55

"Program MARK: A gentle introduction" by Evan Cooch and Gary White

is a guide to using a particular piece of statistical software. It is available in print form as well as in a free, online version.

The foreword to the book contains the following paragraph:

We’re occasionally asked how to properly cite this book. Easy answer – please don’t. This book is not a ‘technical reference’, but a ‘software manual’. The various ‘technical’ bits in the book (i.e., suggestions on how to approach some sorts of analysis, guides to interpreting results...) are drawn from the primary literature, which should be cited in all cases.

Now, I don't have any problem with the last sentence - the authors are quite right to recommend citing the primary literature. But I don't understand why they recommend NOT citing their own work. If I have found it useful (I have) then the usual arguments for citation seem to apply:

  • Citing the guide may be helpful in pointing others towards this useful resource
  • The guide has performed an important function in my work, even if that function is "just" pointing me towards the primary literature, and that function should be acknowledged. It is in a small way dishonest if I imply that I researched the primary literature without any help.
  • While any mistakes in my use of the software are of course my own, knowing the source of advice that I've used might help others to spot or trace those mistakes.

Against this is the fact that the authors have politely asked me NOT to cite the work. I have no wish to be disrespectful to the authors who have created a valuable resource. But is it reasonable for them to ask me to break from academic good practice?

Should I respect the authors' wishes and omit a citation?

  • 33
    You can accomplish your purpose and respect the authors' wishes with a bibliographic entry or footnote rather than a formal citation. – Bob Brown Oct 17 '17 at 11:01
  • 3
    @BobBrown That's an interesting suggestion, thanks! Would you consider turning it into an answer, and expanding a bit on how that would look? Particularly in a journal paper where they are normally quite strict on citation formats. – user2390246 Oct 17 '17 at 11:11
  • 12
    For a journal you could also add that you learned many things from this book in the acknowledgments. This way you can work around any problems with footnote and citation formats. – skymningen Oct 17 '17 at 11:59
  • 11
    I think I would make a joke out of it and say something like "on the subject of MARK, we would have referred the reader to [Cooch & White], but following the authors' request in op. cit. not to be cited, we will refer to [some other reference] instead". – Gro-Tsen Oct 17 '17 at 14:47
  • 9
    @BobBrown Huh? A bibliographic entry is a citation. – JeffE Oct 17 '17 at 15:00
63

No.

If you use a source, you must cite it. The authors deserve credit for their work, even if they don't believe they deserve credit for their work.

Of course, you should also take the authors' advice and also cite the appropriate primary sources (after reading them, obviously).

  • 12
    I don't think it'd necessarily be an also, unless other fields work differently from mine. If author A references an idea from author B but doesn't bring anything new to it, we would just cite author B. If author A quotes/cites B, but we have no access to the original work by B, we cite the latter by way of the former (B qtd. in A). If we're only interested in what A says that's new, we sometimes are okay with only mentioning B's name (mainly for famous authors/general knowledge), or in many cases, not mentioning B at all. – guifa Oct 17 '17 at 16:20
  • 36
    Also, citation is not just about "who gets the credit" - in fact, one could argue it's not even the most important - what is important is being open and honest about your sources so others can investigate those sources to ensure you represent them accurately and so they can explore the topic more. This notion about credit (and the corresponding respect and funding) is meta to the notion of performing good science. – corsiKa Oct 17 '17 at 17:09
  • 4
    @guifa There are lots of good reasons to cite secondary sources. Suppose author A references author B without bringing any new technical content, but with a much clearer exposition. Or suppose author A references authors B, C, D, E, F, G, H, X, Y, and Z without adding any new technical content, but had the foresight to survey those other authors in a single document. – JeffE Oct 17 '17 at 18:12
  • 13
    Your “no” verdict is correct but I disagree with your reasoning. If an author expressly wishes not to receive credit, it is not meaningful to say they still “deserve” credit. For example, if someone I knew published a book under a pseudonym and kept their identity hidden, stating a desire not to become known as the book’s author, but I accidentally found out that they were the author of the book, would I have a moral obligation to publicly expose them because they “deserve credit for their work, even if they they don’t believe they deserve credit for their work”? Of course not. ... – Dan Romik Oct 17 '17 at 23:23
  • 9
    ... The reason why “no” is correct has nothing to do with the authors “deserving” credit. By not citing their book, it’s not that OP would be doing the authors a disservice by denying them credit that they “deserve”, but rather that OP would be doing OP’s readers a disservice, by depriving them of the knowledge of a potentially useful resource, and of the ability to evaluate the novelty and context for OP’s own contribution. – Dan Romik Oct 17 '17 at 23:26
22

Just cite them because (in my opinion) they only do not want to have any serious, academic citations.

They made these remarks to prevent you from citing them for ideas that are not originally theirs and facts that might not be as sourced as in a "proper" academic publication.

I think what they don't want you to do is cite their book like you'd cite other sources for ideas and facts. Like:

We got the idea from [Evan et al] or used the formula from [Evan et al] to prove what was proven before [Evan et al].

What I think is perfectly appropriate is something like the following:

We used fact a and fact b to get started [other guy et al], based on the idea from [this girl et al]. A helpful source in the implementation using the software xxx [proper software citation] was the textbook by [Evan Cooch and Gary White].

And then take care that you actually cite primary sources where needed.

