47 ms·

No paper, no PhD? India rethinks graduate student policy

66 points by NN88 7y ago

mcguire 7y ago
This seems like a poor answer to the wrong question. The problem is not that students are being preyed on by publishers, the real question is why they aren't publishing in reputable journals and conferences. I was a volunteer for the IEEE SoutheastCon one year. It was amazing; almost none of the authors showed up for the conference. It was enough for them to get the publication and enough for the IEEE that they'd payed the registration fee.
- Fomite 7y ago
  Because the incentives aren't aligned. If the requirement is "One paper", then why would you try to publish in a reputable journal, go through peer review, need to revise and resubmit, etc. when you can just pay your money and check the needed box.
  - mcguire 7y ago
    Because that's what you need to do for a career in research and teaching. -- the optimist, probably
- physicsguy 7y ago
  In my field it’s standard that if you don’t show up, your paper is not published in proceedings - and normally those proceedings are in an IEEE Transactions journal.
  - mcguire 7y ago
    There's apparently quite a bit of difference between the single-subject conferences and the regional, general conferences. The latter are primarily about funding the regional groups, while the former are only partially about funding the groups (yeah, I'm still bitter about ICDCS '98).
- musicale 7y ago
  I've seen some "official" IEEE conferences that really look like junk conferences/spamferences.
anbop 7y ago
One thing to note here is that India has massive credential over demand. An international hotel can require Master’s degrees in English for its front desk staff. So it doesn’t matter that these PhDs can’t do original research because 99% of them won’t be.
- noobermin 7y ago
  As usual then it seems they are trying to solve the wrong problem. By "as usual" I mean it's like a lot of problems in society, we try to fix the symptoms instead of the illness.
- booleandilemma 7y ago
  So pretty much where the US is headed then? (I’ve met a technical support person with a master’s degree in CS)
  - malshe 7y ago
    I met an Uber driver who was doing his PhD in CS and specifically in machine learning!
    - slashcom 7y ago
      That should tell you how grad students are paid.
  - rxhernandez 7y ago
    As a counter-anecdote, I've met a bunch of highly paid software engineers in the Bay area with no degree.
- valarauko 7y ago
  Isn't the demand due to an overabundance of highly qualified applicants in the job market? Employers see no harm in asking for preposterous qualifications, and there are enough desperate people looking for jobs.
carlob 7y ago
Wouldn't it be somewhat better to maintain a list of international peer reviewed papers for each discipline? My country (Italy) has something of the sort for evaluating prospective full-time researchers. It doesn't even really need to be based on impact factor, especially in what are considered non-bibliometric disciplines like humanities.
vikramkr 7y ago
I have mixed feelings on this. On one hand, I like most things that involve moving away from the publish or perish culture in academia. On the other hand, this doesn't seem to be a productive way to do that. This seems to be trying to attack poor quality publishers through an indirect method which likely will not solve the problem - people are still incentivized to publish in other ways and the root cause needs to be hit to make any real progress. I also think that having at least one publication during an entire PHD is not an unreasonable request - ending publish or perish shouldn't be pursued by trying to curtail the importance of publishing overall since it is still a critical part of the scientific process. There need to be good journals that accept null results and in progress results and replication studies so scientists can continue to be incentivized to share their work without forcing them to build their life around achieving something "publishable" in the current paradigm.
williesleg 7y ago
Just do the needful and send your cash.
kazinator 7y ago
That proposal basically lowers the value of an Indian Ph. D. If you already have one and had to publish, you should be upset at a proposal to have your existing credentials blemished in this manner.
- tsjq 7y ago
  i fail to understand why is this comment downvoted .
musicale 7y ago
US universities seem to do fine without a government-mandated Ph.D. publication requirement. However, helping grad students publish and present their work in a high-quality, peer-reviewed context is one of the most important things that graduate advisors can and should do. I am certainly OK with institutions making that a requirement for advisors.
- sytelus 7y ago
  US system computes hell out of statistics for ranking and validation of quality. Other countries, including India or China, has no such things. There, universities/institutes are "ranked" by sort of public perception and media image. It is not uncommon in these countries to buy PhD degrees by simply contracting ghost writers and/or publish in to shady for-fee publications. Government mandates have typically evolved in these countries for a reason.
