Hacker News Reader

Comment by throwawaymath | Hacker News Reader

With regard to your first paragraph, I don't think there are any American universities that admit students by advisor. As far as I'm aware, your admission is a decision made solely by the graduate admissions committee. Even if you have an advisor ready to take you on the most they can do is recommend you to the committee, not supplant it. When you're admitted there are still criteria aside from research to work on.

I think this is a fundamental difference between American and European universities. In European universities, you can hypothetically join a PhD program without any prior degree at all if an advisor chooses to take you on personally. No such sidestepping can happen in an American institution, and regardless you'll have to take some courses (even if you're only auditing them in effect) before you can begin research. Unsurprisingly, American PhD programs are longer.

View on HN

I've been on the PhD admissions committee for the Berkeley CS PhD program. Students in many areas are, for all practical purposes, admitted by an advisor. It's not a formal arrangement in the same sense as European programs, but no one gets admitted unless at least one professor will commit to being willing to advise and fund you. For superstar students this is rarely an issue and you have your pick of advisors, but for many admits there is exactly one professor who's identified your background and interests as a good fit, so you're essentially admitted with the expectation you'll work with that professor.

Nothing you say is wrong; you're correct that everyone has to do course requirements. But culturally the course requirements are treated as a distraction from research, and students are expected to start working primarily on research from their first semester. I think this is more or less the case in all top American CS PhD programs, and there are reasons to set things up this way, but it's a very different environment from PhDs in most other fields.

Thank you for the insight. Would you be willing to elaborate a bit on finding a good match for advisor, and perhaps any thoughts on approaching someone about exploring an advisorship?

First step is to be acquainted with their previous research. Understand the direction they are taking and what their interests are. Find advisors whose work you think you can build on through your PhD. Write down what you are interested in working on and how it relates/expands/builds on your potential advisor's. Then contact them sharing just that and start having a conversation - they will surely have feedback and ideas for directions that your research could take.

I'm at an American university, and while technically this is true, an advisor going to the committee and saying "I want this person and I have the grant funding to support them" is profoundly influential, to the point that I would be shocked if the committee said no.

True, but:

1) It's still ultimately up to the graduate committee,

2) Once admitted, students will have requirements outside of research, for better or worse.

I'm not making an indictment of the American PhD system, I'm just pointing out that direct advisor admission and unilateral research focus are very uncommon (non-existant?) in the US.

This is about 180 degrees from what I advise anyone asking advice regarding PhDs in the US. Specifically, I tell them that it is best if they know before they even apply who they want to work with and in what capacity. This accomplishes a few things:

- Removes several potential surprises about what they end up researching. It is very sad to see someone expecting to do research on topic A (which they love) but end up doing research on topic B (one they care about much less) because that’s where the money is or that’s where the open advisor is.

- Helps admissions considerably, especially in competitive schools. If an admissions committee denies an applicant that the professor wants, there will be a lot of explaining to do. I assume that you are not talking about “professional” doctoral degrees like JDs or MDs — those are very different than academic-oriented degrees.

- Helps clarify funding. Will you be a research assistant (the answer should be yes if you want to be an academic), a TA, or what?

- Helps clarify what knowledge needs to be known or learned before entering so they can hit the ground running. This helps with course selection (when there are options), project selection, team selection, etc.

- Know the pluses and minuses of their advisor. The best grad students I have seen have read at least some of the research of all of the tenured professors in the department, and they know who is strong and who is weak. They will also have a feel for the politics. These folks are rare, but they are usually the future big time academics.

As someone alluded to elsewhere in this thread, the biggest winners in the academic game hit the ground running for their phd. The folks who come in with an open mind and are looking to learn often will have extremely dim career prospects.

> I'm not making an indictment of the American PhD system, I'm just pointing out that direct advisor admission and unilateral research focus are very uncommon (non-existant?) in the US.

This might be true in schools or departments with very weak research programs. Even then, an advisor advocating for a specific applicant, even if it is a rare occurrence, would most likely be accommodated.

Note that highly talented (relatively speaking) doctoral students are rare. If a professor can get one that’s a great fit, s/he would not be wise to cross their fingers and hope that person gets in and gets assigned as their advisee. Just thinking of a professor leaving that much up to luck makes me laugh — it would never happen.

"Specifically, I tell them that it is best if they know before they even apply who they want to work with and in what capacity."

I append this with the advice that one should never go to a place where their prospective advisor is the only person they can imagine working with.

Primarily because:

"Know the pluses and minuses of their advisor. The best grad students I have seen have read at least some of the research of all of the tenured professors in the department, and they know who is strong and who is weak. They will also have a feel for the politics. These folks are rare, but they are usually the future big time academics."

Is often something that will only come out in time. You can try to figure these things out, and get a good idea, but there are subtle nuances to people that only comes by knowing them.

"This might be true in schools or departments with very weak research programs. Even then, an advisor advocating for a specific applicant, even if it is a rare occurrence, would most likely be accommodated."

My PhD department worked on the "No specific advisor when admitted, lots of coursework" model, and characterizing it as having a "very weak research program" would make you look like an idiot.

