writing a journal article – how many references?


I’ve been asked about how many references go in the literature section of a journal article. A supervisor had offered a view – one reference per sentence is best, perhaps two. But, the person asking me said, they had seen papers with lots more.

How many is too many? What is not enough? What to do?

This is both an easy and a pretty tricky question to answer. The easy answer is that there are no rules. There aren’t even any strong conventions – different disciplines have different habits when it comes to referencing. And different reviewers have different views. What’s more the actual topic of the paper matters too – some papers just need more literatures than others.

My starting point is that citations are actually little actors in your text. They do things for you – or against you. Before thinking about numbers, it helps to consider what is the work that you want your citations to do.

So here’s three things to think about. Just to start off.

(1) Where do you want to place yourself?

A journal article isn’t the place to do a comprehensive literature review. You’ve done that before you write the paper. What’s needed in a paper is enough references to ground your research in the field and make the contribution of the paper clear.

The implication of the need for brevity is clear. If you can’t write something comprehensive then you have to be economical. You have to cite just enough to signal the location of your paper – where the paper sits in relation to others – and what the paper calls on that’s already been done.

And this may mean referring to some key literatures in the field. Citing key texts and authors shows the reviewers and readers that you know the topography of the field and who’s who and what’s what.

Choice is crucial when you’ve only got a few cites to play with. If you refer to a bunch of secondary sources, or attribute a key term to someone who isn’t considered the key source, or cite a bunch of people all of whom have different views but you suggest that they are the same by grouping them together, it’s bad news. You are actually saying to the reviewers that you haven’t much idea about the field.

A corollary of the key figure citation strategy is that you will end up referencing the same old bunch of people, those who already have loads of citations. You may not want to do that.

So counter to this approach, you may want to situate yourself in relation to another category of researchers in the field. This is where the politics of citation comes in – you may choose for example to signal your own position in the field by citing people of colour or women or people from the global south. Or if you are studying a place or texts different from your own location you may choose to cite scholars from that place, drawing on local knowledges and research.

But placement isn’t really about numbers of citations. The politics of citation is about what you cite and what you might think about in choosing who and what to use.

(2) How do you want the reader to read you?

Another important question is readability. This is probably not such an issue for you if you’re using a footnote citation system. Your citations don’t get in the way of the reader. But if you are using something with parentheses, then packing a sentence and paragraph with references can make for hard reading. It can be tricky for a reader to pick their way through a very dense bit of prose, particularly if it also has long sentences, multiple clauses and lots of nominalisations.

My preferred method in choosing how many citations to use connects with the look of the prose.

  • If my text looks like it is choking and it’s hard to pick out my words among the references to other people, then that’s probably a sign that there are too many citations.
  • If the balance of citations to my text is heavily in favour of other people then I am not really managing the argument. I’m standing in one spot lifting up placards and not actually moving anywhere at all.
  • If it’s all them and not me I’ve lost my voice and become a serial lister.

Yes, you may need a bit of a list somewhere in order to make your case, but not at the expense of what YOU want to say. So the numbers questions relates to your visibility and voice in the text.

(3) How do you want to express your scholarship?

Now I know that some people think that citation loading is a sign that you are well read and a proper researcher. Bristling with brackets means knowledgeable. Stuffed to the gills means erudite.

Well, to some extent that’s true. But quantity is not the only way to show you know what you’re writing about. Both citing key figures, or offering a counter reading of the field and who gets cited in it, signal that you know what’s what in the field.

However, sometimes too few citations are an issue. I can’t tell you the number of times I read things like – there is widespread agreement in the field (one name and date) – or – researchers generally agree that (one name and date). If you are only going to cite one person in this kind of sentence, there had better be a good reason for there being only one. With these kinds of ‘many’ claims it’s probably better to have a couple or three references, and also say e.g. The moral is that if you say multitudes then you need to evidence that in some way.

And citing yourself? Well yes of course you can, especially if you are building on work you’ve done before. But too much self citation looks like you don’t know anything else, or worse still, you don’t value the work of other scholars.


So how many references to literature in a journal article? Well, it depends.

Im sorry. That’s a non answer. Or perhaps it’s an answer which says that the number of citations follows the way in which you think about the work that citations do for you.

Photo by Gaelle Marcel on Unsplash

About pat thomson

Pat Thomson is Professor of Education in the School of Education, The University of Nottingham, UK
This entry was posted in academic writing, citation, citations, journal article, literature review and tagged , , , . Bookmark the permalink.

7 Responses to writing a journal article – how many references?

  1. MH Thaung says:

    Thanks for this. Do you have thoughts on when to cite review articles vs (more) original works, or what a good balance might be? I tend to refer to review articles when setting the general context, and then choose original articles/case reports to discuss when evaluating my specific work and how it fits into the wider knowledge.


  2. Rahaman Mohammed says:

    Thank you this is really helpful as someone who is new to this.

    Liked by 1 person

  3. Claire says:

    Hi Pat, I’m about to write my first journal paper and I thought that blog was spot on. I completely understood what you were trying to across to your readers. Thank you!

    Liked by 1 person

  4. Jane S says:

    Dear Pat: As ever, timely, cogent, and relevant. Although I’m writing up a thesis, not composing a journal piece, a couple of your points struck home.
    For example, I recognise the comment about citing ‘a bunch of people all of whom have different views,’ to suggest they have something in common simply by grouping them together.’ I can see how such specious logic can be employed to legitimise or underpin an essentially weak thread or argument – which only makes it more conspicuous!

    But the politics in ‘citing key figures, or offering a counter reading of the field and who gets cited in it,’ to ‘signal that you know what’s what,’ is a minefield. (Who in academia doesn’t have their pet theories, and their pet hates?) On the plus side, I quickly learned how serial listing is bad practice – largely because it’s so boring – and also to avoid taking one authority’s opinion or theory as a universal law purely because I agree with it and it’s a neat fit.
    My ideal citations would be concise, e.g., ‘Smith (1999) p. 10; Jones (2011) p. 55,’ or something like. However, the thesis evidences otherwise.
    (Memo to self: It’s not quantity that displays whether you know what you’re talking about, m’girl. It’s quality.)

    PS: I’ve been a part-time researcher for some six years now, but find material’s been published since I began that means I need to cut / update / alter / revise. One cannot be out of date!


  5. sylvia hammond says:

    Dear Pat, Once again, I find timely and appropriate advice. I have just started a draft article – now reviewing against your guidelines. Your contribution to my journey is invaluable – & incalculable :))


  6. Kate Hegarty says:

    I recommend all students read their institution’s guidelines provided to thesis examiners. That document will show how the use of literature is framed to examiners, and the measures of achievement they are asked to attest to.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s