  - jldugger 7y ago
    > US system computes hell out of statistics for ranking and validation of quality. Other countries, including India or China, has no such things. There, universities/institutes are "ranked" by sort of public perception and media image. Having worked in university administration, I can assure you these two things are not so different after all. US News ranking, is for example, a survey of professors and high school councilors, plus some reporting about things like overall budget per faculty member and per student, and yes, some statistics about the incoming class. And... the ghost writing problem is not unheard of in the US. After all, a lot of PhD candidates hail from India and China, so this shouldn't surprise you overmuch. > Government mandates have typically evolved in these countries for a reason. I suspect it's because of government mandates in these countries and others. For example, advanced degrees help in obtaining visa status in the US.
notme77 7y ago
I'd like to see more professional, terminal (doctorate) degrees that aren't PhDs. Run them in parallel. For instance, I feel there should be an M.Eng. equivalent for comp sci that requires mastery of existing theory instead of development of new theory.
- SketchySeaBeast 7y ago
  That's literally what a PhD is for - the fact that you're an expert in your field and breaking new ground. Wouldn't a masters in comp sci be the equivalent to a masters in engineering? You can get a PhD in engineering as well.
- jefft255 7y ago
  You can simply do a course-based masters for that... same classes as a PhD student, no need to produce original research.
- adestefan 7y ago
  There exists a handful of EngD problems out there. It’s to be treated as an applied PhD where the goal is to develop as a leader of industrial R&D.
- sytelus 7y ago
  Exactly. There is something dearly missing between Masters to PhD. Many European PhDs are now 3 years while US PhD tends to be 5 year. It would have been great if PhD degree standardized to 5 years but was broken down into two separate degrees of 2+3 years.
  - schuetze 7y ago
    Although the time to completion seems quite disparate between Europe and US, the difference isn't nearly as stark as it appears. Nearly all Europeans enter PhD programs with a Master's degree in hand. At least in psychology (my discipline), US PhD programs will admit candidates directly from undergraduate. The overall time from bachelor's to PhD is similar when you account for these differences.
chriskanan 7y ago
A PhD in a scientific or engineering discipline is a certification that you know how to conduct research and communicate the research you conducted. Papers are the evidence people use to assess that ability. Graduating without them makes it unclear what was accomplished and if the necessary skills were acquired. Informally, most top programs in the US have students produce 2-6 original papers to graduate, which varies per discipline. Without publications, it will be very hard for graduates to get a job doing research, which is the main career door that a PhD opens. I assume that thought process was what motivated the policy forcing people to publish to graduate. However, very low quality publications are equally useless and do not showcase one's ability to do research, so just requiring publications isn't sufficient. A PhD student's committee is supposed to assess if they have demonstrated the ability to do good research and add to humanity's knowledge. It doesn't sound like that is happening appropriately.
- sgt101 7y ago
  Look - it's straightforward, is there a genuine contribution, is the thesis of sufficient quality to effectively communicate that - there's the standard. Who cares what random referees have / have not decided about the work if you are examining it?
  - splittingTimes 7y ago
    Not sure what profession you refer to but in physics you publish to specialized journals in your field of research. The referees typically have some, if not a lot of expertise. It's not like a random guy is giving a thumbs up or down. It is a critique of your scientific work. Having been a referee for a long time I can tell you that so much submissions are just minuscule variations on existing papers that are sugar-coated to be ohh so important contributions to the field that it is mind boggling. And if you decline it, they try the next journal with a lower impact factor until they get published.
    - apathy 7y ago
      This is accurate in every field I’ve worked in (synthetic organic chemistry, comp chem, statistics, genetics, clinical trials). The top glam journals have an additional layer of politics. It’s the worst system aside from all the others that have ever been tried... (last time I was actually excited to review was for NEJM & was disappointed)
