[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 570 KB, 3148x1400, 1674562507498115.jpg [View same] [iqdb] [saucenao] [google]
15852131 No.15852131 [Reply] [Original]

Does a PhD in STEM actually help if you're not planning to continue in science afterwards? How well do the skills translate outside of it?

>> No.15852156

No. Get an internship, and specially get networking. Studying to become a salesman will bring you more success than studying to become a specialized monkey.

>> No.15852162

You are not planning to continue anyway, as they don't have a postdoc place for you.

You can get extra points if you become STEM teacher in qualification exam loooool

>> No.15852189

STEM PhD's have value in a very specific sub-area of their field and are pretty much useless outside (compared to other PhD's). The main transferable skill you get from a PhD that you don't get from a Master is showing you can work independently and commit to a project.

>> No.15852196

>>15852131
Is it in pure math or computer science? If so you can build the LLM to replace the people who did gay humanities PhDs.

>> No.15852201

>>15852196
Life sciences

>> No.15852204

>>15852131
It's really not. Don't waste your best years.
>t. wasted 3 years for the PhD and even more time trying to find a job

>> No.15852918

>>15852131
>lil bro aiming for that phd for his "career"
LMAOOOOOOOOO all you need is grad and youre good to go
PHDs are for rich nerds who are too autistically obsessed with their field to actually go to work

>> No.15853813

>>15852131
1: go to >>>/sci/scg
2: yes, a PhD will help, about 20 percent of all physicists end up in management consulting firms like MBB. Another similar sized group end up in finance.
3: skills like analytical thinking can be applied almost everywhere, thankfully since I had to leave research when my field folded.

>> No.15853891

>>15853813
I think two questions are maybe getting mixed up here. The first is "are you completely fucked after a STEM PhD if you don't get a career in science?". The second is "should you go and seek a STEM PhD if you don't want a career in science?".

The answer to the first question is that most people doing a STEM PhD end up in careers outside science, so clearly it's not a death sentence. However it's generally acknowledged that in terms of time and effort invested it's a poor return and in almost all cases you're better off just getting a Masters and a few years of experience. Maybe a PhD is better than a 3-year resume gap, but it's not nearly as good as 3 years of even tangentially related actual experience.

"Analytical thinking" is all well and good, but it's very vague and nonspecific. The actual requirements are on the level of "2+ years of experience writing commercial facial recognition algorithms in XXX".

>> No.15853919

>>15853891
I gotta be honest, i developed most of my truly analytical thinking in my Master's. I just felt like a labrat during my PhD

>> No.15853939

>>15853919
That's a real risk especially with experimental PhDs and I think people aren't aware of it enough. God knows the professors aren't going to be in the lab so someone has to keep tha data flowing. If you're not especially careful it is very easy to just hang your brain out to dry and chug through the experiments that someone else tells you to do. Then you write out the papers that someone else told you to write (maybe your PI gets a theory/computation collaborator to write the actually difficult bits), staple them together and look at that, you're a doctor.

In fact, you'd be surprised by how many PhD students not only accept this but actually expect this. I could rant on but better to stop there. You have to actively try and seek out your own projects (they can be in your PI's research area but have you pushing them and doing the thinking).

>> No.15854032

>>15853939
Fuck man, too real. I really blew it with my PhD. I just wasn't ready for it in term of mindset and maturity.

>> No.15854277

>>15853939
I really wish there were more theoretical research projects in life sciences. I love the data analysis and theoretical side of work, but like >>15853919
said, in practice it's just 95% being a labrat.

>> No.15854310

>>15854277
It's insane. Nobody is hiring lab techs anymore so PhD students end up doing lab tech shit all day.

>> No.15854367

>>15854310
Yes! It is pure insanity. I've seen what it looks like when the day-to-day running of labs is left not to lab techs or even postdocs, but mainly to grad students. It isn't pretty.

