How Do I PhD in Physics/Astrophysics?
May 23, 2016 11:07 PM   Subscribe

I have recently gotten rejection letters from all of the PhD programs to which I applied for the coming fall. Now I'm back to square one and I think I need to figure out what's wrong with my approach so far. Basically I want the hive mind to tell me exactly what steps it thinks I should take to find the right school and make my application competitive. Can you help?

First, let me say that I understand there are many out there who may look at the details below and say doing a PhD is probably a mistake, a waste of time, gruelling, won't help my job prospects, etc. Let's take that as read if we can and move on.

Now, about me. I have a Bachelor of Science in Physics from Rowan University. I'm a 4.0 (okay, 3.94) student. I graduated summa cum laude. I'm very interested in just straight up classical general relativity (more on that later), but also cosmology, and some particle physics. I have undergrad research experience, but only in materials science, nothing related to astrophysics/astronomy. This is my second attempt at getting into a PhD program in physics/astronomy/astrophysics. See the link if you want to know what happened the first time. The short version is I got into two schools I wasn't very excited about and decided not to go to after visiting. Instead I got an internship doing detector characterization at NASA Goddard. That has since turned into a full time job. Yay for me but I still want to keep going with academia. So, I'm going to do this thing all over again next year, but I want to get started early this time and do it right (3rd time's a charm?).

There's really not anything I can do about my academic or research record. My current work is certainly related to astrophysics but involves no actual research. That is I'm not going to be producing any papers, nor would I know how if I wanted to.

I think the thing I need help with is the more ephemeral/social parts of finding/picking a school and making myself attractive to a professor/department. For instance, today I got the opportunity to meet a string theorist and I told him the short version of what I'm telling you all and he said something along the lines of, "General relativity? No, don't say that. Tell them you're interested in particle physics.". What? He explained how no one's doing general relativity anymore and it's been subsumed under other research areas for the most part, but how would I ever have known that? I think this illustrates how entirely naive I am about what even the landscape of physics research is and how I find my place in it.

So, how do I:
1. Find out what kind of research I should even be telling them I want to do? Is astrophysics too broad? Is sub-millisecond pulsars too specific?

2. Find out who them are? How do I identify which schools, research groups, professors are doing the kinds of things I'm interested in AND are accepting new students AND I have a chance of getting into. Difficulty: I am absolutely terrible with names. I literally forget the names of non-blood relatives I have known for years at holiday time. That string theory guy? Met him about 12 hours ago. No idea what his name is.

3. Give my application the best chance of getting through the process? I assume the only real things I have control over are essay portions and recommendations. I show my essays to former professors, friends who are good at writing, and colleagues and they basically say "these are fine".

I really need specific concrete examples here. For example, a suggestion of "read current research" in reference to 1 and 2 is too vague. I have an ArXiv feed sent to me daily, but it's just a firehose of papers I have no idea how to parse or assign importance to. Another unhelpfully vague suggestion would be, "talk to people in your field". Who? Where? How do I find them (see above). What should I be asking them about? Something like "go to this website, look for papers that have a blue star next to them and contact the fourth author listed on a Thursday" is more my speed. Sorry if I'm asking for a spoonfeeding, but baby needs a PhD. Thanks in advance.
posted by runcibleshaw to Education (42 answers total) 6 users marked this as a favorite
 
What do your GREs look like? Can you improve them?
posted by kickingtheground at 11:19 PM on May 23, 2016 [1 favorite]


Have you reached out to the professors who you are interested in working with? You need to do that.
posted by mr_roboto at 11:34 PM on May 23, 2016 [3 favorites]


Find out the top five programs in the fields that you're vaguely interested in. Navigate to the faculty list on their website. Go through the papers and books that they've written and their current research activities until you find something that you want to study. Yes, all of them. Contact those people.

The advice in your last question was largely "Don't go to grad school until you have a better idea of what you want to do." It seems like you still haven't done this and could be why you're getting rejected.
posted by youcancallmeal at 11:37 PM on May 23, 2016 [14 favorites]


Do you have a research question that you want to answer, rather than just a topic area within which you want to research? In my field - which is very different from physics, so take this with a pinch of salt - that would be an important factor distinguishing you from the run of applicants: having a specific project/question/aim/curiosity to contribute to the field, and not a generalised interest in the subject + desire for the degree. What do you want to do, concretely, during your PhD? How do you imagine yourself spending your day?

In selecting papers to read within your general area of interest, I would suggest following what speaks to your personal curiosity and fascination, given your current expertise, and taking it from there in defining your PhD project. I do think it's a good idea, if possible, to go into applications with a project ("I want to solve/understand/explain y about x") - however half-baked and open to correction - rather than with just undergrad-level willingness to learn ("I am interested in x").
posted by Aravis76 at 11:50 PM on May 23, 2016 [4 favorites]


Why are you listening to a string theorist for advice on how to get into astro grad school? The advice that string theorist gave is one part out of date and one part irrelevant to you. (I am a string theorist too, so maybe you shouldn't listen to me either..). It sounds like your string theorist gave advice relevant for their sort of theorist.

Numerical GR is a pretty active research area (especially with LIGO). And if you want to do astro, saying you want to do particle physics (unless you are interested in astroparticle) is just bizarre. You should say you want to do what you actually want to do.

Note physics PhD programs are not like many humanities and social science programs; it may not be possible for you to know what sub-subject or question you want to work on before you go. Yes, there is a research statement, but your grades, scores, and rec letters matter more.

That having been said, can you talk to researchers who are actually doing the sort you want to do? It'll be more useful than talking to random string theorists.