kevinventullo 7y ago
The first publication related to my thesis appeared nearly a year and a half after I graduated. Part of the reason is that, at least in pure math, it might take 3-6 months to receive a rejection!
- semi-extrinsic 7y ago
  I made the error, early in my PhD, of naming the git repo for one of my papers "prestigiousJournalYearX". It didn't end up in that journal, and it took something like two years more than Year X. Not a good idea, with the constant reminder of falling short of your earlier goals, and I never could get myself to rename it either.
- omaranto 7y ago
  You are so lucky! I've had math papers rejected after a year!
- gradstudent 7y ago
  I regard 3-6 months as pretty good for journals. In CS competitive journals can take a year to reject something. Even if a paper is accepted, and the period of time from submission to publication can sometimes take close to 2 years. CS conferences are nice though. Only 6-8 weeks!
hooloovoo_zoo 7y ago
It's important to note that publishing a paper can take quite a long time. For instance, it quite commonly takes 2+ years to publish in JASA, the premier statistics journal.
analog31 7y ago
I agree with the others who say that publication is an important aspect of doctoral education. However, grad students already face too many sources of risk and delay. I've known a number of PhDs who had to deal with professors holding back on letting them graduate in order to get one or more papers out of them, at enormous expense to the student, and after the dissertation work was finished. In one case, a student had to lawyer up. I also know one professor who simply can't bring himself to let anything be published, and a student shouldn't be penalized for that. The professors who review the student's dissertation should be able to decide if the work is of publication quality. The dissertation is still the main focus of the PhD.
light_hue_1 7y ago
This is a huge step backward for Indian science and it's really sad to see. India already is far behind China and not even comparable to the US or Europe. Now they'll fall far further behind. You don't get your PhD in US/Canadian/European universities without publications. Not in physics, math, cs, psychology, neuroscience, etc. Publications don't just show you can do things, but they hold your advisers to account to make sure they're also pushing out new ideas, they hold entire departments to account. If a department can't produce enough new ideas and research to get 1 paper per PhD student out, then they are failing and should be dealt with by the university administration. Failing to graduate PhDs is a critical metric in finding such problems and dealing with them. What this does is just entrench India's failure in the research world. Lazy senior researchers will get lazier. Young researchers will have far less incentive to keep moving. Departments won't be held to account anymore. And low-quality PhDs will keep being minted. I'm not a fan of publish or perish, and it's damned stressful sometimes. But 1 paper per PhD student is not a burden. It's a bare minimum and advisers should be ashamed if this is all that they can publish with a student through their PhD. India is pumping out huge amounts of junk research, mediocre PhDs, and is just stagnant overall in research. This is short sighted, lazy, and will hurt everyone there.
- jldugger 7y ago
  > You don't get your PhD in US/Canadian/European universities without publications. This is not a national requirement however. I bet you could find a PhD granting institution in the US that doesn't require publication. Moreover, there's spillover effects from requiring this; I'm told in China that medical journals are facing a dire problem of research fraud, due to research publications being tied to promotions in the medical community (https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(15)60700-0/fulltext https://www.thelancet.com/journals/lancet/article/PIIS0140-6...). It seems reasonable to assume incentivizing publications will reduce the quality of publications, whether the incentive is a promotion or a degree. The question I have though, is what harm there is in reverting to a socially enforced standard. It's pretty easy to screen out PhD applicants without a publication, and I don't know of anyone who's too busy churning through PhDs to give their scholarly output at least a brief glance. If the screening device is available cheaply, who is harmed by this legislation?
Fomite 7y ago
As a faculty member who works with students in other developing countries where this is required, this is a great step forward. The requirement not only pushes students towards rapid publication, predatory journals but it also produces poorer quality science because the focus becomes entirely "What do I have to do to get just enough for that publication?"
tsjq 7y ago
that's interesting. so, what's gonna be the next batches of PhD? pass a few more exams , get that PhD degree, and add a Dr prefix ?

