I had numerous past posts on geography and other criteria in applying for jobs.
There are several interesting recent posts (echoing each other) on this topic in blogosphere: Drug Monkey, Blue Lab Coats, Terra Sigillata, and Uncertain Principles.
Most people on the job market are too selective, a fact they don't realize until it's often too late. Offers to beget offers, and some "lower-level" institutions may seem far more attractive after you visit them (or after getting desperate enough).
And once application package is finalized, the costs of sending one extra copy is minimal, but potential benefits could be quite substantial. Finally, there's nothing wrong with rejecting an offer.
The academic market is indeed very competitive and one is wise to "diversify", which means applying for non-academic positions.
But only half-agree with some aspects of Chad Orzel's rant on "toxic" idea of aiming for academic position, or limiting your choices based on what is (in many cases) bound to be unrealistic expectations.
The reason why so many physics postdocs search primarily for academic positions at research universities is often not because of preconceived toxic ideas, but because they enjoy doing cutting-edge science. For many disciplines, academia provides the only environment that provides such opportunities. Generally, there are three traditional bastions of scientific discovery: research universities, national laboratories and industry.
Industry has been cutting on fundamental research for the past 10 years or so, best examplified by Bell Labs. So for the past few cycles the job search options (for people who want to continue doing research) are limited to academia and national labs. And national labs hiring freeze phase due to continued resolutions weren't exactly creating a lot of openings.
Chad is correct that there are many other career choices for PhDs - "between a tenure-track academic job and living on the street eating scraps from a trash".
Unfortunately, these choices involve giving up on research career (these include many tenure-track, but non-research oriented institutions). For some people quitting research this may be an acceptable compromise, whereas living in undesirable geographic area is not. For others, science beats geography.
One thing to keep in mind though - it IS possible to move from one university to another, and from national or industrial lab to university or back.
It is also possible to move from a lower ranked university to a higher ranked, provided that you are producing quality research. It is also obviously easier to move down the ladder.
However, it is very difficult to move from teaching position at small liberal arts or community college, Wall Street or consulting company back to research.
There are obviously some trade-offs and everyone's mileage may vary. But it's no reason to limit your options early on in application process.
Friday, October 10, 2008
Subscribe to:
Post Comments (Atom)
14 comments:
Hi IP,
Interesting points. I'm new to the academic job search. I was wondering if you could give me some advice. How many places should one apply to? I have the impression that it at times can be bad to turn down offers. For instance, suppose you only got one job offer at a place you don't really want to go to, and you turn it down, and say do another postdoc. Then getting future offers might be a bit more difficult because depts in general don't want to be turned down.
Also, do you think its a good idea to look at teaching as well as research places within the same year? This might be overwhelming...
LFJ
IP, I know of very few examples where a scientist could move from a (permanent) staff member position in a national/industrial lab to a (tenured) faculty position in a university. Universities are reluctant to offer (tenured) positions to these people unless they are quite distinguished (like Steve Chu). Most of the people who go to national/industrial labs for higher salary (foolishly) are stuck there forever.
There are many industrial/national lab "refugees" now in academia. I just thought of 5 in 30 seconds without really trying (they are mostly people I know). And I know there are MANY from Bell Labs who ended up in various universities when they were really shedding people in the 90s.
Most of these folks are very established, very well-respected scientists, who were in high-visibility or leadership positions in their labs. It does seem rare for Assistant or Associate Prof level people to go from industrial/national labs to academia.
I went to a National Lab as a postdoc, and now am a tenure-track professor running a research group, but not at a highly ranked department. I think a post-doc at a national lab won't necessarily get "stuck" but it is really important to work in a group that frequently publishes high profile papers in your field, and also important for a young scientist to have a clear contribution to her/his own projects. If you are viewed by search committees as being just a cog in a large science machine, then you are certainly "stuck."
Of course "stuck" may mean a very enjoyable, comfortable existence. I have lots of friends who are very happy at National labs.
What the anonymous above (11:41AM) said is right on point.
There are actually quite a lot of people who came to universities after spending some time at national labs or industry - in some top departments it could easily be majority of condensed matter people. Not just Bell labs, but also Exxon, IBM, etc.
Most of these labs either abandoned fundamental research entirely or substantially reduced it. So there's a vacuum that may be filled by national labs to some extent.
I don't think postdocs are necessarily hurt by going to national labs - not as long as they still do exciting science and don't become technicians pushing buttons - which applies to university postdocs as well.
However, I think it is more difficult to go from lab to academia, than from one university to another. The reason is that for junior hires universities would rather consider younger (postdoc-level) scientists, and for senior (tenure-level) scientists they are more likely to look for big names with established credentials.
