How hard is it to switch fields?

Diracula

How hard is it to switch fields in science? For example, say you do your Ph.D. in an area of theoretical physics and are very successful, but cannot find further funding at some point (can't land a postdoc in your chosen field, no professorship offers, denied tenure, etc.). How difficult would it be to change to a field where there is more funding available? Can you fairly easily switch to another area of physics? What about an area outside of physics? Bioengineering, systems biology, electrical engineering, chemistry?

I hear conflicting stories on how difficult this is. Is there any hard data?

ZapperZ

I obtained my Ph.D in experimental condensed matter physics, and then switch to accelerator physics after my postdoc. I would say that, in physics, it isn't that unusual.

However, the switching is more desirable if the new field you go into actually has a strong need for your original skills. In my case, my expertise in photoemission spectroscopy and material properties are what exactly was needed when I was offered a job in accelerator physics. Don't get me wrong, you have to do A LOT of catching up to get up to speed. Hopefully, your training in pursuing your Ph.D has taught you how to learn things.

But as far as doing theoretical work, that I have no idea.

Zz.

swaps2000

I guess this is a new generation people can persue multiple careers in there life . Eg . 2/3 years may be just needed to read parallely while in current field . Once you get the basic and primary concepts ..just switch to field. Take up tutions etc in subjects of new field . We can teach what we learnt. Also concept learning is very simple in this generation with techinques like audio/vedio learning, imagination,speed reading, memory maps, online notes,internet chatting and clearing doubts, pdfs etc...so In 2/3 years we can learn the concepts while we are in current field and then once we are through we can immediately switch fields..

Mike H

As the joke goes, the more interdisciplinary a field is, the more likely it is to have refugees from all over the scientific community.

Insofar as chemistry, at least - I would imagine you'd have a fairly straightforward time fitting into a theoretical/computational chemistry group. On the experimental side of things, it'd be far trickier, and in some cases, I'd see it as so improbable to boggle the mind. I have a Ph.D. in chemistry, and even I'd have a hard time reinventing myself as a synthetic chemist at this point in time. But experimental p.chem and the laundry list of "chemical biology" labs out there would probably be the most likely avenues for you in that case.

I would certainly echo ZapperZ's point about the switch being easier if the new field has need of your old field's skills - I know a couple of PhD physicists with backgrounds in magnetic resonance or neutron/x-ray scattering who are doing interesting work in materials science, biology, and so on nowadays.

ParticleGrl

In an attempt to make this sort of switch myself, I talked to about half a dozen former particle theorists who had switched to various other specialties. The common thread seems to be that all of them managed to find a growing area that had a legitimate shortage of qualified people, and figure out a way their skills made them desirable for such work. ZapperZ can correct me if I'm wrong, but I think his switch seems to fit that general mold.

Shortages in science are very temporary, however. i.e. not that long ago there weren't many biologists with the sort of mathematical training needed for bioinformatics. Now there are bio-math and bioinformatics phd programs. So I imagine its hard to plan ahead of time to switch.

I don't think anyone can know what the odds of being able to switch are ahead of time. If you happen to graduate during a recession, the odds certainly aren't in your favor- most fields will have an over-abundance of qualified applicants who aren't trying to switch. If you graduate during a boom, it may be relatively easy to switch. The key needed ingredient seems to be a newly "hot" field that needs some skills you have- this is largely outside your control.

Diracula

Are postdoc positions generally saturated with qualified candidates in a recession? Or are only the more highly desired permanent positions (tenured or at least tenure track professor, permanent industry R&D job, government lab) saturated with qualified candidates?

How would a newly graduated theoretical physicist fair in a poor economy when applying to engineering, chemistry, bioinformatics, synthetic biology, etc. postdocs?

My feeling is that it would be much easier to switch fields via the postdoc route rather than trying to land an assistant prof or industry job that pays 80k+. If true, this strategy would open up a lot more avenues for real R&D jobs to physicists that can't land that tenured professor job in their Ph.D. specialty.

ZapperZ

From my experience (and I've been involved in hiring 3 postdocs the past 2 years), there's a LOT of qualified candidates in a particular field of study. So someone else from a different background will have a severe disadvantage, considering that you are expected to land on your feet and do top-notch work the moment you are hired. You have almost no time to learn the new field, and you're not expected to either.

The only exception that I can think of is, as mentioned earlier, a field that is interdisciplinary. In detector physics, for example, they often require people with many different backgrounds and skills.

Theoretical physicists, I would think, face an even tougher challenge. Unless the skill that you have can smoothly transpose itself into another subject area (i.e. such as computational skills, or that the mathematics that you are familiar with in your original subject area are similar to the one in another field), then I am not sure to what extent you are desirable in another field.

Zz.

Choppy

I don't know of any hard data on the matter in particular.

Obviously it happens though. Lots of people are able to jump into other fields for whatever reason, and do it quite successfully.

As far as advice for a student planning his or her career goes, I would strongly suggest working to develop a marketable skill set from day one, rather than relying on the hope that if your pursuit of your primary goal doesn't work out you could 'always just switch' into a more lucrative field.

twofish-quant

I didn't find it very hard to switch from academia to industry from the point of view of getting a job. It was extremely difficult psychologically, but that's easy to deal with if you "pre-brainwash" yourself. I've known a lot of physics Ph.D.'s that have ended up teaching university courses in business/finance departments.

Also things that seem unrelated often are. My entire career can be seen as numerically solving parabolic partial differential equations. So even though it looks like I'm doing a radical career change, I'm really not since I'm just solving the same equation in a different field.

FYI, in theoretical physics, you need two post-docs before people will even consider you for tenure track.