Do the programs you are applying to ask for physics GREs? If so, improve those scores. This can be done with targeted practice (I.e. Break out your intro series books and do every single problem, training for speed. Learn some basic standardized test skills such as eliminating answers etc, if you eliminate enough guessing among the remainder will still help you on average). It's stupid and I think that exam is stupid, but especially coming from an less known place, a good Physics GRE score can help. (The issue may be that nobody knows what your GPA from your school actually means). I don't like the test, I don't think it tests much that is useful for actual grad school, but until admissions committees understand this, up your scores.
posted by nat at 1:23 AM on May 24, 2016 [7 favorites]


There's probably a massive glut of eager, bright-eyed bushy-tailed undergraduates swamping everything having to do with GR, with the GR wave results. If this is the case, programs are forced to be very very picky and it could be you may only have a very slim chance no matter how polished your essay and recommendations are. Even if you're a good, self-motivated, keen student. This may be what happened to you.

I'm guessing, as I'm not in physics any more but I think my reasoning holds up.

20 years ago, I had the vague sense that my fellow undergraduates were focusing on ... GR, solar cells, and ... generic particle physics? Trendy stuff that they could relate to. And we were fuzzy on the first and last, since we hadn't done coursework on GR or math-heavy particle physics, so our vague expressions of passionate interest are sort of pro-forma from the point of view of the profs on the departments admissions committee. "Eager, likes generic GR, let's admit them, and we'll see how they do two-three years from now once they've done necessary grad coursework." (Repeat 10 times, then the program's full.)

And then when they got into grad school, that crowd got dispersed into slightly random slots that weren't always what they were looking for but it all worked out for the keeners, if they got half-decent advisors. And you sound keen.

I think students going into programs in materials physics etc may have had clearer goals. And admissions committees like that. (And there's a lot of jobs in industry.)

I think there may be a mini GR boom in the next couple years, but that may mean 20 more grad student slots a year and 1000 more applications. So. More butts than seats, and if there's 400 brilliant deserving students in those applications, 380 of them get inadvertently screwed by resource limitation issues.

Also, there may not be a job market for the GR PhDs who don't get jobs as academics. To be very blunt. Which might mean your 40s would suck if you don't win the lottery and get a tenure track professorship, and instead you're an adjunct teaching E&M at two institutions for < liveable wage.

Note that this is guesswork on my part.

Would you enjoy doing materials science / detector development / related in grad school? If so, consider that. Talk to your mentors about it a bit.

If you think materials science / detector development/related would suck, don't that.

That string theory guy? Met him about 12 hours ago. No idea what his name is.

Write things down.

For example, a suggestion of "read current research" in reference to 1 and 2 is too vague.
Some physicists blog. Follow one? e.g. http://www.quantumdiaries.org/

Also, and I'm spitballing here, but it may be GR is a dense subject where you have to do a year of grad-level GR coursework in order to get anything out of the theory papers, especially with no context yet. Likewise a lot of the stuff with feynman diagrams. Try strongly focusing on stuff that's been contextualized in a blog, and reading review articles. And start working through Arfken, or something horrible like that.

Also, try following a material science / detector /etc blog and see if you develop warmth for it. That stuff also changes the world after all. And you could be a perfect fit for material/detector stuff. If you loved it.

Again, I'm not in academia or physics any more, so I may be wrong about this stuff.
posted by sebastienbailard at 1:55 AM on May 24, 2016 [1 favorite]


Are all of your letters of recommendation from Rowan faculty? It can be helpful to have a letter from somebody in a higher profile position. Is there somebody well known at NASA who could write a positive letter about you? I am a faculty member at a school that lists Rowan as a peer institution, but I do research (with undergraduate students) at a prestigious research institution nearby, and when my colleague at that institution (very well known highly accomplished scientist) writes a letter for my students, it makes a big difference.
posted by Seymour Zamboni at 3:12 AM on May 24, 2016 [2 favorites]


The questions you ask would typically be addressed by your faculty advisor in your major or professors in your area of interest, especially if you attended an undergraduate institution or department that sends many of its graduates to Ph.D. programs, and also has an active research program with many graduate students of its own. An academic researcher active and current with their field knows who is doing what where, what the hot topics in the field are, and can even put in a word for you with faculty members at schools you are interested in applying. They would tell you where to apply and how to present yourself in your application materials. This "spoonfeeding" of the "ephemeral/social parts" of how to join the academic enterprise, the first step of which is getting into a Ph.D. program, is something that happens and which is done by faculty members for their students.

Since you are now at Goddard, I would look for one or more mentors there who can play a similar role for you - my own memory of Goddard is that it was full of Ph.D.'s from various disciplines, both Goddard employees and visiting researchers from various universities. And there were also current Ph.D. students, full-time ones at Goddard on temporary research stints, or Goddard employees who were pursuing Ph.D.'s part-time. Talk to them, ask them what they do, write their names down, saying you'd like to read their papers, or ask them what academic papers they have been reading recently, or would recommend you read. You might even end up working with them or otherwise helping them out in their research - if that happens, great, you have additional people you can ask for letters of recommendation!

And if you haven't already done so, I suggest reading Getting What You Came For: The Smart Student's Guide to Earning an M.A. or a Ph.D., albeit with a grain of salt. Some of the advice is dated, and some would say it is too cynical or negative, but there is a lot of good advice there which may be helpful as you strategize about how to apply and be admitted to your choice of graduate programs.
posted by needled at 4:13 AM on May 24, 2016 [5 favorites]


how about doing a masters first? that would give you more of the intermediate knowledge you seem to be lacking?

also, and i think this is very much a cultural thing, so it may not be relevant, but when my partner (astronomy prof u chile, sometime head of enrolment) talks about selecting students what she cares about is students that are self-motivated and smart. people who will finish a phd and do it well. the idea that a recent grad somehow needs to know exactly what to study for research is kind of weird.

related to that last point - have you considered applying abroad?
posted by andrewcooke at 5:09 AM on May 24, 2016


Many programs won't accept you unless your GRE is very good. Often the cut off is missing more than three math questions. That's a firm cut. Study middle school math, get all but three questions right. Check.

Look on the internet for groups you want to work in. Like put university relativity theory into the internet and find specific professors. Contact the prof directly by email. Contact the department recruiter as well. Some profs don't answer their email because they get 300 emails a day. So if the recruiter can get you a phone call, even better.

