47

I'm about to defend my Phd thesis in November 2016 in software engineering. I started my Phd in February 2010. One of my thesis reviewers stated in his review comments that my bibliography references are too old: the most recent work I have quoted in my bibliography is from 2012; that was when I was elaborating my thesis' state of the art. He stated that I have to check main works done in the field of research since 2012.

I am not excited about completing a new literature review two weeks before my defense. Thus, I was planning to say at the defense that the bibliography is old because my thesis began several years earlier. Is this an acceptable response? What would you do in this situation?

  • 124
    Well, basically you are saying that you didn't keep abreast of the literature over the last 4 years. You should not need to do a new literature review, since you should have been reading relevant papers as you went along. I would have an issue with not having recent references as well. – Jon Custer Oct 20 '16 at 17:16
  • 93
    Y'know what would be scarier than being told to refresh your lit review? Finding out -- during your defense -- that someone else has done, and published, "your" research. Refresh the lit review. – Bob Brown Oct 21 '16 at 0:01
  • 10
    @ChrisH even if the review simply hasn't been updated, "the most recent work I have quoted in my bibliography is from 2012" implies that also all of the other chapters done after the literature review have not (according to citations) taken into account any recent publications. – Peteris Oct 21 '16 at 10:29
  • 10
    A bit more related to @BobBrown's comment - Often at first glance it looks like someone else has done the same research, but on closer inspection it's clear that there are significant differences. You probably haven't had someone else publish this research, but it would not be at all surprising if someone could ask you "how is this different from the following paper published 3 years ago?" If you get that question and your answer is "I don't know that paper", you're in a LOT of trouble. If you get it and your answer is "they were focusing on this set of parameters, for which..." you're okay – Joel Oct 21 '16 at 20:02
  • 4
    @Joel Correct. That actually happened to me. I encountered a paper with work very similar to mine and went to my committee chair in a panic. He told me to include that paper in my lit review. I did so, with enough detail to show the differences. Problem solved. – Bob Brown Oct 21 '16 at 20:07
101

The criticism is that you're not up to date on the current state of your own field of specialty, and it is not a light matter. If the "reviewer" is some sort of committee member that can throw a wrench into your exit process, you should address the criticism.

The real issue is how to address it. My assumption here is that your defense will be "typical", with a formal presentation followed by some sort of closed session.

You have a very limited amount of time before your defense, and you need to use it wisely. To start, I'd pick the five most relevant citations to your work that you have, and search them forward to see who cited them, and read those papers-- there will probably be less than 20. If nothing earth-shattering has happened, you will eventually add those to your reference list, and all should be good. If there is something earth shattering and pertinent to your work, you probably have more research to do at this point.

How will your reviewer know you did it? Well, for the closed session, you should have a slide in your rack entitled "review of recent literature". In closed session, make the opportunity to say "in review, a criticism of how up to date my knowledge of the field was came up. This was accurate, and this is how I've responded... I think I've covered this, and will continue to stay up to date with the literature".

To add, they may require a bit of a rewrite-- that's actually fairly common-- but worry about that when the time comes. The strategy is to to convince them that you've done the work, and you're ready to do that rewrite. You don't have the time to do it between now and your defense, and my experience is that the document is largely sealed until it's defended anyway. You're trying to set things up so your revised dissertation will be signed off on without a thorough going-through.

  • 30
    +1 for the pragmatic suggestions instead of "Yew dunne screwed up." – user1717828 Oct 21 '16 at 13:34
  • 2
    Actually, I like this answer better than mine. I hesitated to give advice on a situation I was unsure about, but this answer seems to address what can be done accurately and realistically. – Dirk Oct 21 '16 at 16:47
78

First, if a reviewer stated that you should better do something, then you better do something (in most cases, do what the reviewer says, unless you have a very good reason to do something else).

Second, the reviewer is right here. You have to keep up to date with the literature, and provide a literature review in your thesis which is up to date at the time of submission. Four years is a pretty long time for fast-moving fields, and even for slower-moving fields it is still not negligible. On the other hand, refreshing a literature review for four years of new research is a doable task.

If it really is the case that you did not keep up with the literature in the past four years, it may be that you missed some important developments. Many things may have happened and I can't answer the question "what now?" without knowing more details.

