Thursday, September 17, 2015

10 tips for making postdoc applications (Part 2)

This post is part 2 of a series with unsolicited advice for postdoc applicants. Part 1, which includes a description of the motivation behind the posts and tips 1 through 5, can be found here.


6. Promote yourself

This sounds sort of obvious, but for cultural reasons may come easier to some people than to others. I don't mean to suggest you should be boastful or oversell yourself in your CV and research statement. But be aware that people reading through hundreds of applications will not have time to read between the lines to discover your unstated accomplishments — so present the information that supports your application as clearly and as matter-of-factly as possible.

As a very junior member of the academic hierarchy, it is quite likely that nobody in the hiring department has heard of you or read any of your papers, no matter how good they were. They are far more likely to recognise your name if you happen to have taken some steps make yourself known to them, for instance by having arranged a research visit, given a talk at their local journal club or seminar series, initiated a collaboration on topics of mutual interest, or simply introduced yourself and your research at some recent conference.

Organisers of seminar series and journal clubs are generally more than happy to have volunteers help fill some speaking slots — put anxieties to one side and just email them to ask! And if they do have time for you, make sure you give a good talk.

7. Know what type of postdoc you're applying for

There are, roughly speaking, three different categories of postdoc positions in high-energy and astro (and probably more generally in all fields of physics, if not in all sciences).

The first category is fellowships. These are positions which provide funding to the successful candidate to pursue a largely independent line of research. They therefore require you to propose a detailed and interesting research plan. They may sometimes be tied to a particular institution, but often they provide an external pot of money that you will be bringing to the department you go to. They are also highly prestigious. Examples applicable to a cosmology context are Hubble and Einstein fellowships in the US, CITA national fellowships in Canada, Royal Society and Royal Astronomical Society fellowships in the UK, Humboldt fellowships in Germany, Marie Skłodowska-Curie fellowships across Europe, and many others.

The second category is roughly a research assistant type of position. Here you are hired by some senior person who has won a grant for their research proposal, or has some other source of funding out of which to pay your salary. You are expected to work on their research project, in a pretty closely defined role.

The third category is a sort of mix of the above two, which I'll call a semi-independent postdoc. This is where the postdoc funding comes from someone's grant, but they do not specify a particular research programme at the outset, giving you a large degree of independence in what work you actually want to do.

Image credit: Jorge Cham.

When you apply, it is imperative that you know which of these three types of positions the people hiring you have in mind. There is no point trying to sell a detailed independent research plan — no matter how exciting — to someone who is only interested in whether you have the specific skills and experience to do what they tell you to. Equally, if what they want is evidence that you will drive research in your own directions, an application that lists your technical skills but doesn't present a coherent plan of what you will do with them is no good.

Unfortunately postdoc ads don't use these terms, so it is generally not clear whether the position is of the second or third type. If in doubt, contact the department and find out what they want.

Also, even when the distinction may be clear, many people still produce the same type of application for all jobs they apply for in one cycle. I know of an instance of a highly successful young scientist who managed to win not one but two prestigious individual fellowship grants worth hundreds of thousands of Euros each, and yet did not get a single offer from any of non-fellowship positions they applied to! So tailor your applications to the situation.

8. Apply for fellowships

Applications for the more prestigious fellowships require more work — a lot more work — than other postdoc applications. They will require you to produce a proper research proposal, which will need to include a clear and inspiring outline — of anything between one and twenty pages length — of what you intend to do with the fellowship. This will take a long time to think of and longer to write. They will probably require you to work on the application in collaboration with your host department, and they may have a million other specifically-sized hoops for you to jump through.

Nevertheless, you should make a serious attempt to apply for them, for the following reasons.

