Monday, March 29, 2010

Getting a Job in a Small Subfield

As most of you know I am an aspiring solar physicist. Solar physics is a small sub-field of physics, especially compared to things like condensed matter or cosmology. To give you an idea of just how small it is there are only 58 junior members of the solar physics division of the AAS, which means that there are at most 100 graduate students in solar physics. I don't know how many grad student cosmologists there are out there but it seems like there are about 58 pre-prints published in cosmology every day.

Small fields like solar physics allow for a more congenial and casual atmosphere in some respects since pretty much anybody that's been around for a while knows pretty much everybody in the field. Due to the small size, solar physics has developed it's own online newsletter - the "Solar News" - that anyone can submit information to, including job openings. That means that pretty much every job that comes up in solar physics from NSF section chief to post-doc at Western Montana A&M goes through the Solar News.

Out of curiosity I went through the Solar News archive and figured out the number of jobs each year in solar physics from 2005 to 2009. To start I only included the ones in the US and I sorted those into three groups: post-docs, research positions (temporary and tenure-track), and tenure-track faculty positions. Behold the graph:
Over those five years, there were an average of 17 post-docs, 11 research positions, and 5 faculty positions available each year. If we further assume that a third of those 58 junior members of the solar physics division graduate each year (this is probably high but let's run with it anyway), that means that on average there are 19 PhD's minted each year and for those 19 people, there will be 17 post-docs available. And when those 17 post-docs are looking for potentially-permanent positions there will be 11 research and 5 faculty positions waiting for them. That means that on average 84% of those that graduate in solar physics will keep doing research in solar physics - and this excludes those that get jobs in industry or in primarily teaching roles. That's not bad at all.

However there is one other factor. Europe and Asia have become major players in solar physics. I am unable to find data on how many PhD's they produce each year, but I do have data on how many post-doc and potentially-permanent positions they advertise each year. Behold graph #2:
Foreign countries produce a lot of post-docs, but a comparably small number of permanent positions. That means that there is a large influx of solar physicists with a post-doc or two under their belts into the market for long-term positions in the US.

Overall, however, the job outlook in solar physics is quite rosy. There are few people competing for few jobs, which tends to work out pretty well.


  1. Very interesting Nick. I hope the solar job market stays "rosy".

    As for cosmology, I don't remember exact numbers but I saw a talk at a workshop for cosmologists where the speaker broke down the numbers. Basically his point was:

    1. There was enough money floating around between all the major cosmology experiments that finding a post-doc position isn't that bad.

    2. The majority of cosmologists who were willing to stick it out as many as 2-3 postdocs got permanent positions of some sort. (Some only required one postdoc.) Though admittedly, many permanent positions are not professorships.

  2. Joe,

    Your comments lead me to what I think is one of the biggest failings of physics and astronomy education at the graduate level - the idea that a professorship at an R1 school is the ideal career outcome. I think for many people that is really not their dream job, it's just what they know and what everybody else wants.

    I have found that in general researchers at national labs and professors at smaller universities are as satisfied with their careers as those at research universities - and often at far lower cost in terms of hours worked or overall stress level.

  3. I had a professor here at UNC tell us (the astronomy students) that we would never be able to get a good job because we are going to UNC and not some "good", "big name" school like Harvard, or MIT or Caltech (yes we were actually told this, no I am not paraphrasing or exaggerating). Essentially what this professor told us was that our careers were in the trash bin because we were trying to get our degrees from a "low end" school like UNC. Needless to say, all of the astro students were quite put off by that (and it made us wonder what that professor thought about working at such a "low end" institution). Fortunately I have not found any other professors here who hold the same opinion. As a matter of fact, most have told me that almost everyone who graduates from here ultimately ends up working where they want to.

    But before you get too specialized, just remember what Heinlein said:

    "Specialization is for insects."

  4. Full Heinlein quote:

    "A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects."

  5. Ryan,

    I would refer you to a paper on the arXiv ( that actually attempts to use real data to answer the question of whether those from non-"big name" schools stay in astronomy. Their conclusion is that the percentage of PhD's that stay active in research is pretty much independent of where you do your PhD, however those that go to "big name" schools tend to get jobs at other "big name" schools more than those from less prominently named institutions.

    So if you want a job at Harvard, going to Harvard, MIT, or CalTech is really important. If you want a job at UC-Irvine, Colorado, or UNC, it doesn't really matter.


To add a link to text:
<a href="URL">Text</a>