22

Suppose that a PhD (in engineering) student has 3-4 review/survey papers in their domain, and has received more than 500 citations in total. In a situation like this, can they pass the oral defense and get the PhD degree without a research paper?

What I want to know if review/survey papers can compensate my lack of research output. Any examples and stories that I can refer to?

8
  • 26
    I would say no. A lit review is usually just a chapter in the thesis. It is not counted as a contribution. My advice is to have a look at the examination criteria of your PhD program. At my university, a lit review is only one out of 10 criteria; i.e., if you only have a lit review, you will get 1 mark -- a bad fail. Sep 3 '20 at 4:52
  • 4
    It seems not a very realistic situation.
    – Alchimista
    Sep 3 '20 at 5:03
  • 2
    1) If you know someone who did, the answer is a clear yes, so please post it! 2) Keep in mind that phds are often awarded without any papers (your comment sounds like he/she did some new things but they were not accepted in a journal. 3) In most of the cases, you only have to convince the supervisor -- the commitee often doesn't want to oppose them (for political reasons).
    – user111388
    Sep 3 '20 at 5:26
  • 2
    Review papers are often by invitation these days, and it will be well-established researchers that are asked to author these, not PhD students. A paper is also unlikely to rack up 500 citations by the end of your PhD.
    – RPM
    Sep 3 '20 at 21:11
  • 1
    Prof. Santa Clause, Capt. Emacs, Tom -- please consider converting your comments to proper answers; answers-in-comments are subject to deletion.
    – cag51
    Sep 3 '20 at 22:44
74

It is possible to get a PhD without writing any papers; the formal requirement is of writing a thesis, and many people do just that. So I think the formulation of your question is a bit misleading.

But cutting to what I think is the actual intent behind your question rather than the specific choice of words, the answer is almost certainly that you cannot get a PhD without doing original research. Just making expositions of research already done by others, no matter how detailed or how many citations you get for them, is not what a PhD is about and wouldn’t satisfy the requirements.

9
  • 17
    @user2768 let’s not get carried away with the academic’s bad habit of qualifying every statement with endless caveats and asterisks. I’m sure one can while away many hours arguing about what it means to “tease out new knowledge”, but it won’t change the fact that you can’t get a PhD for writing survey papers.
    – Dan Romik
    Sep 3 '20 at 7:20
  • 2
    I respectably disagree, I think you could get a PhD from the "right" survey papers. For instance, surveys revealing and plugging significant holes in the literature. Especially when holes would likely not be identified without surveying. I certainly wouldn't recommend that route! The OP can chime in on whether there is any novelty in their surveys.
    – user2768
    Sep 3 '20 at 7:44
  • 4
    Thoroughly reviewing the literature can yield new analysis that would be considered original research, but in many cases it would be a particularly inefficient way to proceed. Imagine you want to know if sugar improves tea taste and no single paper exists on the subject. Option one: test (recruit volunteers, make double-blind taste ratings). Option two: find a newspaper survey about sugar in coffee, a figure of a physiological study of tea effects on the brain, and an appendix table of a food chemistry interaction guide, and try to glue it all together. I know which I would prefer...
    – UJM
    Sep 3 '20 at 12:27
  • 8
    In my lab in grad school, it was very clearly expected that a person defending their thesis would be able to plainly answer the question, "What is it that you are doing that is new?"
    – Tristan
    Sep 3 '20 at 15:07
  • 5
    Note: the formal requirement depends on country (and possibly subject within that country?) and does in some countries require you to have published. Sep 3 '20 at 16:35
22

The naive answer is yes. All you have to do is convince a panel that what you did warrants a PhD. This will be different for each school/department/panel/student. Typically this requires contributing something novel to the field. Something novel does not have to be a research paper.

A couple examples:

  • A colleague wrote a review where he extracted related data from a multitude of papers. He did this for a single minor figure in his review. Turns out others wanted that data already aggregated and it spawned many more research papers. He ended up creating a website/database and maintained it throughout the rest of his PhD. This database was a significant portion of his PhD (and probably his strongest portion).

  • Another colleague created a machine to automate a particularly labor intensive process in the lab (2 years of development). This was then used by many other researchers in their own novel research papers. Again this became a large portion of his PhD, and made him very popular among those who wanted to use it.

It doesn't have to be a research paper but it does need to be novel and contribute to the field. A review in and of itself is not necessarily novel, but it can be. As @JonCuster pointed out, the 'new' stuff was not the review nor a research paper but new tools that helped the community do further research.

3
  • 21
    In your two cases, each created a new tool that was a step forward in the field - that was the 'new' stuff they did, and indeed those steps are items worthy of being part of a PhD.
    – Jon Custer
    Sep 3 '20 at 14:14
  • 2
    @noslenkwah Jon Custer's positive comment got more upvotes than your answer. Consider incorporating it. Sep 4 '20 at 5:46
  • 2
    @CaptainEmacs - Agreed. Done.
    – noslenkwah
    Sep 4 '20 at 12:59
5

Information Systems professor here. I chair a dissertation every year or two. Short answer is no.

(1) You need to prove that you can independently execute the scientific method and complete a sufficiently complex research project that extends the body of knowledge in your area.

(2) I wouldn't be doing you any favors if I let you leave the program without a good body of work that you could use for conference papers and grant proposals.

0
4

Survey papers are useful and will give you citations if well written.

But for a PhD, you need to add to the existing body of knowledge, not merely reorganize it. Unless it is a massive contribution, such as a reformulation or neater repackaging of existing knowledge (for extreme examples, consider parts of Euclid, Cartan, Wielandt, the latter of which merely found a shorter proof of a known theorem, but it was much shorter), which itself is a scientific contribution, it's a very clear no.

In fact, if you are able to create a simplified map through a difficult and ill-understood landscape, that's definitely a PhD.

Others talked about new tools as worthy of a PhD, in a way, this is a similar situation. Identifying what makes a field hang together is a new tool. Done well, it can change a field.

However, this is most likely to be far harder than to pick a concrete problem and study it. However, it is also far more than just a literature review.

3

Your advisor and your university determine whether your work meets the requirement for the degree. That's where to go for guidance.

I suspect (as @DanRomik says) that a literature survey, no matter how thorough and useful, will be insufficient.

0

It seems to me that what you're actually asking is can a person be classed as a researcher if they only write review papers. I think that you know that the answer to that is no, unless the review papers have significant research results embedded in them. Generally though, review papers are requested, and just randomly writing up a survey paper and getting it published is very difficult, and actually not recommended unless you are an expert and have been studying the subject for a long time.

-1

You might be able to get a PhD in metascience.

To quote Wikipedia, metascience is "the use of scientific methodology to study science itself", performing research on the process of research. Part of this process would involve writing a bunch of literature reviews in order to analyze aspects of the scientific literature to determine information about the scientific process. For instance, you might code a thousand recently-published Information Technology papers to determine what the most popular scientific methodologies are, and how the methodology chosen affects other aspects of the paper like reproducibility or the availability of software artifacts produced during the process of writing the paper.

As a result, when you perform a metascientific analysis of the literature, you're not just taking existing knowledge and packaging it into a more convenient location, but producing new knowledge about the scientific process.

This isn't just speculative, either; when looking up the publications of authors who created popular scientific methodologies and gotten thousands of citations on their papers introducing them, you'll sometimes find that they've created a career out of publishing papers on metascience and scientific methodologies.

2

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.