If no one is funding the research you want to do, times will be bad. I don't know this about physics but funding areas do entirely dry up in my field. Ask these profs you're contacting about it. It's possible you could TA every semester for five years to get into a project you want. That's a hard road. No funding means little interest, and can make papers harder to publish because the area is not "in" right now. But not impossible. It's worth it to do what you love, but know what you're getting into.
posted by Kalmya at 5:11 AM on May 24, 2016 [2 favorites]


What sort of places have you been applying to? Are you shooting too high? (Do you have to for the job to be useful? I'm not a physicist either, but I am involved in graduate enrollment in my program, mathematics. )
posted by leahwrenn at 5:13 AM on May 24, 2016


I have a PhD in Astrophysics and work at a national laboratory. I also had a job after my B.S. (at Fermilab in particle astrophysics), so I understand the situation that you're in.

My advice, since you want some really specific concrete examples:

The vast majority of graduate students that are attending top 10-20 physics programs are there because they know someone. Yes, even over applicants that had very good grades and GRE scores.

You could ask some of the people that you work with if they know someone at LocalStateU that you could talk to about attending their PhD Physics program while working part time at Goddard. The thing is, graduate schools don't want to let in anyone that they don't believe will complete the program. If you can show that you are already set up to do research that you could turn into a thesis, you have a leg up. If you know people who know people at the university, you have a leg up. If all you have been doing is visiting universities and sending in applications, well yes, people do get into programs that way, but it's not necessarily the best way to get what you want.

Special note about physics PhDs: it's not necessarily true that you need to be involved in your prospective department's specific brand of research to be admitted. My experience has been that they care more that you have a clear path through the program, and that you have acquired the resources to make it happen. This may mean that you have applied to fellowships and grants. Wining a NSF transferable fellowship, for instance, is a ticket into almost any graduate program that you want, since you would essentially cost the school nothing, and also have shown that you are skilled enough to win a fellowship.
posted by Shouraku at 6:45 AM on May 24, 2016 [1 favorite]


Okay, so first off, your GPA is fine, obviously, the fact that you actually have experience now doing real physics stuff is great!

But yes, I'd second the advice about making sure your GRE general/subject test scores are high enough that they won't disqualify you. But I wouldn't stress about, say, getting higher in the 90th percentile on the physics subject test if you're that good. I think a lot of professors already know that test performance is not very well correlated with graduate student performance.

Also, I don't think anyone's said it explicitly yet, and while maybe you already know it, it's something I really wish I had known when applying to graduate school: Graduate admissions essays are not college admissions essays. You are not trying to convince them you are an interesting and well rounded person. You are trying to convince a bunch of professors that you are super excited about research and are willing to spend a significant chunk of your time focused on a narrow research topic. More specifically you want one professor to look at it and think "I have funding. I need a graduate student. Hey, this person is interested in what I do! Hey, they actually have skills doing something that I need to get done! Let's accept them!" And even better if they think "Oh, that's that person who I talked to who was super interested in my stuff!"

The good news is you do have specific interests (sub millisecond pulsars is not too specific) and you do have skills - you've been characterizing detectors. Astrophysics uses a lot of detectors. So you need to figure out where that stuff is going to be the most useful. What sort of detectors do you work on? What science are they being used for? Is it part of a collaboration? What other professors at what universities are part of that collaboration? What sort of science can be done with that or a similar type of detector? Who's doing it? Does anyone you even remotely work for/with know those people?
posted by Zalzidrax at 7:08 AM on May 24, 2016 [1 favorite]


Hi again.

From your stated interests, I can't tell if you're interested in Astronomy or Physics. In the previous thread, you said you were interested in exoplanets, high energy astrophysics, and black holes, but now you list your interests as GR, Cosmology, and particle physics. To me, it looks like you're realizing you're more into Physics than you are into Astronomy. I mean, they're related, but even the type of astronomy you want to do is usually studied in astrophysics groups within physics departments.

So, some concrete advice: Make your Physics GRE as good as it can be. Good general GRE scores are fine, but the physics GRE is much more important. If you're not scoring above, say, the 50th percentile on the physics GRE, you have basically no chance of getting into any even reasonable physics grad program. You need to be above the 70th or 80th percentile to get into a good program. As a side note, your general GRE scores don't help your application, but they can hurt it. If you're not getting 90th percentile or above on the math GRE, that's likely to throw up a big red flag.

Go to every one of these talks. Looking at that list, I am supremely jealous that you have those talks where you're working. I mean, Wendy Freedman talking about Cosmology?! If you're not going to those talks, you're making a huge mistake. Write down names of the people talking, but also of the big, important people that the speakers mention. It also appears that there are archived videos of old colloquia; watch the ones that are closest to your areas of interest.

And, yes. I agree with those above. Being bad with names is not an excuse for not jotting down names in a notebook or on your phone.

To answer your specific questions:

1. You should tell them you're interested in whatever you're interested in. Don't listen to the string theorist. Given what you have expressed as your interests, I would say something like, "I have broad interests, but I think I'd like to focus on GR. A recent result that really excited me is XXXXX."

2. The way you get to know people is by putting yourself in situations where you can meet them. So, 1) read papers - you should look at the ArXiv every day, read the abstracts, and look in more detail at the papers that look interesting; don't worry if most of it flies over your head initially - it will get better. 2) Go to every one of these talks (yes, I'm mentioning it again) 3) Go to conferences in the field if you can.

3. If you're not already, get recommendations from your boss at NASA Goddard. She/He is going to be able to best talk about your abilities to do research, which is the most important part of graduate school. And, yes. Detector characterization is a form of research. You're not asking your own questions yet, but you're working on ways to figure others' answers.
posted by Betelgeuse at 7:43 AM on May 24, 2016 [1 favorite]


A few things that have been said in here and in your other graduate school threads...