They generally don't know what they're doing, because why would a physics grad know the ins and outs of spectrometer maintenance or cryostat best practices? It's the kind of shit that takes years of hands-on experience to get a feel for, and by the time that happens they graduate and get out. That's assuming they gave a shit in the first place, which they usually have no obligation to do.

Meanwhile the setups are consistently so fucked up that half the data is questionable, and damages from improper usage rack up to easily cover the salary of a lab tech per annum. And I am not exaggerating in the slightest, this was literally how it was. The students just roll with it and pretend they don't know that the data is fucked. The PIs make a point of never descending to the labs and have zero idea of what goes on day-to-day, they just want their underlings to bring them finished datasets and manuscripts they can put their name on and spend all day applying for grants and promoting their work.

I repeatedly told them to hire a dedicated lab tech. It was never even considered, apparently "there wouldn't be enough work for them" and nobody wanted to foot the bill. Meanwhile it's taking the total effort of 5 clueless PhD students trying their best full time and failing miserably to keep the place going, all of which could be taken over by one person who knows which end of the wrench to hold.

I have developed such a deep disgust with the entire sector. I honestly hope it crashes and burns, it deserves to.

>> No.15854369

>>15853891
>However it's generally acknowledged that in terms of time and effort invested it's a poor return
In terms of money, perhaps. For experiencing new things it was good, regular jobs come soon enough.In times like these where the economy is tanking hard and only unemployment is taking to the skies, doing a PhD will be better than a big gap in your CV/resume.

>> No.15854423

>>15854310
>>15854367
I seriously feel like the entire Bachelor/Master's/PhD system needs to be revamped and rebuilt from the ground up. It was fine back in the 19th century when the Catholic Church was dealing them out, but i really feel like the current (academic) model is not longer doing what it is intended to do.

>> No.15854429

>>15854367
Pretty much my experience. In the case of PhD students it's even more egregious because maybe half of them actually *do* know what is fucked up and how but are also too smart to involve themselves in the clusterfuck.

>> No.15854444

>>15854423
I blame the "hurr durr let's manage everything with projects just like companies because companies are efficient" approach. Everybody is running after the hot new thing to get funding. We need a return to the Ivory tower approach. Academics should be aloof and disconnected from everyday reality, that's the only way to get good research done.

>> No.15854494

>>15853919
With college degrees being like high school diplomas now, a master's degree is like an old bachelor's degree. I think it's probably better to do slowly or after a stint of working. May be i'm wrong, but banging out a bachelor's degree and getting a job in the field then starting a master's seems better than just going bachelor's to masters immediately. Once you get the master's you'll already have experience and possibly also be working in the field while you're going to school. Internships might help, but nothing beats just having legit job experience in the field to employers. With job experience out of the way you can secure a job easier and then start trying to search for even better jobs.

>> No.15854517

>>15854444
Thing is, companies have a shitload more funding. It's not even close.

>> No.15854522

>>15854517
Companies have to produce something useful to justify being funded. "Scientists" just suck on the government teat and steal that money from the unwilling taxpayer.

>> No.15854581

What I've gotten is that as someone with lab, construction, and machine shop experience looking to do a PhD, I would be invaluable to almost any department

>> No.15854724 [DELETED] 

join the official /sci/serverino
https://discord.com/invite/xCZfa7zM

>> No.15854870

>>15854724
>become a discordtroon

I may be at my lowest point but I'll never sink to that level.

>> No.15855498

>>15854423
How would you change it, /sci/?

>> No.15856188

>>15855498
Academia should (still) provide the foundation onto which you build up your real world or industrial experience. Academia should still have some insight in what industry needs.

Secondly, industry should reign in HR that uses master's degree requirements as a cheap crutch to cimplify the recruitment process. Europe has a "masters disease" except Germany that moved the threshold to having a PhD.

Thirdly academia has to eliminate the current endemic fraud. It permeates academia, and is so deep rooted that it is hard to undo. Only mass investigations and decades of jail time will repair this. hopefully the LLM AI can be used to hunt down decades old fraud. Nothing good comes from fraud and total moral collapse, you will have to burn the filth down to the ground.