But national labs still provide a very good environment for doing research, despite all the bureaucratic and safety-related management BS, which could make one feel sometimes as if they have been transported to "Office Space" or perhaps "The Office".
lfj - I think applying to ~20-30 places seem to be the average number from people I talk to, but I don't see why someone couldn't apply to 40 or 50 places. I don't think declining an offer is anything that is frowned upon, or something that should affect your application decisions.
I think it's perfectly fine to look at both teaching and research, even though obviously the application packages will need to be targeted accordingly. There's actually a nearly continuous spectrum of departments in terms of research/teaching importance: top research, not-so-top research departments, departments with only masters degree program and finally undergrad-only programs - liberal arts colleges, etc. Even though even undergrad-only universities now may be interested in your research experience.
So it's more of covering a wide range of the spectrum, rather than just "A vs. B".
I know of many examples of professors moving from industry to the university. They're cherished in lower-ranked research institutions that aren't looking to propagate professors.
I know of two in particular that worked in industry for four years with six figure salaries out of grad school and walked into associate professorships.
Tell me that they were fools. Tell me that they were fools by showing me the postdocs that walked directly into associate professorships four years out of grad school.
A problem I've seen is that most people who've worked in industry for a decade or so and are pulling in $175k or so aren't willing to walk into an assistant professor's job at $60-75k or so.
The candidates are right for expecting more, and the schools are right for insisting on less. Only in one of them blinks (usually the candidate) does a deal get made.
Even so, this year might be a bad year for CM positions, since Seagate is closing down its research facility in Pittsburgh (which is only ten years old and spawned at least a half dozen professors already).
****
Someone once told me, and I later found out that he was right, that the best way to look for professorships is to look at it on a two-year approach.
The first year, jump at anything that moves. You need to have enough applications out there to expect a couple of interviews so that the next year you know both (1) what kind of institutions are interested in you and (2) what kind of things they're looking for.
[Not everything is lost research-wise if you go to a university that has no graduate program in physics, especially in CM. You can often get grad students from engineering departments.]
So how many is the right number?
You can do the math yourself (it's fun). You know, given there are (say) 200 applicants for each position and every institution interviews 4 applicants, what is the likelihood of getting M interviews when you apply to N institutions. Since the expected number of interviews needed before an offer is four, the answer is very disturbing...
...and it's too optimistic.
The number of applications varies widely from institution to institution depending on its desirability in terms of reserach and location and its specificity in the job announcement, and more importantly, universities are biased in their selection process.
They're biased toward good researchers.
the math from the post above sounds about right - in many cases it's 300+ applications, and the number of people interviewed varies, but is typically 5-8. Keep in mind, it's often the same 300+ applicants, or at least with substantial overlap (and there might be substantial overlap among interviewees too - a small fraction of candidates will get a large fraction of interviews).
Going on job market multiple years may be a necessity dictated by the market, rather than a strategy. Because it's a small number game, I am not sure much can be learned from a small number of interviews - so I would advise treating every interview as the real thing, rather than a test-run.
Getting tenure positions 4 years after PhD sounds way too extreme of a case for Physics (this is perhaps more usual for engineering though). I don't see how it would be possible at least in my department to offer tenure to someone roughly the same age and record as people being considered for tenure-track assistant professor positions - there has to be a highly compelling case, and even then it might not be possible.
Seagate exodus is more likely to affect engineering department than physics departments - most of their research is in applied nanomagnetism/magnetic devices, and this will affect only a small fraction of departments looking to hire in that area. This is nothing like Bell Labs exodus.
Thanks for all of the feedback. This is very useful.
I've heard significantly varying results concerning the number of places to apply to. My gut reaction is that 30+ places seems high. A senior PI at my current institution (though it is one of the top ones) said I should apply to around 8. But then I bet there is a good chance I'll be on the street jobless next year.
Maybe this is something you can give me feedback on too since you're a PI. 30+ for me is fine. I don't mind spending weekends writing applications and (e)mailing stuff out. But this does seem to be a bit of an overburden on letter-writers. For instance, they have other students and commitments, and addressing this many letters for each of them year after year would probably get a bit overbearing.
I wish that application packages wouldn't solicit letters at the initial stage, but would wait and contact your reference only if they were fairly interested in your application.
I don't think my letter writers would be too happy if they got an the email from me asking for 50 letters.
Also, about the continuous spectrum of schools from research institutions to liberal arts colleges. That seems about right to me. The main difference in opinion that I have is that some liberal arts colleges can be better places for research than masters or even Phd granting universities. For instance, these places (Williams for instance) have some of the most motivated and creative undergraduates in the country. On the other hand, I can picture some of the public masters-granting universities to be run on more of a business model, and have students there who just want to get a degree to get a job, where learning for the sake of learning isn't a top priority. Anyway, minor difference ...