If you were my pal, I wouldn't recommend that you do a PhD at your age. I know that you say that you don't care about money, but you do need enough to live on. I am one of the lucky lucky people with a tenure track job and I fear that anyone starting in this line of work even slightly later in life is not going to make enough money for retirement. I became a faculty member in my early 30s and with the hit that I took in saving for retirement in my 20s while in grad school is haunting me and will continue to haunt me. And I am so afraid for my many many many PhD classmates that have not found permanent work. (And I was in a top 3 program in my sub-area.) They are taking on tons of personal debt and cannot find "normal" work like food service, retail or an office job because the PhD signals something to those employers.

But you need to figure out what the top journals are for your sub-area (many the disciplinary association has a list? Ask your undergrad professors?) and read the last few years of each of them cover to cover. Learn who the big researchers are. When you like an article, read everything else by that person. Find out where they are faculty and read everything by everyone in their department. Go to conferences. Listen to talks. Network.

But truly, a lot of admissions are done by personal connections. When I read a letter of recommendation from someone that I know/respect, that means something. Your undergrad professors' networks are really important.

Like everyone else, I want to firmly say that I don't think that it is a good idea for you to get a PhD.
posted by k8t at 8:47 AM on May 24, 2016


I'm an experimental physicist who sits on the admissions committee at a place that's pretty highly regarded in physics (read bottom end of the top 20 most years). We don't have astronomy faculty, rather we have astrophysicists which are part of the theory tract, so I can't speak to what is necessary for pure astronomy.

There's a big difference in the admissions criteria for the experimental and theoretical students. We divide our admissions between the experimental applicants and the theory applicants, with the former group being much larger - last year's group was about 25 experimental, 7-8 theoretical - the latter number chosen because they actually needed 4 and typically about half of the theory students switch to experimental after they don't perform all that well in the coursework. We have the theory people on the committee select the theorists. From what I have seen in theory you need to have a GPA in math and physics above 3.5 (seriously, I've seen single Cs used in a math class as a reason to reject someone). You need >90th percentile on your math AND your subject GRE. And you need good letters, preferably from someone that someone on the committee knows and respects. These standards are lower if you're female, and even lower if your an underrepresented minority applicant, as the committee tries to accept a diverse student pool.

So you have the grades, seemingly, but what about the other two? At Goddard you should be able to develop a relationship with someone who can write a good letter. And if your GREs are too low take them again and spend time preparing for them.

And yes, go to those talks that Betelgeuse mentions.
posted by overhauser at 8:51 AM on May 24, 2016


What do your GREs look like? Can you improve them?
Huh, I sort of forgot about the GREs. My scores are not great. My general GRE is 96th percentile verbal and 64th percentile quantitative. I got in the 39th percentile on the Physics GRE after taking it twice. I'm not sure I can actually improve my score. Both GREs seem to me to be testing how fast you can do math, and the answer for me is: not very fast. I spent months taking practice tests for the physics GRE, trying to improve my score, using tricks I read online, and 39th percentile is my improved score. I'm willing to do it again, but from my perspective it seems fruitless.

Have you reached out to the professors who you are interested in working with? You need to do that.

Mhmm... if you'll read my post you'll see that I'm having a hard time even finding who those people are. In addition, the few professors I did get in contact with basically told me that I'd have a couple of years before I picked a professor/group to work with. Perhaps they were blowing me off. I don't know.

Find out the top five programs in the fields that you're vaguely interested in.
Do I just use the US News & World Report List?

The advice in your last question was largely "Don't go to grad school until you have a better idea of what you want to do." It seems like you still haven't done this and could be why you're getting rejected.
I don't understand how I'm supposed to know exactly what I want to do without already having some experience doing it. I know there are people who somehow know exactly what they want to research and have a proposal and all that but I don't understand how I would do that having no previous research experience in the area I'm interested in.

Do you have a research question that you want to answer, rather than just a topic area within which you want to research?
No, and this part feel very catch22 to me. How am I supposed to know what specific research question I want to answer if I've never done any research in the field I want to go into?

Why are you listening to a string theorist for advice on how to get into astro grad school?
Same reason I'm asking a group of strangers on the internet. His point was that if you do particle stuff that you also learn GR along the way.

Note physics PhD programs are not like many humanities and social science programs; it may not be possible for you to know what sub-subject or question you want to work on before you go. Yes, there is a research statement, but your grades, scores, and rec letters matter more.
I seem to be getting conflicting advice here.


The questions you ask would typically be addressed by your faculty advisor in your major or professors in your area of interest, especially if you attended an undergraduate institution or department that sends many of its graduates to Ph.D. programs, and also has an active research program with many graduate students of its own.

The only person in my undergrad doing astro stuff was a planetary scientist. He did not have much advice that I found helpful, and he was not able to point me toward any particular school or researcher. My physics department was undergrad only.

You could ask some of the people that you work with if they know someone at LocalStateU that you could talk to about attending their PhD Physics program while working part time at Goddard. The thing is, graduate schools don't want to let in anyone that they don't believe will complete the program. If you can show that you are already set up to do research that you could turn into a thesis, you have a leg up. If you know people who know people at the university, you have a leg up. If all you have been doing is visiting universities and sending in applications, well yes, people do get into programs that way, but it's not necessarily the best way to get what you want.


how about doing a masters first? that would give you more of the intermediate knowledge you seem to be lacking?

Yes. JHU has an applied physics masters program for professionals that some of my colleagues have done. However, it seems to me focused on practical applications rather than theory. More of an engineering lite it looks like. I have no idea if it would help my prospects, but it is an option.

related to that last point - have you considered applying abroad?
I have but it seems an even more daunting prospect than applying domestically.

If you can show that you are already set up to do research that you could turn into a thesis, you have a leg up.

My experience has been that they care more that you have a clear path through the program, and that you have acquired the resources to make it happen. This may mean that you have applied to fellowships and grants. Wining a NSF transferable fellowship, for instance, is a ticket into almost any graduate program that you want, since you would essentially cost the school nothing, and also have shown that you are skilled enough to win a fellowship.

I'm not sure how to show I'm "set up to do research" if I've never done research in the areas I'm interested in before. I did try to apply to some fellowships. That all came to naught. The NSF Graduate Fellowship specifically wants a general research proposal which, again, I don't know how I propose research without already having some experience doing the research. It seems like a catch22 to me but I'm probably missing something.

