Length of postdoc before getting faculty position?

[TwinStar]

What is the longest amount of time you can remain a postdoc before you are no longer a viable candidate for faculty positions?

Have you heard of anyone getting a faculty position if they've been a postdoc for more than 5 years? 8 years? Longer?

 

[twofish-quant]

What is the longest amount of time you can remain a postdoc before you are no longer a viable candidate for faculty positions?

Have you heard of anyone getting a faculty position if they've been a postdoc for more than 5 years? 8 years? Longer?

Postdocs for physicists are typically three years. Typically you need two before you become eligible for an junior faculty. You might be able to squeeze out a third, but no one is going to hire you for a fourth.

 

[ParticleGrl]

It varies a bit by field, in particle physics theory, your window to get a faculty position is between 5 and 8 years postdocing. After that, universities will start looking for people "younger and hungrier." As an anecdote, I know of one person who was shortlisted a few times at good schools after his 6th and 7th year of postdocing. He hasn't been shortlisted since, and is on year 11.

For liberal arts colleges in physics, most seem to want one postdoc 2 to 3 years, but those jobs typically won't go to theorists (again, based on anecdote).

 

[caffenta]

He hasn't been shortlisted since, and is on year 11.

11 years as a post-doc?? The horror...

 

[Vanadium 50]

I don't think it's a desire for "younger and hungrier". I think you are mixing cause and effect.

Suppose the top 20% of postdocs move up. If you have someone who wasn't in the top 20% in years 3,4,5,6,7,8,9 and 10 of their postdocs, it's not very likely that he will be in the top 20% in year 11. Every year new smart people enter that pool.

 

[ferm]

In many countries of Europe it is quite more common to get in a tenure track after three postdocs rather than two. Doing four postdocs is not so common, I wouldn't say it is impossible to get tenure then, but not many people are willing to apply for postdocs being 35 years old...

 

[ParticleGrl]

Suppose the top 20% of postdocs move up. If you have someone who wasn't in the top 20% in years 3,4,5,6,7,8,9 and 10 of their postdocs, it's not very likely that he will be in the top 20% in year 11. Every year new smart people enter that pool.

First, no one in particle theory moves up in year 3 or 4, and rarely do people move up in the 5th. So this guys first years on the competitive market, he made short lists. The difference between the candidates on the short list and the guy who gets the job is often luck. If we assume a uniform quality distribution (which seems fair), then he should consistently be in the top 20%, as long as he is keeping his output up.

All of this, of course, supposes there is an objective, reasonable criteria for top postdocs, which might not be the case (trends in the field can have much bigger impact on who gets hired then any individual's publication rate). If the top 40% are equally qualified, but only the top 20% can get jobs, then its basically a lottery. In this case, arbitrary criteria (like age) probably come into play.

 

[atyy]

Alan Guth was a postdoc for 8-9 years.

 

[Vanadium 50]

And Lance Dixon was a postdoc for one.

There is no magic answer, but after a certain number of cycles, people usually get the hint. My point is not that universities are thinking "too old", it's that they are thinking "this other person is better". And there are excellent people always entering the pipeline.

 

[ferm]

And Lance Dixon was a postdoc for one.

There is no magic answer, but after a certain number of cycles, people usually get the hint. My point is not that universities are thinking "too old", it's that they are thinking "this other person is better". And there are excellent people always entering the pipeline.

IMHO, there is a point where it is quite difficult to order people by their quality, especially when you are trying to estimate the future scientific output of this person. Naively, it sounds as a much better strategy to give professorships to people with demonstrable experience publishing papers, and with a good network of contacts, which is easier to have after two or three postdocs. Given the amount of excellent people in the market I don't think competence between universities trying to get people is really an issue.

Of course this sucks for the researchers, but I don't think universities are caring about that.

 

[twofish-quant]

My point is not that universities are thinking "too old", it's that they are thinking "this other person is better".

Personally, I think that people presume that the other person is better because he isn't too old. When you are talking about post-docs, the criterion for "better" is extremely subjective, and if you've had a lot of post-docs that's taken as evidence that you are not good.

It works the other way in industry. The longer you've worked the "better" you are assumed to be.

 

[Vanadium 50]

Yeah, yeah, yeah. I've heard this before - university faculty are a bunch of stupid yokels that are simply prejudiced against this guy who's spent 11 years as a postdoc. Prejudiced, I tell you!

While there are subjective aspects, there are objective ones as well - papers published, papers cited, talks given, and so on. And even the subjective ones are not purely subjective - a letter that says "best postdoc I have ever had, or am likely to have" is stronger than one that says "clearly in the top two-thirds".

If you look at a stack of applications and divide them in half - strongest and weakest - you will find that the people who have been around for 11 years are preferentially filling that second stack. Are there exceptions? Sure - right off the top of my head, there's one strong candidate in her 9th year who has declined to even apply for many positions.

