My thesis has around 100 references. What do you think if I highlight most important references (for example with bold text)? does it add more value to the work?

  • 3
    What field is your thesis in? Have you seen this done before in other theses? In my field, this would be very odd. – NMJD Feb 9 '17 at 16:19
  • Computer Science. Well, I also never seen it done in other thesis. – Woeitg Feb 9 '17 at 16:26

Generally, this is not something you do in your thesis, and should be solved in the text (as mentioned in the comment of NMJD).

But it's not completely strange, the entire series of Current Opinion in XXX journals have marked references:

• of special interest

•• of outstanding interest

They're also marked in the text like:


But keep in mind that only reviews are published in these journals, and those are generally not comparable with a thesis.

  • 1
    Additionally, the bibliography contains a short blurb describing why these references are flagged, like This is the first prospective study investigating whether hair cortisol and experimental cortisol stress reactivity measured prior to trauma predict development of PTSD symptoms." I always find these indicators incredibly helpful and wonder why more journals don't include them. – Matt Feb 10 '17 at 16:42

Your thesis should have an introduction. You can highlight the most important references by mentioning them in the introduction and explaining why they are important.

  • 3
    This is done not only in almost every thesis but also in almost every paper I have read. – Alexander Woo Feb 9 '17 at 16:52
  • 3
    @Woeitg I think what Alexander Woo means is something of the form: "The central challenge has been well-described by Smith et al. [cite]: there is simply not enough priority given to the blah blah blah...." -- or -- "The problem of [issue] has been deeply studied by Smith et al [cite]. To briefly summarize: [blahblahblah]" – NMJD Feb 9 '17 at 17:33
  • 1
    Good point - and if your citations use numeric coding and are sorted in order of citation, that also means they are listed first in your citations, making them the most likely candidates for further research. – Tobias Kienzler Feb 10 '17 at 8:40

You're probably thinking by analogy with the indices of some books, which boldface the pages that cover the concept in detail. So, for example, an entry might read "Widget 17, 25–28, 34, 41, 50", meaning that widgets are discussed in detail on pages 25–28 and 50, and also mentioned on some other pages. People look at the index because they've just picked up the book and they want to find information about a specific topic; they look at the index and then start reading. The boldface guides them to the most important things to read.

The references section works in exactly the opposite way. People who are already reading the book (thesis, paper, ...) consult the references to find out more information about some topic. Since they were already reading the book, you've already had the opportunity to guide them to the most important references first by writing text that does that. ("Widgets were introduced by Einstein [14] and the theory was fully worked out by Woeitg [73]; see also [12, 44, 56–58].") Further, the information that a reader seeks is often spread across the whole references section, as in the example I just gave. You'd have to check seven different entries to see which were the important ones. So I don't think that would add much information. In addition, people don't read the references section: they just refer to it, so they won't expect to find extra information hidden in there.

Having said that, boldfacing the actual citation would add information, in a concise way. I've never seen it done but "We now discuss widgets [12, 14, 44, 56–58, 73]" does seem to have exactly the benefits of boldfacing in indexes.

  • 3
    Yes, I think that last paragraph is a great suggestion. And a thesis gives you more flexibility to try out something like that than does a journal paper, in which you will need to conform to the house style. – user2390246 Feb 10 '17 at 10:51

If you are not forced to order your references e.g. alphabetically or in citation order, you can simply sort them by what you consider importance (and state so). Then it's rather clear that few-digit references are more important than many-digit ones.

  • 1
    The disadvantage is of course that you have to rank them, which may have unwanted side-effects of hurt feelings etc... – Tobias Kienzler Feb 10 '17 at 8:39
  • Anything other than alphabetical order makes it impossible to determine quickly if a particular paper is cited, so please don't do that. – David Richerby Feb 10 '17 at 10:02
  • @DavidRicherby Aside from a hopefully digital distribution where full-text search is possible (and the potential ambiguity of "first" author to sort by) it really depends on the amount of citations and the publication format. But I am not claiming sorting by some subjective rank is actually better, I merely provide an alternative solution... – Tobias Kienzler Feb 10 '17 at 10:16
  • 1
    Yeah, there are infinite discussions about who should be first author. I'm glad I work in an alphabetical field. – David Richerby Feb 10 '17 at 10:42
  • 1
    Ranking every one of the citations is going to be much harder than flagging a few as "particularly important" and unless this is very prominently mentioned, I'm not sure anyone would pick up on this scheme on their own. Therefore, it may not be the best idea. – Matt Feb 10 '17 at 16:32

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.