Are all of your letters of recommendation from Rowan faculty? It can be helpful to have a letter from somebody in a higher profile position. Is there somebody well known at NASA who could write a positive letter about you? I am a faculty member at a school that lists Rowan as a peer institution, but I do research (with undergraduate students) at a prestigious research institution nearby, and when my colleague at that institution (very well known highly accomplished scientist) writes a letter for my students, it makes a big difference.
They were. I could certainly get a recommendation from my branch head this year. She has an astrophysics PhD.

Many programs won't accept you unless your GRE is very good. Often the cut off is missing more than three math questions. That's a firm cut. Study middle school math, get all but three questions right. Check.
I'm not sure this is possible for me as I'm not even able to complete the quantitative section in the time allotted. Many of the things on the GRE math section I never learned in any level of math.

Look on the internet for groups you want to work in.
This does not really help me narrow it down. People in the field seem to know who is doing the "hot research" or whatever, but I can't seem to figure that out on my own. How do you even know if the person/group is doing good work?

What sort of places have you been applying to? Are you shooting too high? (Do you have to for the job to be useful? I'm not a physicist either, but I am involved in graduate enrollment in my program, mathematics. )
I applied to a few top 10 schools mostly because of advice I received in the previous thread. That's yet another area where I don't really understand what the best strategy is in terms of finding a school I can get into. The thing that really got me is that I applied to the University of Maryland after contacting a professor in the department who was doing GR research and mentioned them specifically in my application and still didn't get accepted. They're listed as top 20, but I really thought I had a shot.

From your stated interests, I can't tell if you're interested in Astronomy or Physics. In the previous thread, you said you were interested in exoplanets, high energy astrophysics, and black holes, but now you list your interests as GR, Cosmology, and particle physics. To me, it looks like you're realizing you're more into Physics than you are into Astronomy. I mean, they're related, but even the type of astronomy you want to do is usually studied in astrophysics groups within physics departments.
Hi. You're right. I'm kind of flailing here. I'm interested in all those things. I guess I'm finding that my interests are somehow too broad, but I'm at a loss on how to narrow them down or find the thing I want to do without having already tried doing some of them. It's very frustrating.
posted by runcibleshaw at 8:58 AM on May 24, 2016


For networking: write things down immediately. I have to do this too or things immediately fly out of my head. I would suggest getting business cards so you can hand them to people when you meet them - they will often give you one back, which means you're guaranteed to have a record of their name! If not, write it down as soon as possible, along with some brief notes on what was said during the conversation and what you want to follow up on. If you reach the end of the conversation and can't remember their name, a good tip is to ask for their email address so that you can follow up -- most people's academic email addresses are some variation on their name. I have also resorted to looking up the person's institution/department and scanning the photos so I can grab their name that way. In any case, have an organized way of keeping track, either with a simple Word/Google Doc file or something fancier such as Evernote.

For figuring out who you want to work with - literally pick the top 5-10 schools in your field (ask your undergrad advisor or NASA supervisor if you need help) and go browse their websites. Click on each individual faculty member, and look at what they're working on. If it sounds interesting, read some of what they're writing. Look at who their current students are and what they are publishing. Another piece of good advice I got in this area was to think about which younger scholars you admire, and look up where THEY went to grad school. You wouldn't want to do this with someone who's 70 (because programs change over time), but someone who's graduated in the past 5-10 years can at least give you a good idea. Check out who their advisors were.

Something like "go to this website, look for papers that have a blue star next to them and contact the fourth author listed on a Thursday" is more my speed.
I really don't want to be a total dick. But. If this is really your speed, then you should probably work on figuring this out before you try and go get a PhD. If you're not okay with being very very self-directed/self-motivated and dealing with lots of uncertainty but plunging in anyway, a PhD probably won't make you very happy. Getting a PhD is pretty much ALL ABOUT doing those things. I would think really hard about why you want a PhD. If you are the sort of person who likes a really clear list of what to do to succeed, you probably will not be very happy with this career path.
posted by rainbowbrite at 9:02 AM on May 24, 2016 [6 favorites]


On refresh/update, GRE is the place you absolutely have to improve if you want a shot at getting into a top 20 school. For the places you're applying, scores that low for math/physics probably mean that they are not even looking at the rest of your application. There's not really a way around that. If you're encountering math on the GRE that you've literally never seen before, I would consider taking some math courses at a nearby community college just to get up to speed. You will also probably need a GRE-specific prep course, but I would make sure your math skills are 100% solid first. They're expensive, but my sense from talking to people who have taken them is that they can be effective. You're correct that speed is a big part of it -- this means your skills just have to be really solid and you need to practice to get your speed up.

Again, if these are things you don't want to do, you probably aren't going to like or do particularly well in a PhD environment. Your classes are going to be lots of math, and if you can't do the basics quickly, you'll fall behind really fast. So if this is truly your dream, I would get this squared away now.
posted by rainbowbrite at 9:15 AM on May 24, 2016 [8 favorites]


One more thing... Networking is definitely another weakness of mine. I'm trying to develop connections by attending local APS events but I'm often at a loss with where to go from there. Also, most of my colleagues at Goddard are engineers who don't care about theory or research other than for how it applies directly to their job. I have no interaction with anyone on the science side of things.
posted by runcibleshaw at 9:17 AM on May 24, 2016


[Couple of comments deleted. runcibleshaw, maybe take a step back from the thread for a while; people are giving you answers about what it takes to get into and be successful in a PhD program, and you can mark the answers that seem most useful to you.]
posted by LobsterMitten (staff) at 9:24 AM on May 24, 2016


Your Physics GRE (and quantitative GRE) scores are sinking you. As rainbowbrite says, with a 39th percentile on the Physics GRE, many admissions committees are probably not looking closely (or at all) at the rest of your application. There's evidence that scores on the physics GRE are not strongly correlated with success in graduate school (in Astronomy, anyway), but for a nominally evidence-based crowd, graduate admissions committees are surprisingly resistant to change.