Ifj, let me offer my opinion too -- first of all, this is a conversation that you should be having with your own PI. They are the ones who know best, in my opinion. If they say "Oh ? Wow, gee, I dunno... apply wherever you want ...", tell them , politely but firmly, that you really expect guidance from them. If on the other hand they ask you pointed questions, aimed at narrowing down the spectrum of institutions, and after a couple of conversations they come up with a list, that's where you can apply.
Secondly, let the records of individuals speak for themselves.
What I mean is: if the senior PIs who tell you to apply for 8 positions have a good record of placing their advisees in positions that would be desirable to you, they probably know stuff that you and I don't, and probably tell their advisees exactly where to apply...
On the other hand, if these are people who have not supervised postdocs for a while, perhaps are themselves not terribly active in research anymore, then there is a good chance that they talking through their hat.
Finally, I don't think you should worry about asking people to send out 100 letters of recommendation. They will be happy to do that. If they come and tell you "I don't think you should apply to these many places", you can always say something like "Well, you know, I am sorry for the inconvenience but I would really like to get a job, and I am willing to go to each and everyone of these places... what's your reason for suggesting that I don't ?". But honestly, I doubt if you will run into such a problem.
In 1997, when I got my job, I applied to around 30 places -- each and every one of those I could apply for, including at liberal arts colleges and many places I really was not crazy about.
As for moving to national labs to academia -- of course it's very common and happens frequently. Especially over the past ten years, with national laboratories becoming increasingly unfriendly to basic research, there has been an exodus toward academia. Particularly conspicuous are cases of entire research groups (e.g., at Livermore or Los Alamos) which disappeared almost overnight, with each member thereof finding an academic position without much trouble.
I personally know a dozen of them, and in many cases the pay cut was substantial but they were all willing to endure it because they wanted out of the lab.
Okham, thanks. You're right that these are the sorts of issues I should resolve with my own PI. But it's always useful to get as many perspectives as possible. For instance, the folks within my direct advisor network had the fortune of not really having to apply.
One final thing which I was kind of wondering about are research proposals. For experimentalists, it really seems like there needs to be something concrete for future plans. After all, when they arrive they need to start ordering stuff and setting up experiments. Here a detailed picture of future activities is imperative. On the other hand, I'm a theorist. Correct me if I'm wrong, but it seems like the most useful stuff you can write about in a 'statement of research interests' is the past research accomplishments and their impact. Any theorists can make a long list of projects they would like to do in the future. But how valuable would such a list be for the search committe? I'm thinking not very. So a good statement here would be a summary of past research and impact combined with future directions.
Do you agree?
lfj - Just to throw in my $0.02.... In my experience research statements are just as important for theorists. Your past accomplishments are probably fairly clear from your publications and letters. There's nothing wrong with establishing some context and your bona fides by briefly summarizing past work, but the committee really does want to see what you want to do. At the very least, they want to see some evidence that you've given the matter some real thought (e.g., where can I have a real intellectual impact given my background and skills? What problems are appropriate for the near-term and the longer term? Which problems would work better for a postdoc vs. grad students vs. undergrads?). In a highly competitive market, it's important to make clear that you actually do have a vision of where you want your research to go. A candidate that seems like they're just winging it won't do well relative to the competition.
I definitely agree with Doug - you want to specifically mention in research statement which topics interest you the most, what you expect to work on, why it is important, what new elements/techniques/approaches will you bring, what you plan to accomplish etc.
Not doing so would be a serious mistake that will put you at a serious disadvantage.
Research statement is one area that can be improved on very easily over a short period of time - whereas other aspects (publication list, recommendation letters, CV) are to large extent out of your control.
Research statement should not talk about what you have done, it should be forward oriented. Imagine faculty members asking you: "why should we hire you? what will you work on? what will you accomplish?"
You can talk about your past experiences of course, but only in as much as it supports your claims about the future. But definitely make it future-oriented. The committee wants to make sure you have some ideas of your own - your past accomplishments might as well be your advisor's ideas, with you being a good soldier.
Doug, Thanks, your remarks are useful and this makes sense. As you're on the committee I'll take your $0.02 :)
Sylow,
Once again, you are making up the facts as you see them. I am in a TT position after spending some years as a postdoc and then as staff in a national lab. FWIW, I negotiated my starting salary up to just under my last salary as a Fed, so I didn't have to go back to $60-70k after making my higher salary as a postdoc and young PI.
I know at least 4 people who went from staff positions at my national lab into TT positions. If I remember correctly, 2 went in as Asst Profs and 2 went in as Assoc Profs w/tenure. At least 1 in 4 or 1 in 5 postdoc fellows go back to academia after 2-3 years at my lab (not all to research universities, though--some decide they prefer teaching after all). It depends on what you do at the lab whether you can make the move.
Post a Comment