Do scientists really need a PhD?

http://www.nature.com/nature/journal/v464/n7285/full/464007a.html

by mbrubeck • 15 years ago

52 47 15 years ago

Does Bill Gates, Mark Zuckerberg or Jack Dorsey need a BS in Computer Science? No, apparently not.

The purpose of science related PhDs has always seemed to be a means to an end that wasn't financially available anywhere else. Academically, there are a lot of pursuits that you simply can not pursue outside of that setting. The science PhD is ultimately the direction you want your career to take, but you know it can't realistically take that path without some luck. And of course, the safety of academia is intoxicating on all levels. Why dive head first into research that requires funding and years of preparation when you can basically do the same thing at school without any repercussions? The PhD is a safety net for abstract thought that has little to do with the real world.

It may seem like a crazy comparison, but the sort of struggles scientists go through with their research is not entirely unlike what an artists goes through. No one really wants to acknowledge your absurd existence, and there's not really anyone to talk to when you do get on the outside. So no, success, be it financially or scientifically does not require higher education. But socially and emotionally, you can't underestimate the importance of things like a PhD.

zerostar073• 15 years ago

Are they scientists? No.

oldstrangers7• 15 years ago

Is computer science not a science? Someone should alert all the major colleges of your discovery.

wisty• 15 years ago

They oversee R&D programs. They might not be good scientists, but they do a very good job running an R&D intensive organization. Unless your definition of science is "writes papers nobody reads about stuff no-one cares about while being payed tax dollars", industrial R&D can also be science. Of course, Facebook is a terrible example. Google would be a much better one. Intel or Boeing would be even better.

There'll be people who say that computer R&D isn't science, because it's just gimped math or hand-wavy pretend engineering, but they'd be the same people who say that most of the stuff computer science academics do is also crap.

grovecai• 15 years ago

Apparently, they are not scientists.

kevinalexbrown• 15 years ago

I'm planning to get a PhD in a science, so this is a question I've asked myself a lot.

From a theoretical standpoint, a PhD is obviously not a magical "get good at science" button: there are plenty of other ways to gain competence. An example is Rick Stevens at Argonne National Labs, [1], who I'm pretty sure doesn't have a PhD (my friend worked for him once, and his bio doesn't mention it, but I don't know him personally). People like that show that a PhD is obviously not necessary to do first-class work.

On the other hand, if you want to be a principal investigator on a grant of any decent size, almost no one will believe you have that competence without a PhD. Part of this is the ridiculously rigid US scientific job market: all the arbitrary structure of a resume, even less of the rationale. You can't get a postdoc, almost trivially, without a doctorate. Even if someone wanted to give you a job as a postdoc, the university wouldn't like it, and why choose the guy without the PhD when you have other talented people who also have those three letters? And it's really hard to get a faculty position without a postdoc, so there you go.

So if I could go about becoming a principal investigator or having a nice faculty position without getting a PhD, I'd seriously consider it. But practically speaking, that's not a real possibility yet. Hopefully someday it will be.

[1] http://www.anl.gov/Administration/Bios/stevens.html

p4bl02• 15 years ago

I'm also planning to get a PhD in CS starting next year. But I think this question is irrelevant, in the sense that it is not the right one. Because of course the answer is formally "no", one can be a very good scientists without having a PhD. But do society needs people getting PhDs and people training them and the whole university system. Yes, definitely yes. And a PhD, even if not formally needed to be a good scientists, is an effective way to become a scientists since it's basically learning to be one with others who already are.

aroberge• 15 years ago

Sorry for stating the obvious but, Postdoc == Post-doctoral (fellowship). So, of course you can't get one without having doctoral degree. However, it is sometimes possible to work as a research assistant and end up doing very similar work, without requiring a Ph.D.

gaurav_v• 15 years ago

Well, I think to be a PI on a grant of any decent size what you need is a tenure or tenure track faculty position, which requires a "PhD or equivalent."

As far as I can tell, "or equivalent" means that you've made so many contributions to your field that institutions are fighting with each other to give you a tenured position anyway. Freeman Dyson, for example, never got a PhD.

Like everything in science, it's just about publications. If you somehow manage to publish a lot of significant papers, nobody is going to blink over the "PhD" qualification. That being said, the easiest way to build up a publication record is via the PhD-postdoc pathway (at least given the current system).

raphman2• 15 years ago

One thing to note:

There are a lot of differences between PhDs in different fields of research. Especially in biomed research, PhD students often seem to do extremely specialized research. Therefore, what they learn during their PhD may be much less generalizable than e.g. what CS PhD students learn. I guess it depends on the definition of a PhD whether it is necessary for a certain job or not.