Your strong qualitative / weak quantitative general GRE scores are, perhaps, more concerning. Doing well on the qualitative GRE shows that it's not general test-taking issues and doing that poorly on the quantitative general GRE is likely to be pretty concerning to a graduate admissions committee.

I think your only shot in is through some kind of "back door." I had a student who worked with me who, after graduating (undergrad), went and worked as a research assistant at an excellent university. She managed to gain admission to the PhD program at that university without even taking the physics GRE. She was, however, one of the strongest student researchers I have ever encountered. You have to be that level of researcher to get into a strong program with so-so physics GRE scores.
posted by Betelgeuse at 10:03 AM on May 24, 2016 [3 favorites]


With GRE scores in the 30s your application would not even be looked at at my school unless you were a minority candidate. And as for what Betelgeuse said, we had a student who a professor tried to 'back-door' into the PhD program and her lab because he was an excellent researcher (experimental) despite his mediocre GRE scores (which were better than yours), and the Dean of the graduate school blocked his admission on the basis that low GRE scores would hurt our rankings. Something similar happened to a labtech who had four first-author publications in the lab at the ivy where I was a postdoc. So that might not be an option at many places.

You need, badly, to up your GRE scores if this is what you want to do.
posted by overhauser at 10:19 AM on May 24, 2016 [3 favorites]


The US News and World Report stuff is crap. Figure out what the major disciplinary association is and see about their rankings or reputational reports. Note that people move around though, so old rankings may not reflect the current composition of a department. Here's an example one from my discipline. Also NRC rankings can help.
posted by k8t at 10:19 AM on May 24, 2016


Your GRE scores are the major problem here. Your quant score would exclude you from many quantitatively-oriented social science programs, much less a hard science one. Beyond the actually admissions committee in a department, at my grad university, there was a minimum university-wide GRE requirement for TAs and there probably is one at most schools.
posted by k8t at 10:30 AM on May 24, 2016 [1 favorite]


You need to improve your GRE math and physics scores to attend any US institution. Even a bottom 25% program will likely decline you with these scores. Study study study. Take a course and practice doing math faster. Some people don't have the analytical brain to do more. This is what the tests are designed to suss out.

You can try to find a job as a technician at a lower ranked university if you want to back door your way in. This will also let you experience the work and paper writing etc. It is a long hard road from one of these universities to a tenure track position. You need luck, dedication, and some great talent. If you can't improve you gre scores and make it look easy, this is likely not in the cards for you.
posted by Kalmya at 1:07 PM on May 24, 2016


catch22

Sorry to be blunt but this is first gate towards being accepted into a PhD program; you'll need to embrace the firehose and learn how to figure out how to filter that stream and delve into the papers that you are interested in.

That's what you're essentially going to be doing the first year (in addition to classes) and for the rest of your career. Reading, filtering, and interpreting published research is one of the core skills that you need to even start being an academic. Your undergrad training should have been sufficient to start you down the road and self-directed learning is absolutely required.

Shortcut; go through the physics faculty website for your local universities (or universities that you'd be open to moving to). Faculty pages will have a summary of that researcher's research interests, their lab team, and recent/highlighted publications.

Read those publications. If someone's publications are interesting to you, contact them via email saying that you found their research interesting and whether they are currently in a position to add a PhD candidate to their team.

Scenic route; read papers, identify research areas that you are interested in, pay attention to first authors and the last two/three authors (this might be different in physics). Follow individual researcher's publications over time. Those first authors? They might just be starting their own lab and need to churn successful grad students in order to stay on the tenure track; email them saying that you had found their research interesting and whether they are currently in a position to add a PhD candidate to their team.

But yeah, retake the GRE and get your quantitative scores up.
posted by porpoise at 1:10 PM on May 24, 2016 [2 favorites]


One approach is to get a master's first (to go from a less-known school, with a less-than-great scores, to eventually end up in a better-known PhD program; at least this is true in math, I don't know about physics). So, you'd apply to master's programs at mid-rank schools, get a 2-year master's, and then apply to better-ranked PhD programs from there.

The advantages: you'd have two years to figure out your research focus, and two years to improve your record, improve your scores, get good references from the professors, demonstrate you can hack it, and you'd get the name of the mid-rank institution on your app's to the PhD programs. The disadvantage: funding is often more chancy in master's programs - still, often you can find teaching assistantships etc that will cover your costs (and little else), and as always, you should probably not go if you'd be taking loans.

Much of the advice above still holds even if you go the master's degree route -- You'll still want to scope out the field, in order to select master's programs that are focused in the sub-areas you want, and to find which master's programs have a good track record of placing their graduates in PhD programs. And you'd still want to improve your math.
posted by LobsterMitten at 1:25 PM on May 24, 2016


Your GRE scores are probably getting you dismissed out of hand. You need a near perfect general GRE (especially on the quantitative side) and a much higher score on the subject exam. You've got a nice GPA, but there are some basic thresholds for qualification that you currently don't meet.

Why did you apply to the schools you applied too? Did you talk to anyone beforehand in the departments? I was a physic's PHD applicant about 8yrs ago, and I talked to professors in at least one or two labs per school, ideally in person (yes I spent my weekends traveling on stupid Chinatown busses to talk to professors) before I even considered applying to the school; in this day and age, every single science department has a homepage, and most labs have a bio page/summary of research. I was not the strongest candidate (I had a fractured undergrad experience with a rather weird GPA; but I had near perfect general GRE scores, and mediocre subject GRE scores. (~60%)) but talking to professors around the country gave me a good idea of 1. what I actually wanted to study 2. what people were excited about 3. where the money actually was and perhaps most importantly. 4 what my chances in that particular school for the coming year was.

Disclaimer: I ended up not going to physics grad school for REASONS and it turned out ok for me.
posted by larthegreat at 1:41 PM on May 24, 2016


They're listed as top 20, but I really thought I had a shot.


I guess you could look for programs that don't require GREs for their applications, or have no minimum scores...