  • 2
    I don't think it's always true that you must do something because the reviewer says so. It's not unheard of for them to say something should be done when there is a good reason not to. – Jessica B Oct 21 '16 at 6:48
  • 12
    @JessicaB: That's true in the general case, but it's hard to see how that applies in this specific case – Chris H Oct 21 '16 at 9:01
  • @JessicaB Agreed - since my I formulated my first paragraph in a general way, I took that into account now, thanks! – Dirk Oct 21 '16 at 16:50
  • @ChrisH I'm not disputing that. But it's stated in this answer as a general truth, and I'd expect others to read the question and answer at some point. – Jessica B Oct 22 '16 at 9:02
  • 2
    @JessicaB As a bit of background: At least in the German system, a reviewer can officially request (as opposed to recommend) changes in the thesis, and if the committee concurs (which usually happens, because this is a step not taken lightly), these become mandatory -- meaning that even if the defense is passed, the degree won't be conferred before a suitably revised thesis is (re)submitted. – Christian Clason Oct 22 '16 at 23:15
13

While writing a lierature review early is a good idea, it's a mistake to regard this as the literature review. The early literature review is better though of as something for your own use (plus your supervisor and any internal requirements).

It's quite possible that you've kept up with the literature but not demonstrated that in the thesis -- which is essentially the same as not putting the thesis in context.

From where you are now, it seems like you really should update (not rewrite) it. If you have kept up this shouldn't be too hard. Even if you haven't you could be lucky and find a good recent review paper that you can use as an index to developments with some context and weighting. Even if you're unlucky and there isn't such a paper, looking for it is helpful to writing your review. Check for recent/relevant publications from anyone involved in your defense.

Actually getting stuck in to recent literature on your subject is a good way to prepare; although the writing side of it is a chore, you're not in such a bad position as you seem to fear.

8

Many of the answers above are quite good, and I don't want to re-state them except to highlight things that they have already said:

  1. Surely you have been reading papers over the past 4 years, and have used this scholarship to enhance your own research. Part of your response to this reviewer can highlight this fact.
  2. You should take this reviewer's request seriously. This person gets a vote on whether you graduate or not. The only reason not to do so is that literally every other person on your committee thinks this guy is an idiot and that they will vote to support you regardless.

In addition to this advice, I would also add:

  1. You should also find out what your advisor thinks. Do what your advisor says.
  2. Do not waste too much time on this. Do good work, but do the minimum work necessary. Unless software engineering is wildly different from physics, this is not a "book field," that is, you do not get tons of credit from writing a scholarly book. Instead, peer-reviewed articles are the preferred form of scholarship.

My own thesis was printed and bound in a gorgeous black volume with gold embossing. It has sat on my shelf, lo these (not so) many years. Instead, I cite the 3 peer-reviewed articles that form the basis for this thesis. In fact, my own dissertation was written using the "staple method" taking these three peer-reviewed articles and weaving them together in about an hour. This may make @qsp's disgruntled examiner even more disgruntled, but that is the way that my field is moving...

  • "Surely you have been reading papers over the past 4 years, and have used this scholarship to enhance your own research." ...exactly! – cProg Oct 23 '16 at 15:08
  • Given that only a couple of weeks are available at most, is there really danger of wasting too much time on this? – cfr Oct 24 '16 at 3:21
6

It is a serious problem. It means that you don't know what happened in your field during the last four years. You should try to update the literature review by adding newer references, including at least some 2015 and 2016 references.

It does not matter when you started your work. Even if you started your Ph.D. in 2010 and did the literature review in 2012, you should have updated your literature review during the last few years. You need to keep the literature review up to date.

2

There are two very different things in your question, and I get confused because you mentioned it like the same thing.

  1. Your literature review is only until 2012.
  2. The latest paper you cited is in 2012.

IMHO, 1) is OK, and 2) is not. Actually I did 1). In my thesis, I had a small subsection of only 3 pages, called "state-of-the-art", which briefly summarized all the techniques and more importantly their limitations up to 2011, the time I started my PhD. This helped showing my motivations to choose the topic, and what I had achieved in the thesis (in one paragraph).

After 2011, there were papers that developed alternative approaches to mine. There were papers that were incremental to my work. My reasoning is that if I put them all in the literature review, one may ask why I propose the slow technique in Chapter 1 (published in 2012), while there are already much faster technique in the literature review (published in 2015, and based some parts on my work).

For each chapter, I have a related work section that cited all the papers up to that time. Therefore, my bibliography is complete until 2015, although literature review is only until 2011.

My thesis is not a good one, an examiner in particular commented that "thesis reads in place like a bundle of papers rather than a thesis". However, both examiners are very established in my field (also Software Engineering), and none of them complained about the literature review.

1

I'm surprised he or she did not encourage you much sooner to include more recent literature. Other than seminal work, often including research older than 3-5 years is discouraged by advisers. I don't think you can get around "modernizing" your literature review and always do what your dissertation chair says if you want to finish on time!

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.