  • 2
    While I think this is a reasonable approach to the issue, I'm not convinced that this was the authors' intention, given that they refuse to answer the question of how to properly cite the book. – user2390246 Oct 17 '17 at 16:21
  • Technically they don't refuse to answer the question, they just write it like that in their message (in the book / on the website?). If you'd like to be really sure, send them an email to ask. If you show you're just including the citation for educational purposes they might easily change their minds. – VonBeche Oct 18 '17 at 8:39
  • This is the correct answer. The authors of the manual state about being cited: "Easy answer – please don’t." But that is only the easy answer for those who do not understand why: they can stop reading there and take the advice. According to the question OP understands perfectly that his case is not one of the "easy" cases, and citing the software manual together with the original papers is the best solution. – louic Oct 18 '17 at 12:03
  • @louic: I didn't read it like that at first, but it makes perfect sense. – VonBeche Oct 18 '17 at 15:26
12

I think you may have misread what they want. I admit it is a matter of interpretation, but to me it seems that they are asking you to not use them as a citation for certain types of information, namely statistical techniques.

So, if you are doing a t-test and want to say "I did a t-test as described in (Cooch and White)", they are asking you to not do that, but to go find a textbook on statistics and say "I did a t-test as described in (Bamberg and Baur)". This should not exclude citations where their book is the authorative source. For example, if you are writing a comparative article on statistical packages, you can write "The MARK software supports both paired and unpaired t-test (Cooch and White)". This would be a proper use of their work, for which they deserve a citation. They wrote a book on what a given program supports, and you can cite them for that.

So why did they express it in this unusual way? I think they had a specific target population in mind. There are a lot of students out there who are eager to do right, take the time to perfect every single citation they use, but don't know yet enough about different types of literature sources and which one is appropriate for which case. So, they do a statistical analysis by clicking the appropriate button in their software tool, their supervisor remarks that they need a citation, and the first place they naively look is the book describing which button to click. And then ask the authors how to make the citation "right". If this happened often enough to Cooch and White, I can understand why they chose to put this information in the book, hoping to save the students from doing the wrong thing.

As for everybody else, who does not belong to that population but is using the book as a citation for the things it actually is meant for (e.g. information on the functionality of MARK), I think they are free and even obliged to cite it in the usual way.

  • 5
    "you have misread" - Assuming that your interpretation holds, I wouldn't put the blame on the OP, but on the authors for not expressing clearly what they want. They straightforwardly say that one should not cite the book. – lighthouse keeper Oct 18 '17 at 10:09
4

Whenever you make a statement that expresses someone else's idea, you should provide a citation for the idea, even if the originator of the idea ask you not.

Citing the guide may be helpful in pointing others towards this useful resource

This is not a reason for citing work. The purpose of citations are to give credit to others for their original ideas and not because you think something might be useful to the reader.

The guide has performed an important function in my work, even if that function is "just" pointing me towards the primary literature, and that function should be acknowledged. It is in a small way dishonest if I imply that I researched the primary literature without any help.

No one expects you to research the primary literature without any help. No one cites Google Scholar and Pubmed. I often read an article and find other articles from there and in the end do not cite the entire chain that got me to the idea I needed. Again, this is not a reason to cite something.

While any mistakes in my use of the software are of course my own, knowing the source of advice that I've used might help others to spot or trace those mistakes.

This is a reason to cite something. Saying what software you used is critical. For example, and I think it is relevant and correct, the SVD is well defined mathematically, but can be implemented in a number of different ways that behave very differently for nearly singular matrices (or something like that). It becomes important to provide readers with which version you used. Two common ways are to state the software used and to reference the technical manual. The author's have asked you to not reference the technical manual. In this case, I would then cite the software itself.

  • 3
    "helpful in pointing others towards this useful resource (...) This is not a reason for citing work." - strongly disagree. It is the primary reason. Authors receiving credit is at best a beneficial side-effect. The purpose of publications is to disseminate information, both by directly providing it, and by pointing to related publications that disseminate more (background or otherwise) information. It is not "look at how great we are, and how great these others are", accompanied by some pages of text. But basically, that's also the conclusion from your last paragraph ("This is a ..."). – O. R. Mapper Oct 18 '17 at 11:07
  • @O.R.Mapper Nope. StrongBad is right on the money here. If all you got from someone was literally and explicitly the resources that they quoted, you don't cite them. If you use any of the book's argument (&, in practice, it seems you should, if you're quoting several of their resources), then you cite them. I'd tell the OP to try harder to find some argument to cite in addition to the primary refs. Otherwise, nope. O.R.'s take is exactly like saying you have to cite the Library of Congress Classification System when you pull every well-bound book off the shelf for [your subject here]. – ruffin Oct 19 '17 at 1:30
  • @ruffin: "literally and explicitly the resources that they quoted" - unless we're talking about a pure listing of bibliographical entries, any text that cites other works typically adds something on top of that. This "something" can be a different way of explaining the topic, or even just establishing a connection between several different publications, or juxtaposing them in some way. Depending on the situation, it can be exactly this "something" that you want to point out by your citation - even if it is not a "hard finding" in the traditional sense. ... – O. R. Mapper Oct 19 '17 at 6:58
  • ... With this said, "common knowledge" usually remains uncited. Generally known sources like the "Library of Congress Classification System" thus would not be cited (unless, of course, the author wants to make a statement about something specific related to this classification system). – O. R. Mapper Oct 19 '17 at 7:05
  • @O.R.Mapper That's right -- it's near impossible not to pull something from the source that presents the sources, but OP needs to cite that thought, not just sources that source cites. Even if just to say in a footnote, "That this confluence of sources is most easily found in a source that tries to forcefully proscribe being approached academically... (PatOriginalSource 5)" Though you're now approaching it as a cultural artifact rather than an academic source, which is perfect. Thought experiment: If I pull a source from a Wikipedia article that turns useful, do I cite Wikipedia? – ruffin Oct 19 '17 at 13:49

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.