>> No.15856527

>>15856188
>Thirdly academia has to eliminate the current endemic fraud
This is pretty much impossible to do, because the entire system gives incentive to do only novel research (so no reproduction to check if previous research was correct), and because research that seems more groundbreaking is more likely to be cited.

>> No.15856774

>>15856188
>Academia should still have some insight in what industry needs
You work for a PI who's has an objective to keep industry money flowing in. They know what the current industry wants, this isn't necessary what the industry wants in a decade or so though.
>German PhDs
They started revoking some of the PhDs due to surfacing accounts of plagiarism & lack of developing a science. Granted you have to be rich to pull this off.
>fraud
Kind of a funding issue, i.e. you have to put out x amount of research paper over this time period rather than heres some money develop something new, if you don't that's ok. Couple this is Administration taking there pound of flesh to become a hedge fund for the 'future generations of students' and you have some of the most educated people on the earth becoming over worked and facing unemployment.

>> No.15856855

>>15854423
>>15855498

As I see it, this all revolves around the fact that there is no long-term academic research position which is not a tenured professor. These professors are also not de facto researchers anymore, they're more accurately described as managers. As noted, nobody hires technical staff anymore.

This means that all the actual research is done by PhD students and postdocs. To keep the wheels spinning requires a large number of these low-paid slaves (who sometimes even pay to work!), much more than can ever make it to the coveted professorship. They are all on a timer, forced to find a new job every few years and eventually getting smoked out entirely.

This, in turn, makes the competition so fierce that you HAVE to play the game of h-index massaging and politics if you ever want to stay in research. So you have to publish shit that doesn't warrant publishing, and not waste time on things that can't be hyped. Therefore the kinds of people who would e.g. reproduce results of others simply do not stay in the system and everything is flooded with garbage.

If there existed an appreciable quantity of permanent researcher positions which were more senior and secure than postdocs but less managerial than professors I feel a lot of these issues would be alleviated. First, there would be less need for a constant influx of PhD students hypersaturating everything just to keep the data flowing. This, in combination with more long-term research positions, would make the competition less extreme. Therefore not every long-term researcher would need to minmax their h-index and journal editor connections. Academia would not be a dead-end career path for 99% as it is now. As a bonus, hands-on research would get contributions from people who may have more than three years of experience in what they do.

The world is full of highly skilled autists who would happily do this for any liveable salary, myself included. Most universities do not have such positions though.

>> No.15856869

>>15856855
On writing this, I started thinking about how I've spent years in an "up or out" career tube with uncountable overtime, a stupid amount of politics/asslicking/corruption, endless need for self-promotion and bullshitting and 900 simultaneous projects. I realized I might as well have done management consultancy, probably would be surrounded by as many rich sociopaths and at least I'd be getting paid.

>> No.15857666

>>15856527
>This is pretty much impossible to do
I know. the question was a bit hypothetic so I replied in the same spirit.
If the Altzheimer fraudsters are sent down for 100 years in jail will we see some hope. It is sadly more likely they will get off the hook, thansk to massive networking (as in organised crime).

>> No.15857978
File: 729 KB, 684x1938, 1598017108-20200821.png [View same] [iqdb] [saucenao] [google]
15857978