  • Any kind of successful academic career requires you to write lots of such proposals, so you might as well start practising now.
  • Writing a proposal forces you to prepare a serious plan of what research you want to do over the next few years, which will help clarify a lot of things in your mind, including how much you actually want to stay in the field (see point 2 again!). 
  • You will often write the proposal in collaboration with your potential host department. This makes it far more likely that they will think favourably of you should any other opportunities arise there later! For instance, I know of many cases where applicants for Marie Curie fellowships have ended up with positions in the department of their choice even though the fellowship application itself was ultimately unsuccessful.
  • Major fellowship programmes are more likely to have the resources and the procedures in place to thoroughly evaluate each proposal, reducing the unfortunate random element I'll talk about below. Many will provide individual feedback and assessments, which will help you if you reapply next year.
  • Counter-intuitively, success rates may be significantly higher for major fellowships than for standard postdoc jobs! Last year, the success rate for Hubble fellowships was 5%, for Einstein fellowships 6%, and for Marie Curie fellowships (over all fields of physics) almost 18%. All of these numbers compare quite well with those for standard postdoc positions! Granted, this is at least partly due to self-selection by applicants who don't think they can prepare a good enough proposal in the first place, but it is still something to bear in mind.
  • Obviously, a successful fellowship application counts for a lot more in advancing your career than a standard postdoc.

9. Recognize the randomness

Potential employers are faced with a very large number of applications for each postdoc vacancy; a ratio of 100:1 is not uncommon. Even with the best of intentions, it is just not possible to give each application equal careful consideration, so some basic pre-filtering is inevitable.

Unfortunately for you, each department will have its own criteria for pre-filtering, and you do not know what those criteria are. Some will filter on recommendation letters, some on number of publications, some on number of citations. (As a PhD student I was advised by a well-meaning faculty member at a leading UK university that although they found my research very interesting, I did not yet meet their cutoff of X citations for hiring postdocs.) Others may deduce your field of interest only from existing publications rather than your research statement — this is particularly hard on recent PhDs who may be trying to broaden their horizons beyond their advisor's influence.

Beyond this, it's doubtful that two people in different departments will have the same opinion of a given application anyway. They're only human, and their assessments will always be coloured by their own research interests, their plans for the future of the department, their different personal relationships with the writers of your recommendation letters, maybe even what they had for breakfast that morning.

You can't control any of this. Your job is to produce as good and complete an application as possible (remember to send everything they ask for!), to apply to lots of suitable places, and then to learn not to fret.

10. Don't tie your self-esteem to the outcome

You will get rejections. Many of them. Even worse, there will be many places who don't even bother to let you know you were rejected. You will sometimes get a rejection at the exact same time as someone else you know gets an offer, possibly for the same position. (Things are made worse if you read the postdoc rumour mills regularly.)

It's pretty hard to prevent these rejections from affecting you. It's all too easy to see them as a judgement of your scientific worth, or to develop a form of imposter syndrome. Don't do this! Read point 9 again. 

I'd also highly recommend reading this post by Renée Hlozek, which deals with many of the same issues. (Renée is one of the rising stars of cosmology, with a new faculty position after a very prestigious postdoc fellowship, but she too got multiple rejections the first time she applied. So it does happen to the best too, though people rarely tell you that.)


That's it for my 10 tips on applying for (physics) postdocs. They were written primarily as advice I would have liked to have given my former self at the time I was completing my PhD.

There's plenty of other advice available elsewhere on the web, some of it good and some not so good. I personally felt that far too much of it concerned how to choose the best of multiple offers, which is both a bit pointless (if you've got so many offers you'll be fine either way) and really quite far removed from the experience of the vast majority of applicants. I hope some people find this a little more useful.

Sunday, September 13, 2015

10 tips for making postdoc applications (Part 1)

Around this time of year in the academic cycle, thousands of graduate students around the world will be starting to apply for a limited supply of short-term postdoctoral research positions, or 'postdocs'. They will not only be competing against each other, but also slightly more senior colleagues applying for their second or possibly third or fourth postdocs.

The lucky minority who are successful — and, as Richie Benaud once said about cricket captaincy, it is a matter of 90% luck and 10% skill, but don't try it without the 10% — will probably need to move their entire life and family to a new city, country or continent. The entire application cycle can last two or three months — or much longer for those who are not successful in the first round — and is by far the most stressful part of an early academic career.