It's a sad fact, but it's nevertheless a fact that some people will never get faculty positions, no matter how many postdocs we take. We had a fellow here a couple years back who complained that the field couldn't find a job for him. After grad school and his first postdoc, he has one publication, which was co-written with his advisor, and it has 15 cites, 8 of them self-cites. It's very unlikely he will be moving on.

 

[atyy]

http://particle.physics.ucdavis.edu/rumor/doku.php?id=where_are_they ?

I think Kari Karhunen ended up in insurance. The anecdote I read says a mathematician friend asked him how he could stand it. His reply was that he still solves problems, just different ones.

 

[ParticleGrl]

Yeah, yeah, yeah. I've heard this before - university faculty are a bunch of stupid yokels that are simply prejudiced against this guy who's spent 11 years as a postdoc. Prejudiced, I tell you!

I don't think that's the point anyone is making. The point is that the job market is tight enough that there are far too many good applicants for each spot. If you have a dozen really good people and you need to limit to 5 for a shortlist and then 1 for a job, how do you do it? Thats when number of postdocs might come into play.

Keep in mind, not everyone with a solid publication rate will get lucky. Many will, some won't.

 

[ferm]

Yeah, yeah, yeah. I've heard this before - university faculty are a bunch of stupid yokels that are simply prejudiced against this guy who's spent 11 years as a postdoc. Prejudiced, I tell you!

In my opinion, there are many, many, many examples in the past no to have to think about a particular guy. You're going to tell me you don't know many people who have ended up demonstrating they were brilliant, and who struggled a lot to get a faculty position?

And the reason isn't faculty being stupid yokels, it is that the competence has reached a level of insanity in which you have to select between a bunch of brilliant candidates, and in this case randomness (and prejudice) can easily come into play.

 

[twofish-quant]

Yeah, yeah, yeah. I've heard this before - university faculty are a bunch of stupid yokels that are simply prejudiced against this guy who's spent 11 years as a postdoc. Prejudiced, I tell you!'

Hardly my point. University faculty have very good reasons for not hiring someone that has spent too long as a post-doc. The problem is that from the point of view of someone that is in the system *it doesn't matter* if faculty have good reasons or not.

Faculty search committees are doing their best in a broken system.

There are objective ones as well - papers published, papers cited, talks given, and so on. And even the subjective ones are not purely subjective - a letter that says "best postdoc I have ever had, or am likely to have" is stronger than one that says "clearly in the top two-thirds".

The problem with academia is that it's very easy to get into a downward spiral. Once your you make a mistake for any reason, you are going to have less ability to recover. If you have a set of bad papers, your are in a worse position to get grant money and collaborations which leads to even weaker papers.

The basic problem is that there are too few faculty positions to go around. The system *has* to be ruthless because there just are not enough positions. You *have* to be perfect, because if you are flawed, that's one less person that has to occupy the open seat.

The problem with this is that research is all about making mistakes, and we really have to ask whether a system that emphasizes perfection really is the best way of doing research.

It's a sad fact, but it's nevertheless a fact that some people will never get faculty positions, no matter how many postdocs we take.

Sure. So we need to do something about that fact.

After grad school and his first postdoc, he has one publication, which was co-written with his advisor, and it has 15 cites, 8 of them self-cites. It's very unlikely he will be moving on.

Tough luck Albert Einstein.

 

[twofish-quant]

Also if you get a physics Ph.D., you are not an idiot. You clearly are able to to decent research if given a chance to do it, and it's a major problem if you aren't.

The important question isn't whether physics faculty are evil nasty people, because they aren't. Given the basic constraints of the system, they are behaving in ways that make sense, and in ways that I'd likely behave if I were in that situation.

The important question is "does the system make sense" and I don't think it does. The way that the system works just kills any sort of creativity and original work. To give an example, people have been working on string theory for decades, and have got nothing useful. It would be useful for some Ph.D. to at least think about alternatives, but they can't do that in the current system because there is a good likelihood that their alternative just will not work (most new ideas don't), at which point they've killed their chances for a faculty position and are unable to get another bite of the apple.

The reason this worries me is that I've seen the end product of this sort of system in academic finance which is a pretty useless and pointless endeavor that involves pure navel gazing.

 

[caffenta]

It works the other way in industry. The longer you've worked the "better" you are assumed to be.

But you also hit that too-old ceiling eventually. It's just later.

The important question is "does the system make sense" and I don't think it does. The way that the system works just kills any sort of creativity and original work. To give an example, people have been working on string theory for decades, and have got nothing useful. It would be useful for some Ph.D. to at least think about alternatives, but they can't do that in the current system because there is a good likelihood that their alternative just will not work (most new ideas don't), at which point they've killed their chances for a faculty position and are unable to get another bite of the apple.