> My PhD department worked on the "No specific advisor when admitted, lots of coursework" model, and characterizing it as having a "very weak research program" would make you look like an idiot.

My comment was a reply to throwawaymath’s comment that US programs usually don’t allow professor admits and don’t have a research focus.

If you went to a good school (as you seem to claim), I am willing to bet that one or more admits were advocated for by a professor and (most likely) ended up as that professor’s advisee. You may not have been told about this, but I bet it happens. As I said elsewhere, strong doctoral students (relatively speaking) are extremely hard to find, and most good professors won’t leave a good fit to chance. Maybe your field is different, but I somehow doubt it.

I will also guess that your school has a strong research focus, since having a strong research focus is almost a defining characteristic of a strong program. Note that OP said “unilateral focus”, and maybe your program does not have a “unilateral focus” (I’m not sure exactly how that would be defined), but that largely seems to be a minor issue of semantics rather than the major issue of throwawaymath’s thesis that US programs don’t usually have a strong focus on research (“non-existent” was the word he used). I know of no good doctoral program that does not have a strong focus on research.

Lastly, note that I didn’t say anything about “lots of coursework” — this is common in US doctoral programs. The issue I commented on is that folks who come in with less focus (I think i said “expecting to be taught” or something similar) tend to have weaker career prospects. This doesn’t doom a less-focused person to failure (not at all), but it does make it tougher than someone who comes in with a more intense focus on research. Again, maybe it’s different in your field, but somehow I doubt it. If you have specific examples, I would genuinely love to hear about them — I personally know of a handful, and they are all due to the fortuitous hire of a new professor.

"If you went to a good school (as you seem to claim), I am willing to bet that one or more admits were advocated for by a professor and (most likely) ended up as that professor’s advisee. You may not have been told about this, but I bet it happens. As I said elsewhere, strong doctoral students (relatively speaking) are extremely hard to find, and most good professors won’t leave a good fit to chance. Maybe your field is different, but I somehow doubt it."

The program doesn't do direct admit like that - I know because I asked. Faculty can advocate, and everyone was matched with an advisor, but for the most part first year funding was not under someone's direct RA line, and a slim majority of students ended up switching advisors (myself included).

Ironically, one of the professors who did recruit the strong doctoral students did so in a way that was a very useful cautionary tale for my own recruiting - namely that it's very hard to actually determine that from an application.

This terribly punishes military servicemen who wish to return to normal life and attempt a phd

Note:just making an observation

Can you explain why you think this? I entered a PhD program at the start of the summer and I can't think of a way that veterans would be punished in the application process.

Why?

If a qualified member (bachelors in physics, within UW Physics' normal acceptance criteria for grades/GRE/etc.) of the armed services stepped up and said to our research group, "I think gravity is the coolest thing in Physics, I'd like to spend the next 5-7 years measuring it, and here's why!", that person would have a big leg up in admissions to our department.

I've worked with a number of current and former military in our group. To a person, they've all been great and know how to get things done. Indeed, one of our research scientists is a former submariner.

As some other commenters have stated, I'm not really following your logic.

They may not market it that way, but they definitely do. In my program, you are admitted by a specific professor behind the scenes. You may end up working for someone different during your first year, but there is a specific advisor who said "I think this person is good and I have money for them."

Sorry to be blunt throwawaymath, but that's just wrong. I am very familiar with several American universities (including Ivy League) where the advisor can admit directly.

I don't mind you being blunt at all, but do you mind naming those schools specifically? I'm still skeptical.

Not OP, but Harvard, Yale, and Penn are three that I can speak to with firsthand experience.

I will also say that, depending on the specifics of the applicant, you might be shocked by what admissions standards can be bent/broken based on the desire of a faculty member to have a specific student. It’s not common, but it happens.

For my PhD program, I specifically looked for advisors doing the type of simulations I was interested in. There weren't many (1989-1990 time frame). I found one who on paper looked great: lots of good recent publications, assistant prof (she was working on getting tenure, so she was hungry), she had reasonable grants for a theorist, and she needed an apprentice.

Having that helped my application. I started working for her before starting school.

I still had to take courses. I passed the PhD qual test on entry, and had 2 years worth of courses to take. Transferring in some of the masters courses helped some. I offered to test out of the courses, but it was explained to me that the physics department liked their revenue, even though most of the money coming in was research grant based. The BS I had to go through to take a chemistry course which would have been helpful was a side effect of their love of inbound revenue, and dislike of outbound payments.

Basically, I think the economic model at my alma mater was quite broken. This model generally governs behavior of schools, departments, etc. Having a group to start working with helped me clear some of this BS, but I still had to deal with other crap.

Strictly speaking, you're right, students aren't admitted by an advisor. However, often they are admitted to work with a specific advisor. That is, there are schools where an advisor cannot unilaterally get you in (its up to the grad admissions committee), but unless there's someone that's wants to take you as a grad student, you're not getting in.

It can vary by specialty. A friend applied to several schools for graduate work in geology; the admission process was explicitly admission into a designated professor's research group.

Reply on news.ycombinator.com