What I'd like to do here is to provide some unsolicited advice on how best to approach the application process, which I hope will be of help to people starting out on it. This advice mostly consists of a collection of things that I wish people had told me when I was starting out myself, plus things that people did tell me, but that for whatever reason I didn't understand or appreciate.

My own application experience has been in the overlapping fields of cosmology, astrophysics and high-energy particle physics, and most of my advice is written with these fields in mind. Some points are likely to be more generally useful, but I don't promise anything!

I'm also not going to claim to know much about what types of things hiring professors or committees actually look for — in fact, I strongly suspect that there are very few useful generalizations that can be made which cover all types of jobs and departments. So I won't tell you what to wear for an interview, or what font to use in your CV. Instead I'll try to focus on things that might help make the application process a bit less stressful for you, the applicant, giving you a better chance of coming out the other side still happy, sane, and excited about science.

With that preamble out of the way, here are the first 5 of my tips for applying for postdocs! The next 5 follow in part 2 of this post.

1. Start early

At least in the high-energy and astro fields, the way the postdoc job market works means that for the vast majority of jobs starting in September or October of a given year, the application deadlines fall around September to October of the previous year. Sometimes — particularly for positions at European universities — the deadlines may be a month or two later. However, for most available positions job offers are made around Christmas or early in the new year, and the number of positions still advertised after about February is small to start with and decreases fast with each additional month.

This means if you want to start a postdoc in 2016, you should already have started preparing your application materials. If not, it's not too late, but start immediately!

Applying for research jobs is a very different type of activity to doing research, is not as interesting, requires learning a different set of skills, and therefore can be quite daunting. This makes it all too easy to procrastinate and put it off! In my first application cycle, I came up with a whole lot of excuses and didn't get around to seriously applying for anything until at least December, which is way too late.  

2. Consider other options

This sounds a bit harsh, but I think it is vital. My point is not that getting into academia is a bad career move, necessarily. But don't get into it out of inertia. I've met a few people who, far too many months into the application cycle, with their funding due to run out, and despite scores of rejections, continue the desperate search for a postdoc position somewhere, anywhere, simply because they cannot imagine what else they might do.

Don't be that person. There are lots of cool things you can do even if you don't get a postdoc. There are many other interesting and fulfilling careers out there, which will provide greater security, won't require constant upheaval, and will almost certainly pay better. Many of them still require the kinds of skills we've spent so many years learning — problem solving, tricky mathematics, cool bits of coding, data analysis — but most projects outside academia will be shorter and less nebulous, success will be more quantifiable and the benefits of success may well be more tangible.

If you have no idea what kinds of jobs you could do outside academia, find out now. Get in touch with previous graduate students from your department who went that way, find out what they are doing and how they got there. The AstroBetter website provides a great collection of career profiles which may provide inspiration. 

If after examining the alternatives you decide you'd still prefer that postdoc, great. But at least when you apply you won't be doing it purely out of inertia, and you'll have the reassurance that if you don't get it, there are other cool things you could do instead. And I'm pretty sure this will help your peace of mind during the weeks or months you spend re-drafting those research statements!

Image credit: Jorge Cham.

3. Apply everywhere

It's not uncommon in physics for some postdoc ads to attract 100 qualified applicants or more per available position, and the numbers of advertised positions isn't that large. So apply for as many as you can! It's not a great idea to decide where to apply based on 'extraneous' reasons — e.g., you only want to live in California or Finland or some such. 

Particularly if you're starting within Europe, you will probably have to move to a new country, and probably another new country after that. So if you have a strong aversion to moving countries, I'd suggest going back to point 2 above. 

On the other hand, you can and should take a more positive view: living somewhere new, learning a new language and discovering a new culture and cuisine can be tremendous fun! Even remote places you've never heard of, or places you think you might not like, can provide you with some of the best memories of your life. Just as small example, before I moved to Helsinki, my mental image of Finland was composed of endless dark, depressing winter. After two years here this image has been converted instead to one of summers of endless sunshine and beautiful days spent at the beach! (Disclaimer: of course Finland is also dark, cold and miserable sometimes. Especially November.)