66 points by NN88 7y ago

This seems like a poor answer to the wrong question. The problem is not that students are being preyed on by publishers, the real question is why they aren't publishing in reputable journals and conferences. I was a volunteer for the IEEE SoutheastCon one year. It was amazing; almost none of the authors showed up for the conference. It was enough for them to get the publication and enough for the IEEE that they'd payed the registration fee.

Fomite 7y ago
Because the incentives aren't aligned. If the requirement is "One paper", then why would you try to publish in a reputable journal, go through peer review, need to revise and resubmit, etc. when you can just pay your money and check the needed box.
- mcguire 7y ago
  Because that's what you need to do for a career in research and teaching. -- the optimist, probably
physicsguy 7y ago
In my field it’s standard that if you don’t show up, your paper is not published in proceedings - and normally those proceedings are in an IEEE Transactions journal.
- mcguire 7y ago
  There's apparently quite a bit of difference between the single-subject conferences and the regional, general conferences. The latter are primarily about funding the regional groups, while the former are only partially about funding the groups (yeah, I'm still bitter about ICDCS '98).
musicale 7y ago
I've seen some "official" IEEE conferences that really look like junk conferences/spamferences.

anbop 7y ago

One thing to note here is that India has massive credential over demand. An international hotel can require Master’s degrees in English for its front desk staff. So it doesn’t matter that these PhDs can’t do original research because 99% of them won’t be.

noobermin 7y ago
As usual then it seems they are trying to solve the wrong problem. By "as usual" I mean it's like a lot of problems in society, we try to fix the symptoms instead of the illness.
booleandilemma 7y ago
So pretty much where the US is headed then? (I’ve met a technical support person with a master’s degree in CS)
- malshe 7y ago
  I met an Uber driver who was doing his PhD in CS and specifically in machine learning!
  - slashcom 7y ago
    That should tell you how grad students are paid.
- rxhernandez 7y ago
  As a counter-anecdote, I've met a bunch of highly paid software engineers in the Bay area with no degree.
valarauko 7y ago
Isn't the demand due to an overabundance of highly qualified applicants in the job market? Employers see no harm in asking for preposterous qualifications, and there are enough desperate people looking for jobs.

carlob 7y ago

Wouldn't it be somewhat better to maintain a list of international peer reviewed papers for each discipline? My country (Italy) has something of the sort for evaluating prospective full-time researchers. It doesn't even really need to be based on impact factor, especially in what are considered non-bibliometric disciplines like humanities.

vikramkr 7y ago

I have mixed feelings on this. On one hand, I like most things that involve moving away from the publish or perish culture in academia. On the other hand, this doesn't seem to be a productive way to do that. This seems to be trying to attack poor quality publishers through an indirect method which likely will not solve the problem - people are still incentivized to publish in other ways and the root cause needs to be hit to make any real progress. I also think that having at least one publication during an entire PHD is not an unreasonable request - ending publish or perish shouldn't be pursued by trying to curtail the importance of publishing overall since it is still a critical part of the scientific process. There need to be good journals that accept null results and in progress results and replication studies so scientists can continue to be incentivized to share their work without forcing them to build their life around achieving something "publishable" in the current paradigm.

williesleg 7y ago

Just do the needful and send your cash.

kazinator 7y ago

That proposal basically lowers the value of an Indian Ph. D. If you already have one and had to publish, you should be upset at a proposal to have your existing credentials blemished in this manner.

tsjq 7y ago
i fail to understand why is this comment downvoted .

musicale 7y ago

US universities seem to do fine without a government-mandated Ph.D. publication requirement. However, helping grad students publish and present their work in a high-quality, peer-reviewed context is one of the most important things that graduate advisors can and should do. I am certainly OK with institutions making that a requirement for advisors.