Doesn't seem to be the case in finance. Even in the worst days after the crash, physics Ph.D.'s were still getting Wall Street jobs. One difference between industry and academia is that a lot of the positions in industry involve essential jobs. If you lose funding in academia, you can just hire fewer post-docs. By contrast, you have a supercomputer that needs baby-sitting, and you *must* find someone to babysit that supercomputer if you want to stay in business at all.

twofish-quant

It's not a matter of "can" but "must." I've had to make three major career changes so far, and in none of those cases was the change entirely by choice. I very, very seriously doubt that I'll be doing what I'm doing in five years.

This is not necessarily a good thing. I suppose part of it is that I'm getting older, and the prospect of having to redo my entire life every few years is getting less and less appealing. The other problem is that the higher you are, the more room you have to fall.

Also a lot of this involves "marketing issues." Getting yourself up to speed in a new field is nowhere as difficult as convincing an employer that you are competent in this new field.

twofish-quant

And finance is the ultimate in interdisciplinary, and yes there are a ton of science refugees. The one catch is that you have to move to New York City, London, Hong Kong, or Singapore. Trying to do finance outside of a finance center is like trying to do accelerator physics somewhere without a big accelerator.

I've actually never had much of a problem with this. There is the essential "one-ness" of the universe in which the same cosmic principles find themselves in different fields. You just have to see the connection. There is also the "Chinatown" factor.

One reason you have a ton of astrophysicists in finance, is that once you have a few astrophysicists working in finance, they will change the culture and the mathematics of the field to resemble astrophysics. It's very much like an ethnic community like Chinatown or Little Italy. Once you have enough refugees in a field, they will create a community that resembles the "old country" and provides a buffer to the new country.

Getting back to *why* finance happens in a few cities. It turns out that there are cities in the world that are designed for field switchers. If you talk to just about anyone in NYC, you'll find that before they became a taxi driver, janitor, quant, waiter, they were something else back home, and they ended up in NYC because they need to do something radically different with their life.

mal4mac

At the start of my career I moved from computer modelling of physical systems to programming human computer interfaces in a university research environment. The guy in the desk next to me had dome some programming in the context of experimental astronomy. It was never difficult finding a job in another field, as long as the job I applied for involved programming

If you can program seriously well in Fortran, or anything else, I guess, then you've a good chance of getting involved in any kind of research that demands programming skills. Many seriously competitive programmers go into fields like finance, leaving lots of research-oriented places in academia for those who like a more relaxed life, with less money :)

So if you are doing theoretical physics then try and do some programming - it doesn't have to be all programming! But just being able to put some evidence of some serious C++ programming on your CV should open a few doors.

If programming doesn't agree with you, then you might have a problem. As a last resort there's always IT support... that's a lark!

mal4mac

If the "newly graduated theoretical physicist" has programming skills she could just look for a job in these areas, apply and get it :) Just go to http://www.jobs.ac.uk, or your country's equivalent, type "programming" into search and see what comes up. I quite fancy going into bioinformatics, never worked in it before. But, a quick search, and here's something I could have applied for even as "newly graduated":

http://www.jobs.ac.uk/job/ADX243/sof...are-developer/

"Recent graduates or current students in their final year are especially encouraged to apply, although it is essential that any candidate has substantial development experience outside of coursework (e.g. active participation in open source projects or the pursuit of personal coding projects). An interest in scientific research and a strong intellectual curiosity are also necessities, although formal education in biology or genetics is not required."

"Our philosophy is that development skills are independent of language and that programming languages themselves aren't something that need to be formally taught. That said, a successful applicant will likely have developed expertise in at least a few languages as a side-effect of previous development projects, and should be able to demonstrate skill in at least one scripting/interpreted and one compiled languages. A successful applicant will have a deep understanding of core computer science concepts and practices (e.g. abstraction, data structures, encapsulation/object oriented design, testing, complexity in time and space, agile development methodology)."

This is a good job advert! They make explicit what employers like this are looking for... they don't always do that...

So get involved in open source projects in your spare time! That's the way to get experience, if you don't get to do much programming at university.

I guess the main thing is to develop a skill, or skills, that are in great demand in many different fields. Programming is such a skill.

mal4mac

This is the main reason the theoretical physicist can easily get into new fields like bioinformatics, HCI, etc, in universities. These departments are full of refugee physicists, so if you apply to the department you are going to find a lot of your own kind there. My first employer in HCI was a former physicist who had written a lot of code in Basic 'cause he knew Fortran and Fortran is quite similar to Basic. So when I turn up, a physicist knowing Fortran, it's a case of like mind meeting like...

ZapperZ

Bad analogy. One can do quite a bit of accelerator physics without a big accelerator. University of Maryland is one of the best universities in producing accelerator physicists. Yet, heard of any "big accelerator" on that campus?

Zz.

atyy

There's no more chinatown in San Francisco - it's the entire city now. Does that mean astrophysicists are going to Occupy Wall Street?

twofish-quant

I think we already have.

Anyway, one sort of amusing thing is that there is very little financial activity physically taking place at Wall Street. It's all on the internet, and the big banks in NYC are all headquartered in Midtown. In fact, it looks like one of the unintended side-effects of OWS has been to *decrease* protest activity near the big banks, since everyone is downtown.

Also, people seem to have thought of this before. Something that is interesting about Midtown is that there is no easy place to hold a protest, and I have this suspicious that someone either consciously or unconsciously designed the blocks a hundred years ago to be unfriendly to protesters. Certainly modern office design has taken this into account, you see a wall which seems out of place, until you have a large protest at which point it seems very conveniently placed.

This applies to cyberspace. There are walls, barriers, checkpoints in cyberspace that mold human networks. What usually happens when you have different fields is that there is some sort of invisible wall or invisible line. Astrophysicists think a lot about how things move through space and time, and that comes in handy when the topic of discussion involves moving something (people, ideas, money) through cyberspace.