>>15856855
>As I see it, this all revolves around the fact that there is no long-term academic research position which is not a tenured professor.
There is one exception: the postdiocs that keep hanging on for 10+ year, hoping to get that halfway promised tenure, just one contract more, please.
>This means that all the actual research is done by PhD students and postdocs. To keep the wheels spinning requires a large number of these low-paid slaves (who sometimes even pay to work!), much more than can ever make it to the coveted professorship.
True. It is modern slavery.
>This, in turn, makes the competition so fierce that you HAVE to play the game of h-index massaging and politics if you ever want to stay in research.
The hope was that people should have at least some integrity, especially when it comes to reserarch that has real impact on people's health.
>If there existed an appreciable quantity of permanent researcher positions which were more senior and secure than postdocs but less managerial than professors I feel a lot of these issues would be alleviated.
There are the research fellow positions thogh.
>Academia would not be a dead-end career path for 99% as it is now. As a bonus, hands-on research would get contributions from people who may have more than three years of experience in what they do.
Once the rot has set in it will be hard to change. It is a classic case of moral hazard so the fraudsters win.
>The world is full of highly skilled autists who would happily do this for any liveable salary, myself included. Most universities do not have such positions though.
Same here. With basic pay I would probably still be working in labs, and happy with that.

>> No.15859146

>>15857666
Blatant fraud isn't even the most obvious issue, it's more the grey area where people P-hack, or otherwise make their data slightly more enticing than it really is

>> No.15861845

>>15859146
Western blot issues are commonly seen, then again this is a practical way to uncover fraud.
I am hoping LLMs can be used to hunt down fraud. Imagine 100 years of publications examined, and millions of papers retracted? The retraction process alone will take decades.

>> No.15861906

>>15861845
>>15859146
There's different levels of this. To put it coarsely, science is stratified and the pajeet paper mills have separate issues than the "real" research groups anyone cares about.

Looking at PubPeer or Retraction Watch will give you an idea about the actual fraud that is called out. At the most obvious levels this involves researchers from e.g. China/India/Middle East putting out AI-generated/paraphrased plagiarism or simply fabricated data. The journals are usually very low-quality, and the purpose is just to pad their numbers with the expectation that nobody will actually read the papers. I think these kinds of papers actually make up a decent fraction of published science, although nobody serious reads the journals.

Outright fabrication/plagiarism gets rarer in good journals but definitely still happens. Sevveral high-profile researchers have built their careers on falsified studies. Calling these out gets political very fast, unless it's the most clear-cut case ever or you have a massive science dick it is career suicide. There's also reasonably respectable but not super high-profile researchers that take shortcuts, I think a lot of those fly under the radar since it often appears legitimate and isn't interesting enough for anyone to go over with a fine tooth comb.

I agree with anon that the grey area is a much bigger issue than outright fraud in these cases. It's very common that you have data which partially or wholly invalidates your conclusions that you just leave out because it won't get published otherwise.

A separate issue is citation machinations. Basically if you have a good collaborator network your work will get cited more even if it is bad or mediocre. Again at the most blatant levels it's pajeets block citing quantum gravity papers in cell biology publications, but it definitely happens at all levels. Chinese groups make a point to cite the chinese researchers, western groups ignore relevant chinese researchers half the time.

>> No.15862026

>>15852131
I can't speak for plebs in other fields but in my case (robotics) doing a PhD is basically working in a senior C++/Python dev/tech lead role where I also get to write publications about what I do and don't have to listen to faggot pajeet business analysts.

I don't see how it's gonna hurt employment prospects, it's literally just work experience. Plus you actually need a PhD for a lot of CTO or senior engineer roles at robotics companies.
>>15852918
nobody will hire your ass for bleeding edge work if you don't at least have a bunch of publications under your name.

>> No.15862049

>>15852131
The only reason I will pursue a PhD is to get a foot in US companies through research internships etc. The moment I'll get an offer from big tech I'll get the fuck out of my PhD.

>> No.15862054

It's not a great idea if you are like me and persecuted for being yourself (by girl bullies and effective altruists mostly)