One of the problems is epitomized by this:

...with a solid publication rate...

After grad school and his first postdoc, he has one publication, which was co-written with his advisor, and it has 15 cites, 8 of them self-cites.

I understand you need to have some kind of metric to judge people, but quantity is not quality. Technically, I really don't care how academia selects its faculty; I left that field eons ago. But this issue has another negative effect: it pushes people to publish non-results just to keep their paper count up. Or worse, they publish the same garbage in multiple journals with just a few words changed. The signal-to-noise ratio in scientific publications is really low.

 

[atyy]

From one of the greats:

http://web.mit.edu/physics/people/faculty/weiss_rainer.html

"Started in physics with the precept that only very important and finished work should be published. As a consequence, didn't publish much and got hell for it. If you are curious how things became compromised, take a look at my publication list"

 

[Vanadium 50]

If a postdoc has published 1 paper as a student, and 0 papers as a postdoc, and 7 papers that haven't been published (15-8 = 7) this is a problem. Of course, the quest for the "least publishable unit" is also a problem, but that's a different problem. This problem is a person who is writing few papers, unable to get them published, and the rest of the community does not find them important.

 

[ferm]

The important question is "does the system make sense" and I don't think it does. The way that the system works just kills any sort of creativity and original work. To give an example, people have been working on string theory for decades, and have got nothing useful. It would be useful for some Ph.D. to at least think about alternatives, but they can't do that in the current system because there is a good likelihood that their alternative just will not work (most new ideas don't), at which point they've killed their chances for a faculty position and are unable to get another bite of the apple.

I couldn't agree more with that! I have the sense that there is a natural selection in academia against people interested in too many things, since what is rewarded is to have many publications in a particular narrow field where you will be making your career. However, many advancements in science have come from people who were working in a somewhat related field.
Many particle physicists look down on condensed matter guys, they forgot too easily that the Higgs mechanism was inspired by condensed matter! And there are many other examples of this kind.

It is curious to see how even people who started working in an alternative to string theory, and that are proud of doing that as if they were somehow "outside the system", have now adopted a new dogma (loop quantum gravity) that they might never abandon, even if it's stuck with no results.

Another funny thing is how people adopt a kind of faith in the theories they are working in. Many string theorists don't have any doubt that string theory is the ultimate theory of the universe. Many people working in supersymmetry are still convinced it is the theory (although we have been so unlucky not having found it yet). And the same happens in theories called "exotics" by the biggest part of the physics community.
Of course not everyone has this attitude, but it is still quite shocking to me how some very intelligent, respected scientists can have this kind of blind faith for something that isn't even a consensus in the scientific community. Of course, sometimes you need that in order to work intensively and alone on a particular, small topic for so many years.

 

[vici10]

twofish-quant is right. The system that drives quite capable people out of science is broken.

Below is a quote from NY Times (2008) http://www.nytimes.com/2008/10/17/science/16prasher.html"

In a couple of months, Roger Y. Tsien and Martin Chalfie will head to Stockholm to collect the Nobel Prize in Chemistry and $450,000 each in prize money in recognition of their development of a revolutionary technique that lights up the inner workings of living cells.

Meanwhile, the scientist who provided the essential piece that made Dr. Tsien’s and Dr. Chalfie’s work possible — a jellyfish gene that produces a fluorescent protein — is out of science.

Douglas C. Prasher, who conducted his research on the Aequorea victoria jellyfish while at the Woods Hole Oceanographic Institution in Massachusetts in the early 1990s, now drives a courtesy van for a car dealer in Huntsville, Ala., earning $10 an hour.

Wikipedia said that in 2010 Prasher found a job in small research contract firm. It is just one example, but it is quite telling.

 

[Vanadium 50]

The system may be broken, but it was never fixed.

Each professor has maybe 10 grad students over the course of his or her career. Only one will replace him or her.*

There was never a time where a PhD guaranteed you a permanent faculty or equivalent position. Good people end up doing something else - there's no way out of that. And it turns out that this is a good thing. A physics degree is something that industries (even banking!) find useful. The problem would be if the only thing physicists were good for was making more physicists.

* This is, of course, an average. Some professors do much better than average, which mean that some do worse. My advisor graduated 9 students: 6 have faculty or equivalent positions, 1 is still a postdoc, 1 went into industry, and 1 changed fields.

 

[ParticleGrl]

There was never a time where a PhD guaranteed you a permanent faculty or equivalent position.

I would argue that for the generation after WWII, ramping funding and dramatic expansion of higher education meant that the overwhelming majority of phd recipients moved directly into faculty positions. Further, the length of the phd itself was shorter. The "postdoc-holding-pattern" didn't emerge until the 70s, when the system began to hit the steady-state outcome of overtraining.

