is known in
a field! You also get to know the people in the research
community! Also, you will get noticed through your good
work and its presentation! If you cannot get funding, pay for it yourself!
October 15, 2010
Department of Electrical & Computer Engineering
.
About Research FundingYou should get funding to do your research Not the other
way around!You should not do research so you can get funding However funding that supports your research is
important– Supporting graduate students (not too many …)– Travel to conferences!– Getting equipment for your research – computers, lab, ..!– Support staff to save your time, so you can do your
research! Funding is not the performance metric!
October 15, 2010
Department of Electrical & Computer Engineering
.
Getting Funded Getting funding for your research is not magic! It is a question of writing proposals, talking to program
managers! You just need to talk to peers, senior faculty, etc.! Find out all the opportunities that there are, and target all
of them in a systematic way! It is just a question of approaching it in an organized
way! In the long run do good work and everything else will
follow -funding, students, glory, …
October 15, 2010
Department of Electrical & Computer Engineering
.
About Tenure
Do great research! Teach well! Whatever service you are assigned,
execute it well! Be reliable with respect your service
activities!
October 15, 2010
Department of Electrical & Computer Engineering
.
Later in Your Career You will be known by your best work! Not by how many
papers you have published! H-factor but not exclusively…. In the beginning, aim to get published, and get over that
threshold first! The norm by which your accomplishments are measured
is L∞ not L1 – Max {Papers} rather than Σ Papers!
October 15, 2010
Department of Electrical & Computer Engineering
.
About your Resume
Not too long ago, The New Republic received a near-perfect resume from a young man named Adam Wheeler, who was vying for a literary internship at the magazine. The three-page document boasted a 4.0 GPA from Harvard, 18 prize and scholarship awards, invitations to lecture, two working manuscripts and proficiency in Old English, Classical Armenian and Old Persian.
October 15, 2010
Department of Electrical & Computer Engineering
.
Sample 1
Title: Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity
Abstract: There are many natural scientists, especially physicists, who continue to reject the notion that the disciplines concerned with social and cultural criticism can have anything to contribute, except perhaps peripherally, to their research.
October 15, 2010
Department of Electrical & Computer Engineering
.
Sample 2
Title: Rooter: A Methodology for the Typical Unification of Access Points and Redundancy
Abstract: Many scholars would agree that, had it not been for active networks, the simulation of Lamport clocks might never have occurred. The notion that end-users synchronize with the investigation of Markov models is rarely outdated. A theoretical grand challenge in theory is the important unification of virtual machines and real-time theory. To what extent can web browsers be constructed to achieve this purpose?
October 15, 2010
Department of Electrical & Computer Engineering
.
Sample 3
Title: Guaranteed Margins for LQG Regulators.
Abstract: There are none.
October 15, 2010
Department of Electrical & Computer Engineering
.
Professor Bernstein's Ten Steps to Writing a Paper
1. Recognize that you have a result: At some point in your research you must recognize that you have sufficient results to warrant a paper. Your paper must have a "reason for being." Although experience is needed to make this decision, be aware that such a decision is necessary before you begin writing your paper.
October 15, 2010
Department of Electrical & Computer Engineering
.
2. State your main result: You will write your paper "inside-out." First, write out your main result as precisely as you possibly can. Keep honing it until the result and its proof are perfect. This result will be the foundation upon which the rest of your paper will be built. Since any errors or ambiguities here will propagate throughout the paper, be sure that your main result is absolutely correct. Everything else in the paper will flow from this "high potential energy" result. Try to make your development of your main result as self-contained as possible without relying on results from other sources.
October 15, 2010
Department of Electrical & Computer Engineering
.
3. Work out the principal consequences: Next, write out the principal consequences of your main result. Work out special cases of your main result in as much detail as possible. Remember that your readers may only be interested in special cases. Many papers were ignored because readers could not relate to the setting in the paper. Therefore, try to connect your main result to as much related and specialized work as possible. Revise your main result if that is suggested when you work out these consequences.
October 15, 2010
Department of Electrical & Computer Engineering
.
4. Assemble your reference list: Collect together the references that you will cite. These reference will provide a frame of reference for your paper. They will show the reader that your work was influenced by such-and-such school of thought, a particular applications area, or philosophical point of view. Be sure that you research the literature thoroughly, including both conferences and journals, foreign, domestic, English, foreign language, etc. Don't assume that work of "obscure" authors is less worthy of citation than the work of "famous” researchers.
October 15, 2010
Department of Electrical & Computer Engineering
.
5 . Collect notation: Make conscious, careful decisions about all of the notation you choose to use. Decide if you will use traditional notation or opt for your own variations for good reasons even though your reader will not recognize things as readily.
