Friday 5 April 2013

A short rant about numbered journal references

Well, I had not planned a blogpost this morning, but I am goaded to do so by growing frustration as I attempted to read a review that was published in Trends in Neurosciences. The authors were proposing a novel theory in a controversial area, and came out with definitive statements, such as “Imaging studies have revealed the involvement of brain regions including the basal ganglia and cerebellum”. Quite appropriately, they supported such assertions with references. But the journal format meant that these references were identified only by numbers, so I had to flip to the reference list at the back of the paper to find out what was being referred to. The paper had over 70 references, and in the course of reading it, I reckon I did this about 20 times. I would have done it more, but it was distracting and I had to put down my mug of tea each time while I hunted for the relevant page. If I forgot the number on the way (a not unusual occurrence at my age), I had to go through this loop again.

Needless to say, TINS is not alone in adopting this referencing format: it is especially common in journals that have strict length limits on papers. I recently had a paper accepted for such a journal, and I found that any mention of the authors of a referenced work was frowned upon, and I was required to rewrite so that the only identifier of a reference was the number.This often meant turning a short active sentence (Jones [54] showed that.X..) into a longer passive one ( X was found by another study ... [54]).

Why does this matter? Well, if the reader knows the topic pretty well, the referencing will indicate whether the author knows the relevant literature and whether the evidence that is cited is balanced or involves cherrypicking. References also give clues as to where to allocate one's attention. If I see a set of statements associated with a list of familiar names,  I can read more casually because I already know the subject matter; if there is reference to novel material, I’ll focus in more depth. Maybe this is just an obsessive tendency in me, but I want to know what is being referred to, and I find that numbered referencing impairs the readability of an article. Journals may save a bit of space by using numbers rather than names, but as more and more articles move to on-line only, this is not a major consideration, and what is gained in space is at the expense of the reader experience. Hrrmph!

P.S. 13.52 5/4/13
Thanks to Neil Martin (@ThatNeilMartin) for drawing my attention to this v. relevant article. Might explain why psychology journals, whose editors understand how cognition works, tend to prefer the non-numbered system.
Clauss, M., Müller, D., & Codron, D. (2013). Source References and the Scientist's Mind-Map: Harvard vs. Vancouver Style Journal of Scholarly Publishing, 44 (3), 274-282 DOI: 10.3138/jsp.44.3.005
Copy of the article available from the 1st author: mclauss@vetclinics.uzh.ch 

18 comments:

  1. This comment has been removed by the author.

    ReplyDelete
  2. You want to play at the big school then you need to follow the grown up rules. One of the greatest problems in autism is people; without scientific research or similar high-level long term training, often read and misreport information. This is then taken as fact by other parents - and causes huge problems. Using specific page numbers helps limit this as the references must be accurate. Also the snake oil salespeople who are out there in the autistic world in large numbers cannot easily claim 'whatever' from a paper and must identify specifically where information was obtained. Finally; sometimes only sections of journal articles are copied (you and I know that is a bit on the naughty side) and these documents often take out the page numbers and enter page numbers suited to that document. This creates further confusion. You like to write and obviously get papers considered so why not extend beyond the academic world and try your hand at been a journalist-no referencing; but you do have to prove the validity of everything you report so it probably is just as annoying.

    ReplyDelete
  3. GK Moorebanks, I have absolutely no idea what you're on about. The article isn't about (mis-)interpretation of papers, or about "page numbers", it's about the use of numbers (eg. "[5-6]") to label specific references in a bibliography. Your remarks about "playing at the big school" are simply baffling.

    ReplyDelete
  4. I'm not a fan of numerical referencing either. I find it incredibly distracting and I frequently lose my train of thought. I end up having to keep rereading an re-evaluating the information.

    ReplyDelete
  5. There must be something in the water, or maybe it's the weather, but rants anout numbered citations seem to be the order of the day. Mike Taylor complains about them here http://svpow.com/2013/04/02/a-pox-on-your-numbered-references-redux/

    The advantage of numbered references of course is that they make papers shorter, a hang over from the days of print (and for journals that still have a paper version). To get around this look at this paper http://www.nature.com/nature/journal/v496/n7443/full/nature12031.html, when you hover your mouse cursor above a citation a little pop up shows you the reference.