I'm not a fan of the GREs, and got rejected from at least one graduate school on the basis of my math subject GREs, but I don't see how you can possibly expect to be successful in a physics program with 64% on the quantitative section of the general GREs. Maybe if you had an amazing recommendation letter that in addition to saying you were amazing, explained how come you can't take tests or something---but look, I just skimmed a GRE general sample quantitative test. If you're planning on getting a Ph.D. in a quantitative field like physics, you should be able to answer all of these questions: you should know how to do them all, and you should practice enough that you can do them all quickly.

(I have a lot of sympathy for this; I have a math PhD and am a professor, and I don't know what score I would get if I sat down cold to take the test. Probably crappy, because I'm slow at arithmetic and make mistakes. But you have the opportunity to practice! Which makes a difference. If you take a sample test under timed conditions, no books, no notes, etc, once every 2 days for a month, I bet you'd speed up.)

You said you'd never seen some of the stuff on the test. If that referred to the general, there seemed to be roughly four categories of material: college algebra/precalculus, sophisticated basic arithmetic, a surprising amount of geometry, and a chunk of basic introductory statistics. Did you identify some area you'd not seen? If I had to guess, it's probably the statistics, so you could sign up for an introductory stats class (even online or over the summer), or just work through some basic problems (maybe using something like Aleks?).

(But still, I keep coming back to 'how are you going to succeed in a PhD program without being able to quickly do basic math stuff?' I know, arithmetic != math, so probably arithmetic != physics, either, but....)
posted by leahwrenn at 1:41 PM on May 24, 2016 [7 favorites]


Since you said that you were already accepted into a couple of schools, which you declined, it seems that you should be able to get accepted elsewhere. Or, you could revisit your decision about those schools.

I wouldn't worry about not having a research topic in mind beforehand - I didn't have a clue when I started my Ph.D. in particle physics, but my supervisor certainly did, and had plenty of suggestions. Working as you do on detectors I think may well give you an edge when applying in experimental particle physics - I suggest focusing on universities having physics departments with faculty participating in detector developments for the LHC or FNAL experiments. I'd also suggest thinking about capabilities of the detector(s) you work on, and how they could be improved or repurposed - that might give you some research ideas.
posted by julianb at 1:54 PM on May 24, 2016 [1 favorite]


Astronomy PhD here.

If your interests are broad enough that you think you could be happy doing research that is *either* astronomy or physics, then I would concentrate on the astronomy departments. They don' t have as strong of a cut on GRE scores. I've seen lower from successful applicants. I really think it's the combination of the fact that you have unexceptional GRE scores *and* no one famous writing your letters. The best way to get into grad school is to get a letter from someone the committee knows. Luckily you are at Goddard, so you can begin to work with someone.

The issue is funding. Since you aren't a current student, and have no training, you fall in this strange category. There might be rules against you working for free. I would look for a data analyst position, if you have coding skills. That will get you familiar with a project, and get you interacting with someone well-known. Do a great job, and get a great letter. I think it will really, really help.

Good luck!
posted by pizzazz at 1:57 PM on May 24, 2016 [1 favorite]


If you go to the results board at GradCafe, you can see the scores of candidates accepted to (and rejected from) various PhD programs. It's self-reported, so take it with a grain of salt, but it might give you some ideas about the standards that different programs have.
They also have a whole lot of forums, including ones specifically about discipline areas and the application process. The quality of information and discussion is hugely variable (I was debating whether to even suggest it as a source), but you might luck out and find some useful contextual information.
posted by une_heure_pleine at 2:56 PM on May 24, 2016


I got a master's in my professional field (a STEM field but not physics or astronomy). Before that, I had absolutely no idea about how its research is done and how to understand its academic writing and journal articles.

The master's degree program gave me that training and experience, both in the classroom and in practical projects. It gave me the basic research vocab so I could ask questions and understand the responses. It helped me to understand my own interests and to come up with project ideas myself. It gave me the tools to evaluate the research and sort out the junk from the gems. Most of all, it helped me see that I do not want to work in a research position. I'm relieved that I learned this after just a 2-year program and not after a 5+ year commitment.

I'm seconding the people above who recommend that you pursue a master's degree first.
posted by cadge at 3:11 PM on May 24, 2016


I can't speak for physics programs, but there are not astronomy master's programs. I mean, you get a masters along the way towards a PhD, and you can bail afterwards, but it's not typically a separate thing you apply for.
posted by pizzazz at 4:28 PM on May 24, 2016


there are not astronomy master's programs

This is not true. There are at least two excellent terminal masters programs in Astronomy in the US - one at Wesleyan U. and one at San Diego State. They can be effective ways for students with not-great undergrad experiences (or those returning to the field after some time away) to get into good PhD programs. I know several people who went to Wesleyan's program that got into top-tier PhD programs after getting their Masters.

That said, these are very much Astronomy focused programs. So, star formation, protoplanetary disks, extragalactic X-Ray astronomy, etc. They don't have really anyone doing the more Physics things that the OP is interested in. I also think that this program wouldn't be very effective in getting the OP into physics PhD programs.
posted by Betelgeuse at 4:50 PM on May 24, 2016 [1 favorite]


Many of the things on the GRE math section I never learned in any level of math.

I don't want this to come across as harsh, but if there is anything on the GRE quantitative section that you've never seen before with a B.S. in physics, then your undergraduate education did not prepare you for postgraduate training in physics. Honestly this is a little shocking to me. I got a B.S. in physics, and the quantitative section of the GRE was a breeze even with essentially no studying. I only missed a few questions, and that probably put me behind most of my classmates; it was expected that anyone coming out the department should have no trouble getting a perfect or near-perfect score on the general quantitative exam (this was at a well-regarded state school, not an Ivy). Fortunately I wasn't trying to go to grad school in physics, and my quantitative section score was perfectly adequate for neurobiology. My classmates who did go on to grad school in physics all spent a couple of months full-time studying for the physics subject GRE.