6 . Collect background material: State all of the background material you will need. Be precise in your definitions and lemmas. Give accessible references for this background material. Be specific in your citations, such as "Theorem 12.3 on page 723 of [19]."
October 15, 2010
Department of Electrical & Computer Engineering
.
7. Produce numerical results: Illustrate your results with numerical results that clearly illustrate your contribution. Your numerical results should be as transparent as possible. Choose examples that are as simple as necessary to demonstrate your results. Redo examples by prior researchers to compare your results.
October 15, 2010
Department of Electrical & Computer Engineering
.
8. Write the introduction: Only after all of the above steps are completed should you begin to write your introduction. Think extensively about what you want to say before you write it. First outline your thoughts into a story line that reflects the order of ideas as you would describe them to someone in a presentation. Better yet, imagine you are explaining to your colleague what your paper is about. Be sure that in the Introduction you clearly state the contribution of the paper. For example, have a sentence like "The purpose of this paper is to go beyond earlier work by ..."
October 15, 2010
Department of Electrical & Computer Engineering
.
9. Write the conclusion: Repeat a few points from the introduction in the conclusion section and then add some additional perspective on the work. For example, you may comment on the potential applications of the work, possible shortcomings, and directions for future work.
October 15, 2010
Department of Electrical & Computer Engineering
.
10. Write the abstract: Last of all, write the abstract. Compress the introduction into a few key sentences. Imagine that you are a researcher in China who only has the abstract of your paper. Could you figure out what is in your paper and what its contribution is? Or imagine you have 30 seconds to describe your paper to a busy person.
October 15, 2010
Department of Electrical & Computer Engineering
.
Professor Bernstein's Top Ten Tips for Giving a Presentation
1. Introduce yourself, acknowledge your coauthors, and thank your host: Don't forget to do these. This will help you get off on the right foot.
2. Warm up your audience: Say something interesting to capture their attention. A story or joke will do fine.
October 15, 2010
Department of Electrical & Computer Engineering
.
3. Communicate with your audience: Look at your audience to get feedback from them. Look at their eyes and read their expressions. Talk to them. Communication is two-way even if you're the only one who's talking.
4. Think of your talk as a story, and be entertaining: Describe your work as you would tell an interesting story. People like to be entertained even if it's serious research.
October 15, 2010
Department of Electrical & Computer Engineering
.
5. Give adequate background and motivation: Tell them what you're doing and why you're doing it. Motivate them to care about what you're doing. Relate what you're doing to something they might care about.
6 . State your main results carefully without lots of detail: Be clear about what your results are. Minimize detail in favor of clarity. Make sure they get the main idea. They can always read your papers for the details.
October 15, 2010
Department of Electrical & Computer Engineering
.
7. Have uncluttered slides: View your slides as emphasizers and reminders about what you want to say. The focus should be on what you're saying, not what the slide says. Don't be a slave to your slides. NEVER read more than a few words on a slide. The audience can read much faster than you can speak and it's annoying for them to have you read what they have already read.
October 15, 2010
Department of Electrical & Computer Engineering
.
8 . Speak clearly, not too fast and not too slow: Both too fast and too slow are bad. Keep their interest, but don't lose them.
9. Time your talk and select your slides in advance: Have an idea how long your talk is. Running over is annoying to the audience. Choosing your slides in real time shows a lack of respect for the audience while they wait for you to decide what to tell them.
Polish your slides and your talk: Have a polished, professional presentation.
October 15, 2010
Department of Electrical & Computer Engineering
.
Trust (Faith) & Research One of Hamming's lessons is never trust
without question someone who claims to be giving you highly accurate data to analyze – not because they're deliberately lying to you but because the data is never as accurate as people think. Another is: the ability to tolerate ambiguity.
I will go a step further: Never trust anything from anyone without question!
October 15, 2010
Department of Electrical & Computer Engineering
.
Why God could not get tenure
1) He had only one major publication2) It had no references3) It was not published in a refereed journal4) Some even doubt he wrote it himself5) It may have been true that he created the world,
but what has he done since?6) His cooperative efforts have been quite limited7) The scientific community has had a hard time duplicating
his results8) He never applied to the Ethics Board for using human
subjects
October 15, 2010
Department of Electrical & Computer Engineering
.
9. When one experiment went awry he tried to cover up his mistakes by drowning the subjects
10. When subjects did not behave as predicted, he delete them from the sample
11. He rarely came to class, but told students to read the book
12. Some say he had his son teach the class13. He expelled the first two students for learning14. Although there were only ten requirements, most
students failed his tests15. His office hours were infrequent and usually held on a
mountaintop. October 15, 2010