And it turns out that this is a good thing. A physics degree is something that industries (even banking!) find useful. The problem would be if the only thing physicists were good for was making more physicists.

The thing physicists are best for is doing physics. Unfortunately, if you add up the physics jobs in industry, universities, and national labs, its not enough to employ even half of physics phds, according to the APS data.

Sure, physicists get jobs in other fields. But, just because a physics phd gets a job in a field, doesn't mean the job utilizes any of the training. Yes, a physics phd can work as a business consultant. Does a business consultant require a physics phd?

People like to think of the phd as an apprenticeship, but an apprenticeship is an implicit bargain- low pay for skilled labor now, in exchange for a chance at a good job later. The problem is the number of years of low pay keep increasing, and the chance at the job in the field keeps decreasing. Is this bad for the field- probably in the long run. As an anecdote, I know probably half a dozen talented undergraduates who decided not to go to graduate school after conversations with graduate students and postdocs.

 

[Fizex]

The only solution to this problem is to limit the amount of physics PhD's or increase funding to keep up. This won't happen and I think we have gotten off topic.

 

[atyy]

But it's a good thing that bankers, congressmen etc know some high-level physics.

What's not good is if grad students don't know that the chance of getting a faculty position is 0.

But I'm sure grad students being bright enter with their eyes open.

 

[twofish-quant]

The system may be broken, but it was never fixed.

Sure.

And it turns out that this is a good thing. A physics degree is something that industries (even banking!) find useful. The problem would be if the only thing physicists were good for was making more physicists.

Don't disagree.

However...

1) Since it's the case that most physics Ph.D.'s will not end up getting a formal research position, to set up support systems that help Ph.D.'s make the switch and to make it clear that ending up in industry is the *normal* career path. Moreover, the is the message that people that end up with faculty positions are *better* than people that don't.

2) It would also be nice if people whose first job is in industry could switch back and forth between academia. If you presume that people that end up industry are not "worse" than people that don't, then there really isn't any reason why that shouldn't be the case.

What I'm saying is that the message should be given that the post-doc is not the *normal* career path for the Ph.D.

 

[twofish-quant]

The only solution to this problem is to limit the amount of physics PhD's or increase funding to keep up.

You could increase the social status of non-academic careers, and then have better support services for people to move back and forth between industry and academia.

 

[twofish-quant]

People like to think of the phd as an apprenticeship, but an apprenticeship is an implicit bargain- low pay for skilled labor now, in exchange for a chance at a good job later.

So let's stop thinking of a Ph.D. as an apprenticeship.

As an anecdote, I know probably half a dozen talented undergraduates who decided not to go to graduate school after conversations with graduate students and postdocs.

Have them talk to me. Something that happened to me was that I had a phase in which I was discouraging people from entering science, but then I figured out that the maybe I could get the system to work, so I've really stopped discouraging people from entering sciences.

 

[ferm]

The thing physicists are best for is doing physics.

There are many physicists working in fields other than physics (both in academia and industry) that enjoyed a success they couldn't achieve in physics. The interesting thing is that many physicists work and succeed in a wide range of fields, which is not true in many different fields of knowledge.

But, just because a physics phd gets a job in a field, doesn't mean the job utilizes any of the training. Yes, a physics phd can work as a business consultant. Does a business consultant require a physics phd?

I would say for some parts of the work you require skills that a physics phd must have. Otherwise they wouldn't be hiring physics phds. Of course you don't necessarily need to have a physics phd to get these skills, it's just a necessary condition.

People like to think of the phd as an apprenticeship, but an apprenticeship is an implicit bargain- low pay for skilled labor now, in exchange for a chance at a good job later.

I don't see the PhD as a bargain. Sure, you have low pay, but in exchange you have a lot of flexibility. Also, I'm happy to get a degree that will allow me (I hope) to work in a wide variety of fields. I have been studying and working in physics for some years of my life, which I really enjoyed to do, and now I feel I should move on to something different that I also find interesting. I think of that as a very good thing.

However, knowing that I might be able to get a job in business consultancy, or in the energy industry, or finance, or medical physics (to list some jobs my ex-colleagues are doing now), the option of doing a postdoc in a different branch of physics is closed. By changing fields I would be making a terrible mistake, according to academic standards (unless the fields are very closely related, or a completely new field is appearing). On the same logic, it would as well be impossible to return to my field after some years in industry.

In my opinion, the situation only promotes a kind of endogamy that reduces the efficiency of academia. Many great discoveries have been made by merging knowledge from different areas of knowledge, or by people not influenced by the prejudices in a particular field. This fact is well known and becoming more and more exploited in industry, why is it not the case in academia?