So for every advertised position, unless you are absolutely 100% certain that you would rather quit academia than move there for a few years — don't think about it, just apply. For the others, think about it and then apply anyway. If you get offered the job you'll always be able to say no later.

4. Don't apply everywhere

However, life is short. Every day you spend drafting a statement telling people what great research you would do if they hired you is a day spent not doing research, or indeed anything else. If you apply for upwards of 50 different postdoc jobs (not an uncommon number!), all that time adds up.

So don't waste it. Read the job advertisement carefully, and assess your chances realistically. There's not much to be gained from applying to departments which are not a good academic fit for you.

When I first applied for postdocs several years ago, I would read an advert that said something like "members of the faculty in Department X have interests in, among other things, string theory, lattice QCD, high-temperature phase transitions, multiloop scattering amplitudes, collider phenomenology, BSM physics, and cosmology," and I'd focus on those two words "and cosmology". So despite knowing that "cosmology" is a very broad term that can mean different things to different people, and despite not being qualified to work on string theory, lattice QCD, high-temperature phase transitions etc., I'd send off my application talking about analysis of CMB data, galaxy redshift surveys and so on, optimistically reasoning that "they said they were interested in cosmology!" And then I'd never hear back from them.

Nowadays, my rule of thumb would be this: look through the list of faculty, and if it doesn't contain at least one or two people whose recent papers you have read carefully (not just skimmed the abstract!) because they intersected closely with your own work, don't bother applying. If you don't know them, they almost certainly won't know you. And if they don't know you or your work, your application probably won't even make it past the first round of sorting — faced with potentially hundreds of applicants, they won't even get around to reading your carefully crafted research statement or your glowing references.

Being selective in where you apply will save you a heap of time, allow you to produce better applications for the places which really do fit your profile, and most importantly leave you feeling a lot less jaded and disillusioned at the end of the process.

5. Choose your recommendations well

Almost all postdoc adverts ask for three letters of recommendation in addition to research plans and CVs. These letters will probably play a crucial part in the success of your application. Indeed for a lot of PhD students applying for their first postdoc, the decision to hire is based almost entirely on the recommendation letters - there's not much of an existing track record by this stage, after all.

So it's important to choose well when asking senior people to write these recommendation for you. As a graduate student, your thesis advisor has to be one of them. It helps if one of the others is from a different university to yours. If possible, all three should be people you have worked, or are working, closely with, e.g. coauthors. But if this is not possible, one of the three could also be a well-known person in the field who knows your work and can comment on its merit and significance in the literature.

Having said that, there are several other factors that go into choosing who to get recommendations from. Some professors are much better at supporting and promoting their students and postdocs in the job market than others. You'll notice these people at conferences and seminars: in their talks they will go out of their way to praise and give credit to the students who obtained the results they are presenting, whereas others might not bother. These people will likely write more helpful recommendations; they also generally provide excellent career advice, and may well help your application in other, less obvious, ways. They are the ideal mentors, and all other things being equal, their students typically fare much better at getting that first and all-important step on the postdoc ladder. Of course ideally your thesis advisor will be such a person, but if not, find someone in your department who is and ask them for help.

Somewhat unfortunately, I'm convinced that how well your referees themselves are known in the department to which you are applying is almost as important as how much they praise you. If neither you nor any of your referees have links — previous collaborations, research visits, invitations to give seminars — with members of the advertising department, I think the chances of your application receiving the fullest consideration are unfortunately much smaller. (I realise this is a cynical view and having never been on a hiring committee myself I have no more than anecdotal evidence in support of it. But I do see which postdocs get hired where.) So choose wisely.

It is also a good idea to talk frankly to your professors/advisor beforehand. Explain where you are planning to apply, what they are looking for, and what aspects of your research skills you would like their letters to emphasise. Get their advice, but also provide your own input. You don't want to end up with a research statement saying you're interested in working in field A, while your recommendations only talk about your contributions in field B.


That's it for part 1 of this lot of unsolicited advice. Part 2 is available here!