sytelus 7y ago
US system computes hell out of statistics for ranking and validation of quality. Other countries, including India or China, has no such things. There, universities/institutes are "ranked" by sort of public perception and media image. It is not uncommon in these countries to buy PhD degrees by simply contracting ghost writers and/or publish in to shady for-fee publications. Government mandates have typically evolved in these countries for a reason.
- jldugger 7y ago
  > US system computes hell out of statistics for ranking and validation of quality. Other countries, including India or China, has no such things. There, universities/institutes are "ranked" by sort of public perception and media image. Having worked in university administration, I can assure you these two things are not so different after all. US News ranking, is for example, a survey of professors and high school councilors, plus some reporting about things like overall budget per faculty member and per student, and yes, some statistics about the incoming class. And... the ghost writing problem is not unheard of in the US. After all, a lot of PhD candidates hail from India and China, so this shouldn't surprise you overmuch. > Government mandates have typically evolved in these countries for a reason. I suspect it's because of government mandates in these countries and others. For example, advanced degrees help in obtaining visa status in the US.

notme77 7y ago

I'd like to see more professional, terminal (doctorate) degrees that aren't PhDs. Run them in parallel. For instance, I feel there should be an M.Eng. equivalent for comp sci that requires mastery of existing theory instead of development of new theory.

SketchySeaBeast 7y ago
That's literally what a PhD is for - the fact that you're an expert in your field and breaking new ground. Wouldn't a masters in comp sci be the equivalent to a masters in engineering? You can get a PhD in engineering as well.
jefft255 7y ago
You can simply do a course-based masters for that... same classes as a PhD student, no need to produce original research.
adestefan 7y ago
There exists a handful of EngD problems out there. It’s to be treated as an applied PhD where the goal is to develop as a leader of industrial R&D.
sytelus 7y ago
Exactly. There is something dearly missing between Masters to PhD. Many European PhDs are now 3 years while US PhD tends to be 5 year. It would have been great if PhD degree standardized to 5 years but was broken down into two separate degrees of 2+3 years.
- schuetze 7y ago
  Although the time to completion seems quite disparate between Europe and US, the difference isn't nearly as stark as it appears. Nearly all Europeans enter PhD programs with a Master's degree in hand. At least in psychology (my discipline), US PhD programs will admit candidates directly from undergraduate. The overall time from bachelor's to PhD is similar when you account for these differences.

chriskanan 7y ago

A PhD in a scientific or engineering discipline is a certification that you know how to conduct research and communicate the research you conducted. Papers are the evidence people use to assess that ability. Graduating without them makes it unclear what was accomplished and if the necessary skills were acquired. Informally, most top programs in the US have students produce 2-6 original papers to graduate, which varies per discipline. Without publications, it will be very hard for graduates to get a job doing research, which is the main career door that a PhD opens. I assume that thought process was what motivated the policy forcing people to publish to graduate. However, very low quality publications are equally useless and do not showcase one's ability to do research, so just requiring publications isn't sufficient. A PhD student's committee is supposed to assess if they have demonstrated the ability to do good research and add to humanity's knowledge. It doesn't sound like that is happening appropriately.

sgt101 7y ago
Look - it's straightforward, is there a genuine contribution, is the thesis of sufficient quality to effectively communicate that - there's the standard. Who cares what random referees have / have not decided about the work if you are examining it?
- splittingTimes 7y ago
  Not sure what profession you refer to but in physics you publish to specialized journals in your field of research. The referees typically have some, if not a lot of expertise. It's not like a random guy is giving a thumbs up or down. It is a critique of your scientific work. Having been a referee for a long time I can tell you that so much submissions are just minuscule variations on existing papers that are sugar-coated to be ohh so important contributions to the field that it is mind boggling. And if you decline it, they try the next journal with a lower impact factor until they get published.
  - apathy 7y ago
    This is accurate in every field I’ve worked in (synthetic organic chemistry, comp chem, statistics, genetics, clinical trials). The top glam journals have an additional layer of politics. It’s the worst system aside from all the others that have ever been tried... (last time I was actually excited to review was for NEJM & was disappointed)

kevinventullo 7y ago

The first publication related to my thesis appeared nearly a year and a half after I graduated. Part of the reason is that, at least in pure math, it might take 3-6 months to receive a rejection!

