Finally as promised! This is a #ritaunsolicited very long thread on writing effective abstracts for methods papers. (thread 1/2)

I’ve gotten the criticism of sounding condescending in previous advice threads, so I’ve sat with that for a while now and tried really hard not to do that here. Making these threads is only worth my time if readers find it helpful!

So I will begin by explicitly stating that almost all the abstracts I read are quite good. You all have this pretty much in hand already. Many of you need no advice and indeed should be teaching others.

I was inspired to write this thread based on my interactions with some abstracts recently.

When I was inviting referees to review papers, I found myself summarizing what was cool about the paper in my own words, because I worried that the referees would not get excited by what was in the abstract.

This is not to say that the abstracts were bad – they were clear and factual-- they just didn’t necessarily convey some key exciting points of the work. To me, this is TRAGIC, because I have read the paper, and I found it exceptional, which is why I’m sending it out for review.

I know abstracts are talked about to death. I’m hoping I can give some insight that’s new into this, especially for methods papers in particular. Of course, the best thing about advice is you don’t have to take it!

I have read thousands (and thousands) of abstracts and edited hundreds. This does not make me perfect at writing them. I’ve just observed some common things associated with abstracts, good and bad. I’m sharing some of them with you now.

So here we go. Why are abstracts important? Scientists interact with a lot of papers, and the abstract can function like an elevator pitch, a short way to introduce your work and try to convince someone they ought to engage with the work further.

Abstracts should stand alone from the paper. I like to think of it as a short synthesis of key takeaway points. If anything, they’re married to the title.

Are they important for editorial assessment? This may vary, but for me not so much. I am going to read your paper no matter what’s in your abstract. It’s nice if it orients me to what I’m about to read.

Could your abstract help or hurt you in getting the best referees for you paper? Absolutely. (Circling back to my inspiration for this thread). Beyond that, they are easily searchable, so putting in key details can help lead readers to your work.

The biggest mistake I see in abstracts for methods papers is that they are too vague. It’s that simple. For example, they may describe the problem they are addressing and that they have a tool to fix it, but they don’t get to the heart of how the tool works or why it innovates.

Methods papers are special, the method is the star! The method is the result! This absolutely needs to be reflected in the abstract. The abstract should very quickly describe the problem (why do you need this method?).

From there, it should describe what the method is and the primary innovation. If the method has a new name or acronym, put it right there in the abstract. This is going to help readers know what to expect, and later to find your paper.

Then, to wrap up, authors can mention benchmarking and list the key demonstrations readers will find in the paper.

Building off the fact that the method is the star, for our abstracts, I always like to see more words devoted to the method and what it can do than the biological findings obtained using the method.

I think the point above is unique to methods papers. The abstract should focus on telling readers how the method is generally applicable and/or used on challenging samples, rather than focusing on detailed biological observations. For us, this helps capture a broad audience.

Of course, this will not be universally true. Some methods papers are half method/half-results. This isn’t the norm for us, but if it is the case that should be reflected in the abstract.

Going back to details, I think it’s important to make it as detailed as possible. Don’t use just “super-resolution” when you’re specifically doing single-molecule localization microscopy or STED. Again, this tells the reader what to expect.

I think when writing abstracts, authors should ask themselves, who is the audience for this paper? Who do I want to find it and read it, and how can I use these 4-5 sentences to make that happen? What would compel me to read a similar paper?

When I edit abstracts, I ask myself “what was it that excited me about this paper? Does that come through in the abstract?”

BUT, there is a fine line. An abstract does in part advertise your paper, but it should not be overblown or oversell what follows. It doesn’t need grandiose language or hyperbole. The best of these are straightforward and readable while being information-rich.

Another fairly common thing I see is that authors devote a large number of words to an introductory sentence in an abstract. I actually think the first line of an abstract is an art in and of itself.

For some reason, twitter won't let me make this thread longer, so to read on with this VERY long thread, please find thread 2/2!