So in addition to joining the others saying you need to work on bringing your GRE scores up, I'd also add that this isn't just a question of improving your test scores, but likely reflects a fundamental gap in your background knowledge and training that you would be expected to have as a first year graduate student. If pursuing a Ph.D. in physics (or, I presume, astronomy/astrophysics, though I know much less about that) is a goal you are serious about, you're going to need to identify the gaps in your training and teach yourself what your undergraduate education didn't. At the very least you should be able to complete the quantitative GRE and be confident that pretty much everything you've seen is familiar, even if you miss a question here and there.

As for selecting a program to apply to, I know things tend to work a bit differently in physics, but in my field it's common to apply to a program with a fairly good idea of who in the department you'd want to work with. Having a very specific research topic already lined up is rare, but you should at least have a good enough sense of the big-picture questions in the field that interest you to know who is doing good work on them, and ideally who has a perspective on them that attracts you. As others have addressed, regularly attending a talk series at Goddard is an excellent way to help develop this. Skimming article titles on Arxiv and reading the ones that interest you in greater depth is another good strategy. And for those articles you find particularly interesting, Googling the authors to find out what other research they're doing and what departments they're in is a good way to get ideas as well. But before you actually spend the effort on preparing and submitting an application, you should definitely bring up your GRE scores, and commensurately, ensure your background in physics and applied math is solid enough that your first year of grad school won't wash you out.
posted by biogeo at 5:05 PM on May 24, 2016 [10 favorites]


I agree with those who note that your quantitative GRE score might be a big stumbling block. My most advanced math in college was analysis in R^n, when I was still a physics major; I then changed to history and still got a perfect score on the quantitative GRE two years later, without any test prep. I left physics and math in part because I was good at them but not great at them. I was reminded of a passage in one of Robertson Davies's novels in which a character describes himself as a master of the grammar of money—how investments work, how to make money, etc.—but contrasts himself with a nephew who has mastered the rhetoric of money, and can make it do what he wants. While I was studying math and physics, I had a similar sense: I understood them, on a grammatical level, but I couldn't make them sing, like a rhetorician.
posted by brianogilvie at 7:18 PM on May 24, 2016 [2 favorites]


Physics PhD programs require you to pass a Qualifying Exam after the first year in order to move on to thesis work. This most often (but not always) takes the form of a two day, four or five hour per day, written ordeal where you can be asked just about anything from your prior physics education, up to and including the classes just completed. Most departments will post sample tests on their web sites; I advise you to take a look when researching options.

Not to pile on about the GRE, but I imagine that admissions committees are finding it hard to reconcile your high GPA with your low GRE scores. The Physics GRE is not so much about math as it is about facility with physics concepts. Those evaluating applications want to see evidence that you are capable of passing the Qual.
posted by Standard Orange at 9:13 PM on May 24, 2016


I discussed this with my wife the grad chair in physics.

University of Toronto and Carleton don't require GREs, as they're weakly correlated with future performance.
McGill would like it, but doesn't insist.
Alberta: "The GRE Physics subject test is not required but is strongly recommended."
UBC: "GRE scores (both general and a relevant subject test) are requested when available"
I believe many and perhaps most Canadian physics departments are like this.

Memorial University in Newfoundland does GR. (Note climate.)

Having a very specific research topic already lined up is rare, but you should at least have a good enough sense of the big-picture questions in the field that interest you to know who is doing good work on them, and ideally who has a perspective on them that attracts you.

In GR and particle theory, an incoming student won't have a specific research topic lined up because they haven't mastered (or started) the material yet and haven't done 10 years of current reading on the subject. (Stated somewhat strongly by her.) First-go postdocs are often reliant on their prof peers for a lot of this context stuff as well.

Learn some basic standardized test skills such as eliminating answers etc, if you eliminate enough guessing among the remainder will still help you on average).


For GRE physics one of my profs did a one month prep course. We learned to quickly eliminate wrong answers by doing unit analysis and taking variables like m1, m2 to 0, infinity, -infinity and seeing the answer do non-physical things. We used a concise overview of undergrad physics, which I'll dig up another time. And we drilled for speed a bit.

There are a lot of resources for prep for GRE quantitative. I found the adaptive test ramping up in difficulty to be quite unsettling, and there was some usage and math that I hadn't used or seen in undergrad physics, if I recall.
posted by sebastienbailard at 9:13 PM on May 24, 2016 [3 favorites]


On more thought, I wanted to emphasize this advice from rainbowbrite:

If you are the sort of person who likes a really clear list of what to do to succeed, you probably will not be very happy with this career path.

I think this is probably the single best piece of advice you've gotten in this thread, and I would consider it very carefully. If all you want is to participate in astrophysics research and learn cool things, you may be better off finding a technical job working with astrophysicists and doing a lot of reading and attending talks on your own. I don't know how things work in astrophysics, but in my field, there are a number of research technician jobs available for people who are good at science and love bench work but aren't interested in being responsible for running a lab, developing a research program (for which there is absolutely no clear list of what to do to succeed), or spending much of their time writing grants. Have you considered whether there are such jobs available which would let you work in a lab doing research you find interesting with other people who share your interests, but which wouldn't require getting a Ph.D?
posted by biogeo at 8:16 AM on May 25, 2016 [5 favorites]


Looks like the consensus is my GREs are not good enough to get in basically anywhere. I did send emails to each department that rejected me asking if they could offer any insight into why I wasn't accepted and none were able to. Does anyone know of a physics GRE prep course, either online or in person? Google is failing me. The general GRE courses seem ludicrously expensive, but if that's what I gotta do, that's what I gotta do. I think I do understand all the concepts (except for circuits, I just skip circuits). I don't think being able to answer a question correctly in under two minutes means I don't know the material or has anything to do with being able to do research or do well in grad school (which I assume doesn't involve a stopwatch). But, I understand that it's a requirement whether I like it or not.
posted by runcibleshaw at 11:42 AM on May 25, 2016


« Older Hope me blog   |   Debt collectors want me to prove I'm not the... Newer »
This thread is closed to new comments.