>> No.15862970

>>15854277
lol how about you are the mind (cringe beta shut-in nerd) and write me all my papers and I'll be the muscle (chad vitamin D charged out goer INdiana Jones+mad scientist) and do all the in and out of lab experiments and just write the Materials and methods section? That's what I hope to do in my PhD anyway
>>15854310
any half-decent lab has at least one lab tech if not many, only poorfag labs don't have them (second-third world labs, Profs with no grants).
>>15854367
lol. If you need to do relatively complex lab procedures your prof is a tard for not even teaching people how to. Lab tech is the easiest route to get a job in biology since they are such important part of any respectable lab, if you get a lab tech certification you are fucking golden and can enjoy your life as a lab rat making decent money for the rest of your life. Labs literally keep these guys doing the same routines over and over forever, they offer you 1-2 year contract at first and if you are capable they beg you on their knees to spend your life with them
>>15854444
Utopic, who give people conducting meaningless research money? Maybe if machines truly take over menial jobs, maybe then it could happen, everyone will just do whatever they please

>> No.15862998

>>15852918
>lil bro
this board has indeed been ruined by the melaninated

>> No.15863014

>>15852196
Only handful of AI PhDs actually work on revolutionary stuff (Google, OpenAI, NVIDIA). Most of them just try to make startups that become pointless once OpenAI add new function to ChatGPT, or they just cope with tiny Meta leaked models that cant do shit despite what people would like to convince you otherwise. The main thing with AI is that the vast overwhelming majority of the money is not used for human labor working on it, but on GPUs training these models and even the human labor part is mostly just piss poor wagies handling the training data and doing reinforcement learning shit. You can still at least do something in academia however, like with literally every phd from every major.

>> No.15863021

>>15862970
>Utopic,
This was how academia ran before the whole project-based funding became prevalent

>> No.15863024

>>15862970
>That's what I hope to do in my PhD anyway
There is no confidence like that of the untested

>> No.15863034

>>15863024
I have happily worked for years in total doing lab work and fieldwork during my studies, all of it for free. They are going to pay me to do it now, AND I'll get to do stuff in the amazon forest? yeah I think I'm gonna be good for a while. Fuck being a neet.

>> No.15863068
File: 30 KB, 514x352, 7fe.png [View same] [iqdb] [saucenao] [google]
15863068

>>15863034
>I have some research experience, my PhD is gonna be SO easy

>> No.15863105

>>15863068
doing something you like makes it easier, yes

>> No.15863426

>>15863105

I found that doing something for a living is a really good way to stop liking it.

>> No.15863436

>>15852131
Nah just do a masters for a professional degree

>> No.15864285

Only do a PhD if you want to change the world, don't do it for money

>> No.15864384

>>15852131
>Does a PhD in STEM actually help if you're not planning to continue in science afterwards?
Yes
>>15852131
>How well do the skills translate outside of it?
Poorly
It helps by making people hire you on the basis that you're smart, without putting any thought into it

>> No.15864429

>>15864384
>It helps by making people hire you on the basis that you're smart, without putting any thought into it
That sounds like a recipe for disaster

>> No.15864524

>>15857666
Who are the Alzheimer's fraudsters?

>> No.15864572

>>15864384
what jobs require you to be smart?

>> No.15864577

>>15852131
a PhD is not job training, it is for people who love a subject. if your object is money then go do something else

>> No.15864591

>>15852131
Not all STEM is equal, anon

If you are doing mathematics or something quantitative and programming heavy, you can absolutely find other uses for it, either sell your soul in finance for disgustingly excessive riches or you can find plenty of other high paying jobs doing more or less interesting things.

If you're doing zoology or some shit, yeah you are fucked and should not do a doctorate unless you're sure it's what you want to do forever.

To put it another way, there is a pretty solid demand for jobs that require math statistics and programming up to the level of CFD, which is really a master's level course but the jobs still require a PhD, field of specialization doesn't matter but to show you can use the skills in research before they hire you and pay you 6 figures.

>> No.15864885
File: 1.89 MB, 2560x3840, chickens1.jpg [View same] [iqdb] [saucenao] [google]
15864885

>>15856855
Community colleges could out compete grant money from traditional R&D facilities by paying a living wage to actual researchers, not manager professors, and not burning up grant money on administrative fees. This would provide permanent positions to a lot of researchers.

This wouldn't work for every field, no particle accelerators, but it could cover a lot of mathematics and simulation work.