    For the PDF version click on PDF then select ReadCube, the references are shown down the side (http://www.readcube.com/articles/10.1038%2Fnature12031). You can also install Readcube on Windows/Mac to manage your references (no Linux version AFAIK).

    Unfortunately none of this helps if you're using a print out...

    ReplyDelete
    Replies
    1. Thanks Nico. Getting a source by hovering over link is better than nothing, but it's all added cognitive effort. And I do read from printouts. Had a look at Mike Taylor's blog, but it's more on problems with numbered system if you want to cite specific part of article. Mine is a much broader objection to loss of key information in the main text. Journals that don't value identity of referenced authors tend to be same ones who think Methods are unimportant - see my post of 10/3/13

      Delete
  6. I heard you can get grants to spend on formatting (numbered) references.... http://collectivelyunconscious.wordpress.com/2013/02/27/scientists-receive-12-6-million-dollar-grant-to-format-references-correctly/

    ReplyDelete
  7. Jona Sassenhagen5 April 2013 at 11:31

    "This often meant turning a short active sentence (Jones [54] showed that...) into a longer passive one (Another study [54] found that...)."
    That's not a passive, it's simply an active sentence with an inanimate subject. Passive would have been "... was found by another study". You better fix this before George Pullum finds it and you end up at languagelog :)
    http://www.lel.ed.ac.uk/~gpullum/grammar/passives.html

    ReplyDelete
    Replies
    1. Ooh cripes! Dangers of dashing off a blogpost in early morning. Wondered whether to just leave it as an example of human frailty, but would not want to confuse, so it is now fixed.
      Thanks for the comment - and I have actually done research on passives in my time :-<

      Delete
    2. Heh! Wouldn't want to upset HIM. I know just how mean he can be - see my blogpost of 3rd Sept 2012.

      Delete
  8. I definitely agree that it is frustrating to read numbered references - especially on PDF, where you have to keep scrolling up and down and end up losing your place all the time!
    On the other hand though, I resent it when journals require you to specify the first couple of authors for every paper within the text and then also require that you include all of those words in the word count...

    ReplyDelete
  9. Paper journals still? Linked references with URL preview seems superior to me. A tab or split window with the reference listing would help, or a technically focused PDF reader as mentioned above. Go Open Access.

    ReplyDelete
    Replies
    1. Thanks for this, Paul. I'm a great fan of Open Access, and I use the split screen, and PDF preview if available, but it is still a poor compromise as far as I am concerned - I think the Clauss et al analysis (ref in PS to the post) is fascinating and makes a lot of sense to how I approach papers. Also, I'm often on a small screen (eg Ipad) and splitting is not v feasible. And sometimes I like to read paper mss, as I spend way too much of my life staring at screens. So I will persist with my campaign to abolish Vancouver.

      Delete
  10. This comment has been removed by the author.

    ReplyDelete
  11. You should try reading law papers (43)!


    43. supra n.26

    ReplyDelete
  12. I'm a bit late to the party--actually I read the posting back in April and strongly agree with the (author, year) but I happened across an interesting article that pointed out a really bad (good if you're looking for tenure?) point about Vancouver style referencing.

    It's a lot easier to, somewhat dubiously, self-cite thus boost your H Index without it being obvious.

    With APA style it would have blindingly obvious something funny was going on.


    http://academia.edu/934257/How_to_increase_your_papers_citations_and_h_index_in_5_simple_steps

    ReplyDelete
  13. I am glad that you wrote this blog - I hate numbered references However, while it is (relatively) straightforward to hop backwards and forwards on a paper version (I sometimes attach a sticky to the reference page or even photocopy it to make this easier), it is really irritating with a pdf with hyperlinked numbers that jump you down to the reference but not the reverse. Numbered references break up the comfortable continuity of reading a paper into an interrupted task which is much more like work.
    Tom Berney

    ReplyDelete