Interesting comments here: http://scientopia.org/blogs/science-professor/2011/03/22/rat...

mvzink• 15 years ago

I'm still an undergrad and therefore naïve, but I had always thought PhDs were much more about proving they can perform research than "what they learn"

Radim2• 15 years ago

You are correct; PhD is a degree where you are expected to demonstrate original research.

See http://matt.might.net/articles/phd-school-in-pictures/ for a visual guide to academic degrees :)

raphman2• 15 years ago

Maybe we should distinguish between the doctorate you get awarded for you PhD thesis (the proof that you can perform research) and the stuff you do during your PhD studies (what you learn). Depending on you advisor(s), your topic, your own knowledge and motivation, and many other factors, you may be able to conduct great research or mediocre research. Depending on how narrowly you focus, you may acquire more focused knowledge or more broad knowledge during your research. In fields like biomed research, you work on a very confined topic within a large team. You may only need to care about your specific experiment. Everything else is nothing you should (or need to) care about. Compare this to a typical CS PhD student at a German university (e.g. me). You more or less define the topic you want to work on by yourself, are required to teach courses, supervise students, occasionally work on a commercial project, and have to learn all that stuff which is necessary for your research, e.g. user studies, statistics, machine learning, electronics, etc.

Having a PhD does not necessarily mean that you know how to perform proper research. I know of a good number of PhDs that have little knowledge of core scientific principles like correct attribution. In my opinion, what you learn on your journey to a PhD is much more important (in most cases) than the topic of your PhD thesis.

FaceKicker• 15 years ago

Disclosure: first-year CS Ph.D. student.

I would say, no, a scientist doesn't really need a "Ph.D.", but does need the training that a Ph.D. involves. To me, this question is very different from the question of whether someone who wants to be an industry software engineer needs a bachelor's degree (a popular HN topic). The difference is that in a science Ph.D. program, you are doing science the entire time you are there. Sure, you take a few classes as well, but these are only to reinforce concepts you'll need in your research; the entire reason you're there is to learn to do scientific research by actually doing scientific research. You are doing at school what you'll be doing in your career; you're just getting paid less for it because you're "in training".

I don't see any reason why one couldn't get a similar education by doing scientific research in an industry which does not result in a formal "Ph.D." degree, rather than an academic setting that does result in a Ph.D., it just doesn't seem like it's been tried up until (apparently) now. Personally (all other things being equal) I'd rather do a Ph.D. program in an academic setting because an industry setting would likely be constrained to research areas that are beneficial to the company.

tincholio• 15 years ago

No, but it helps. There are some very bright researchers who don't have one (Simon Peyton-Jones being one, for example), but they do help in forming people as scientists. Of course, working after getting your BSc. as a researcher on a specific area for many years is, in most aspects the same as doing a doctorate, so except for the lack of diploma, there probably shouldn't be a big difference.

timpeterson• 15 years ago

phd is dumb but skeptical china will be informative as a proving ground for this idea,

do people think # of chinese nobel prize winners will increase in years to come? realize nobel prize is a western world creation but this makes like seem no...http://en.wikipedia.org/wiki/List_of_Nobel_laureates_by_coun...

ThomPete• 15 years ago

I guess the question is. Do you need a PhD to create scientific break-troughs or practice science.

Cause that seems to be the only metrics worth considering when answering that question.

I would guess the answer is no.

Radim2• 15 years ago

No need to guess -- the answer to that question is "no".

Universities started awarding PhD degrees at the (end of) the 19th century. If PhDs were a necessity for scientific breakthroughs, then there couldn't have been any scientific breakthroughs before the 19th century, a contradiction.

QED.

edit: but I got myself one just in case, anyway :)

urbanjunkie5• 15 years ago

You don't need a PhD to be a scientist, but you almost certainly need a PhD to be funded to do science.

(there are of course exceptions, but this is very much the general rule)

ams6110• 15 years ago

As well you need one to be accepted as a legitimate peer in science communities where everyone else has one.

nn2• 15 years ago

I doubt the scientists need PhDs, but a lot of research groups need cheap sla^wgrade students and the PhD process provides that. Now if more of the research funding actually was used for paying scientists that system wouldn't be needed, but it is how it is.

WhatsHisName• 15 years ago

It's been my experience that most of these Chinese biotech companies are factories. One guy makes all the money and the rest do all the work. This is not at all like the computer technology model in the US.

hessenwolf2• 15 years ago

Bioinformatics, and chemistry/material science areas seem to me to require way more semi-skilled labour in general than computer technology (even though I could still really do with a basic DBA or two and a low-level html-cranker).