semi-extrinsic 7y ago
I made the error, early in my PhD, of naming the git repo for one of my papers "prestigiousJournalYearX". It didn't end up in that journal, and it took something like two years more than Year X. Not a good idea, with the constant reminder of falling short of your earlier goals, and I never could get myself to rename it either.
omaranto 7y ago
You are so lucky! I've had math papers rejected after a year!
gradstudent 7y ago
I regard 3-6 months as pretty good for journals. In CS competitive journals can take a year to reject something. Even if a paper is accepted, and the period of time from submission to publication can sometimes take close to 2 years. CS conferences are nice though. Only 6-8 weeks!

hooloovoo_zoo 7y ago

It's important to note that publishing a paper can take quite a long time. For instance, it quite commonly takes 2+ years to publish in JASA, the premier statistics journal.

analog31 7y ago

I agree with the others who say that publication is an important aspect of doctoral education. However, grad students already face too many sources of risk and delay. I've known a number of PhDs who had to deal with professors holding back on letting them graduate in order to get one or more papers out of them, at enormous expense to the student, and after the dissertation work was finished. In one case, a student had to lawyer up. I also know one professor who simply can't bring himself to let anything be published, and a student shouldn't be penalized for that. The professors who review the student's dissertation should be able to decide if the work is of publication quality. The dissertation is still the main focus of the PhD.

light_hue_1 7y ago

This is a huge step backward for Indian science and it's really sad to see. India already is far behind China and not even comparable to the US or Europe. Now they'll fall far further behind. You don't get your PhD in US/Canadian/European universities without publications. Not in physics, math, cs, psychology, neuroscience, etc. Publications don't just show you can do things, but they hold your advisers to account to make sure they're also pushing out new ideas, they hold entire departments to account. If a department can't produce enough new ideas and research to get 1 paper per PhD student out, then they are failing and should be dealt with by the university administration. Failing to graduate PhDs is a critical metric in finding such problems and dealing with them. What this does is just entrench India's failure in the research world. Lazy senior researchers will get lazier. Young researchers will have far less incentive to keep moving. Departments won't be held to account anymore. And low-quality PhDs will keep being minted. I'm not a fan of publish or perish, and it's damned stressful sometimes. But 1 paper per PhD student is not a burden. It's a bare minimum and advisers should be ashamed if this is all that they can publish with a student through their PhD. India is pumping out huge amounts of junk research, mediocre PhDs, and is just stagnant overall in research. This is short sighted, lazy, and will hurt everyone there.

jldugger 7y ago
> You don't get your PhD in US/Canadian/European universities without publications. This is not a national requirement however. I bet you could find a PhD granting institution in the US that doesn't require publication. Moreover, there's spillover effects from requiring this; I'm told in China that medical journals are facing a dire problem of research fraud, due to research publications being tied to promotions in the medical community (https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(15)60700-0/fulltext https://www.thelancet.com/journals/lancet/article/PIIS0140-6...). It seems reasonable to assume incentivizing publications will reduce the quality of publications, whether the incentive is a promotion or a degree. The question I have though, is what harm there is in reverting to a socially enforced standard. It's pretty easy to screen out PhD applicants without a publication, and I don't know of anyone who's too busy churning through PhDs to give their scholarly output at least a brief glance. If the screening device is available cheaply, who is harmed by this legislation?

Fomite 7y ago

As a faculty member who works with students in other developing countries where this is required, this is a great step forward. The requirement not only pushes students towards rapid publication, predatory journals but it also produces poorer quality science because the focus becomes entirely "What do I have to do to get just enough for that publication?"

tsjq 7y ago

that's interesting. so, what's gonna be the next batches of PhD? pass a few more exams , get that PhD degree, and add a Dr prefix ?