22

We get occasional questions here from people who are applying to PhD programs, and are worried because they have no publications. This is puzzling to me, because most of my friends in PhD programs did not have any publications when they applied.

We've also previous agreed that it is not generally mandatory to have a publication when applying to PhD, although it certainly helps. I am interested in knowing if there are any fields that are exceptions to this rule.

Are there any fields of study where (in some parts of the world, at least) most students have publications already by the time they apply for PhD admissions?

By "most," I mean that publications are common enough that a lack of one is seen as worrying, and possibly indicative of some deficiency in the applicant.

  • 1
    @mmh Is that standard in neuroscience? – ff524 Aug 27 '15 at 7:31
  • 3
    Just anecdotal evidence. I think most students have at least one paper in preparation/submitted/accepted related to their master's when applying. I'm in Europe. – mmh Aug 27 '15 at 7:37
  • 1
    It seems to be cultural. I may be wrong but it seems to me that US undergrads/master students have lot of free time and are expected to produce something of value on the side. Here in France you are not expected at all to publish anything in Math until after your M2. The courseload is expected to fill your time and only the most brilliant of all students sometimes publish tiny things. – aussetg Aug 27 '15 at 9:47
  • 3
    In software engineering in central Europe, having a reasonable paper before starting your PhD is uncommon enough to be considered a very good thing to have. So no, it's certainly not common or expected. – xLeitix Aug 27 '15 at 11:29
  • 2
    Not confident enough to leave an answer, but: I suspect that the expectation of publications is stronger in Europe than in the US. In Europe: (i) the undergraduate degree is more specialized; in some places and programs it is more similar to the beginning of a US graduate program than a US undergraduate program, whereas many US university students do not study X, they major in it, often taking less than half of their courses in that subject. (ii) It is common to mandatory to do a separate master's program. So European PhD applicants may have the equivalent of 3-4 more years of training. – Pete L. Clark Aug 27 '15 at 12:40
5

I'll point toward the Max Planck School for Computer Science, IMPRS-CS, if I remember correctly they used to have even stronger wording than the current

A successful candidate must... have performed research and published (or submitted for publication) the results

But the MPIs are not the norm. Mostly, you would be fine with just your Masters thesis in math, comp sci, biology, or chemistry.

4

I’ve been a part of a PhD admissions committee for a top school in the US in a field of engineering. Expectations change based on your educational background.

The lack of any accepted/submitted/in process publications is a significant weakness in any applicants that hold an MS. It is generally expected that during this time you were performing research, and its quality was sufficient for a peer reviewed conference or journal.

In undergraduate applicants who are applying directly to the PhD program, the lack of publications is not worrying whatsoever. There is generally little opportunity or time for them to be produced.

As has been said elsewhere, applicants are ultimately judged on their ability or potential ability to perform self-driven, impactful research. An undergraduate who has clearly made good use of their available time (summer internships involving research, extracurricular activities in the sciences, etc.) with no publications will most likely be viewed as a better candidate than an applicant with an MS with no publications. An MS alone in this context is at best evidence of more successfully completed coursework, which is not what committees are looking for. At worst it can suggest deficiencies exist in their research ability.

  • 3
    "It is generally expected that during [Master studies] you were performing research." - could you add a note about the rough geographic area where this applies? For instance, in Germany, depending on the field, Master students rarely do research; they attend classes, participate in lab sessions and projects, and take exams, just like during their Bachelor studies, but with an increase in depth and difficulty of topics. – O. R. Mapper Aug 15 '16 at 10:21
  • This answer is consistent with my experience in computer science in the US. – JeffE Aug 15 '16 at 11:01
  • 1
    In the US there are typically two options for Master studies: course-work only, and thesis track. The latter involves a research project (and a written thesis). – user58322 Aug 15 '16 at 11:09
  • In addendum: In the case of, for example, an applicant with a German MS to a US PhD program, the organization of their typical MS program will likely be considered - but they will very possibly still sit at a disadvantage to their peers with MS publications. Evidence of research ability remains the most important applicant criteria. – user58322 Aug 15 '16 at 11:22
3

For the context, I'm a Canadian doing a PhD in Biology.

In my University and for most Québec (Canadian?) universities, PhD applicant usually have a publication under way, whether it is published, accepted, submitted or in preparation.

The reason why is that most student have completed a master degree before. Altought publication is usually not mandatory in master degrees, it's becoming the norm.

I would dare say that this answer could fit countries and faculties/department where a master degree is required.

  • Interesting. Do you have any rough numbers? (Look at the ten people sitting closest to you, how many of them had a publication underway when they applied?) – ff524 Sep 1 '15 at 7:28
  • Hum...all? OK, probably 9/10, accounting for student from foreing countries who didn't do masters. However, some of those publications might take a long time to get accepted and really published. So the student will have it on his cv during his PhD. I would like very much if someone with the same background could back me up or not. It's a question hard to answered unbiaised. – Emilie Sep 1 '15 at 11:43
  • "I would dare say that this answer could fit countries (...) where a master degree is required." - wouldn't this only be a valid assumptions for research-based Master degrees (as opposed to, say, coursework-only Master degrees)? – O. R. Mapper Aug 15 '16 at 10:20
0

I would add that this is also dependent on the region. I.e. scholarship applications (for a phd) request your list of publications, though the average applicant will only have domestic publications (albeit often several of these), which are on a completely different level (i.e. many of these don't have a real peer-review process).

It should also be noted that what a 'masters' means is different in various countries: in Switzerland and I think most of Europe this is essentially a mandatory part of your studies before entering the industry, and the entire BSc+MSc will last 4-5 years (3+1.5 usually). In many other parts of the world such as the USA, people generally go into the industry after a (4 year) bachelors degree, and the masters is instead an integral (if often optional) part of a phd. Hence, it is much more common that people publish during the masters.

0

In most subfields of computer science, most strong applicants will usually have some sort of publication record, usually with one work under submission/review at the time of application. First authorship is usually not common, though.

  • 4
    Perhaps you can give some context, and some numbers to support your assertion. Have you served on a PhD admissions committee in computer science? If not, on what do you base this answer? – ff524 Aug 15 '16 at 6:23
  • This is based mostly on my experience as an applicant this past year, and the context is the traditional top four: Stanford, UCB, MIT, CMU. – Pratyush Mishra Sep 13 '16 at 15:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.