Irishsteve• 15 years ago

From what I gather, the point of a phd is to train researchers so they can become independent researchers i.e set their own 'big question' and understand methodologies to solve it. This article states that the research direction in China is coming from external parties (I suppose commercial interests).

So all thats happening really is that companies are finding ever cheaper sources of skilled labour to solve their current problems. Academic researcher ideally shouldn't be linked to a companies motivations... or at least not all of it.

hessenwolf2• 15 years ago

Yes - any large research project like the one described needs hordes of gobots to do the labour and testing. This is not the same as having hordes of people who identify both important hypotheses, and the means by which they will be tested.

baddspellar• 15 years ago

A PhD is a credential, as is a BS, an MS, or a CCIE.

As long as employers demand a credential for hiring, that credential will be necessary. In the US, few companies are willing to take on the burden of training scientific researchers. Further, much of the pure research in this country is done at Universities, and virtually all US universities require a PhD for tenure track faculty. So a PhD is a necessary credential to do scientific research in the US.

The linked article states that Asian companies are willing to invest in training smart university graduates to do research. That means the PhD credential is not required. The required credential is instead a "University Degree".

Now, if the question is whether the PhD system is the best way to create scientists, that's entirely different. There's no reason to believe the University PhD system necessarily produces better scientists than a corporate lab would. University professors are not necessarily good teachers, and University labs are not necessarily well equipped.

It's very similar to the "BS or higher in Computer Science or Engineering" credential on help wanted ads in the US. Does that BS or higher really make you a better programmer than someone who learned on his/her own? Of course not. But it's a credential, so you get it.

Disclaimer: I do have a PhD

PaulHoule• 15 years ago

Frankly, the PhD is not a path to a permanent job as a scientist unless your parents were scientists.

What it is is a program to get cheap labor that has the energy and low health insurance costs of youth. At some point you get thrown away and they get somebody new. The reason why P.I.'s are aging is that there are very few permanent jobs for scientists.

Perhaps these people in China are a bit more honest than the rest of us.

andrewl• 15 years ago

Yes and no. Yes if you're talking about having a chance of working in the US at the level lots of PhDs do (faculty at a university, principal investigator of a grant-funded project, etc.). No if you're just asking if you need a PhD to think and work independently at the level of a PhD in a given field. It's my understanding that Freeman Dyson does not have a PhD, for example. He's also a bad example, because he's Freeman Dyson, and most people are not.

Outliers are often bad examples. I think it's still a good idea to go to finish college, although dropping out worked for Bill Gates and Steve Jobs.

larsberg• 15 years ago

All scientists start out without a PhD. Then, they do some research. At some point, they either have published enough novel papers or have done a single piece of novel work that advances the art and warrants granting of a PhD.

If these students truly are primary authors on major papers in serious peer-reviewed journals, I predict that this lab will end up with some sort of joint academic agreement with a local university so that the ceremonial stapling of bio papers and minting of PhDs can be done.

The rest of graduate school (classes, a master's thesis, etc.) is just scaffolding to acclimate you to doing research. If you can skip all of that and jump right in, nicely done!

But, of those 500 kids mentioned in the Nature article, I predict many do not do top-quality published research. I would even bet their success rate (measured as "people who graduate from just running experiments other people designed") will be lower than that of a traditional PhD-granting institution, were I a betting man. The scaffolding is expensive, but is there for a reason beyond just tradition.

spodek• 15 years ago

I have a PhD in physics though I became an entrepreneur. My loose definition of science is that you observe, predict, experiment, and honestly report your results.

Five-year-olds can do that.

I have found exploring and studying nature one of the more rewarding and pleasurable parts of my life. I've done it as a member of academic institutions, but I also do it just walking in the park.

Three things I wouldn't confuse:

a) membership or funding in an academic institution

b) recognition from peers

c) practicing science

I consider them orthogonal. I consider a) a job, b) a nice perk, and c) an enjoyable pasttime or hobby. Great when they overlap but unnecessary.

SilasX• 15 years ago

>My loose definition of science is that you observe, predict, experiment, and honestly report your results.

I would add one more thing: you become aware of others' efforts to do the same and build off of their work rather than "re-invent the wheel". (That doesn't mean you take it uncritically, just that you know what has been tried.)

That part is, AFAICT, what PhD programs most intensely focus on, i.e., make sure you understand the existing work well enough that you can build on it rather than just independently re-discover results in the field.

That part is also (usually) beyond the reach of five-year-olds.

Note: you can certainly still meet this requirement without an official PhD program, of course.

tokenadult• 15 years ago

Previous discussion 662 days ago:

http://news.ycombinator.com/item?id=1183828

Do scientists really need a PhD?

Related Stories