Journals always ask reviewers to recommend whether an article should be published as is, or whether the writer should do small or large revisions. They also ask if the article should be rejected outright.
Making a publication recommendation can feel like the hardest part of the reviewing process.
New reviewers often think they don’t have enough experience…. They don’t know enough about the journal, they don’t yet have a strong, internalised sense of what makes for a good article – after all they haven’t written a lot themselves. They think, erroneously but understandably, who am I to decide this? They forget that they have actually READ a lot. Reviewing can also produce feelings of anxiety, guilt or sadness – we know that someone, somewhere, could be bitterly disappointed by what we have recommended. These negative feelings are not necessarily rational – but they are a logical response to the game of journal publication.
Perhaps this is why newish reviewers most usually opt for what seems to be the safest and nicest option – publish but with some revisions required. Deciding to reject outright CAN feel extremely hard to do – although occasionally as a journal editor I encounter a few early career researchers who err on the side of harshness rather than generosity. I have also observed, although I haven’t taken a particularly organised look at this, that it is often the very experienced reviewer who recommends publication without any changes.
So it seems helpful here, in a blog intended to open a conversation with newer reviewers, to think about the basis for the two publication recommendations that are not so comfortable.
On what grounds might we reject an article for publication?
Aside from the obvious things – it’s a rant not a reasoned piece of argument, it’s a piece of journalism, it’s a blog, it’s been sent to the wrong journal , it’s plagiarised – here are some possible reasons for rejection:
(1) It’s straight from a thesis chapter – it’s a trawl of the literature, has far too much to say about methodology and/or theoretical resources, has no argument and no conclusion
(2) It’s bad research – the quants are wrong, the interpretation of the qual data is dodgy, you can drive a truck through the claims made
(3) There is no analysis – it’s a plodding report of a survey or a set of interviews and nothing else
(4) It’s unethical – people may be harmed if this is published, it’s sexist/racist/homophobic
(5) It’s got too many ideas in it – you can’t follow what is being said at all, there isn’t enough space devoted to each part of the argument, the various bits don’t seem to relate to one another
(6) The argument doesn’t make sense – you can’t follow what is being said at all, there isn’t enough space devoted to each part of the argument, the various bits don’t seem to relate to one another
(7) It’s not significant – there is no answer to the So What question. That is, it’s too local, it’s too small in scope to say anything… it’s naracisstic and self-indulgent, and/or the conclusion is what we already know and there are heaps of other articles which say the same thing and/or it doesn’t seem to say anything much at all.
We ought not to reject something because it’s written in a style we don’t like or it uses big words or we disagree with its party politics. We can raise all of these objections as reasoned arguments in a response which might, in the case of party politics, require revision to recognise different points of view.
We also ought not to reject something just because it’s boring. Again, that’s for revision, unless the reasons for it being boring are any of (1) – (7) above.
On what grounds might we recommend publication without changes?
Setting aside envy we can see that:
(1) The paper is very well written, and it’s well structured. It’s a good read. It’s elegant. It might even be pleasurable!!
(2) You can follow the argument, and the way in which it’s been constructed and on what basis. The claims that are made stack up with the data and the analysis.
(3) It says something significant, it offers important new knowledge, it offers a new way to think about/talk about/investigate something, it offers a healthy challenge to the field. You’ve been trying to sort this out and they’ve done it.
(4) Anything you can think of to improve it isn’t really necessary, and it would just be tinkering for the sake of it with something that’s already pretty darn good. (You wish you’d written it/on a good day you could do half as well/you want to give it to others.)
The thing to do when making a recommendation is to try to focus on the reasons for the decision, not the feelings. After all, it’s the reasoning you have to use to write the feedback.
And for the most common category – revising and resubmitting? I will deal with this in the third and final blog about reviewing; it will be focused on giving feedback to the author and specifying what they need to do, whether this is a few adjustments here and there, or a pretty substantive rewrite.
Pingback: refereeing a journal article. part 3. writing the feedback | patter
Reblogged this on Research Staff Blog and commented:
Here’s the second of Pat Thomson’s blogs about refereeing journal articles.
Pingback: Refereeing a journal article | Adam Loy
Pingback: First Peer Review – And Pizza | Helen Kara
Pingback: reviewing a journal article – are you Jekyll or Hyde? | patter
Pingback: peer reviewing your first paper | patter