Last week, I gave a talk (online, of course) about ‘Writing review articles’. It was aimed at graduate students who, as part of their training, had to identify a topic in the field of developmental biology and write a mini-review on that particular topic. However, my talk contained some general advice about writing review-type articles, as well as some general writing tips, so I thought I’d share a summary of it here.
Types of Review articles
I guess the first thing to point out is that review-type articles come in lots of different ‘flavours’. They all vary with regard to length, scope, style and overall purpose, and are given different names by different journals. But they all aim to summarise and distill research findings. This makes them very different to primary research articles, which aim to present data, although they are handled in similar way, i.e. they are submitted to a journal and peer-reviewed by 2-3 experts in the field.
What’s the purpose of a (good) Review article?
A good review article might aim to:
- summarise key research findings
- highlight ‘must-read’ articles in the field
- act as educational material
However, an excellent review article will also:
- be timely
- provide critique of studies
- highlight areas of agreement as well as controversies and debates
- point out gaps in knowledge and unanswered questions
- highlight current technologies that are helping/can help the field
- suggest directions for future research
But remember that readers are usually a mix of experts and non-experts who will be looking for very different things so a good review will cater for both of these audiences. For example, a graduate student might turn to a review article when they start in a new lab to find out more about the history of a field, or to get a summary of key findings. By contrast, an experienced post-doc or PI might want to read a review written by one of their peers to find out what the current state of thinking in a field is. Ideally, a good review should therefore aim to provide a combination of balanced summaries and critique whilst being authoritative, forward-looking and inspirational. However, note that the exact ‘flavour’ or format of the review will also dictate its purpose, e.g. a ‘Perspective’ article in Journal X might aim to summarise a handful of recent studies, whereas an ‘Essay’ in Journal Y might aim to provide a more comprehensive analysis of the last decade of research.
Where to begin?
The first step is to choose the topic you want to write on and come up with a rough idea of the scope of your article. You may already have this in mind but it’s important, before you begin writing, to really nail the exact purpose of your article. To help you do this, I‘d suggest the following:
- Identify the particular theme/topic/idea that you want to focus on. In most cases, this will be something that’s closely related to the topic you work on, e.g. you might be working on something, or reading up on a particular area, and feel that a review would be helpful. If you need inspiration (i.e. if you want to write but aren’t sure what to write about), read, speak to people, and think about talks you’ve been to. What’s exciting in your field right now? Are there papers that change the way we think about something? Have you seen/read papers that converge on a similar theme/idea?
- Check that there aren’t already reviews on this topic, i.e. something that’s been published within the past year or so. This is important; no-one wants to read a review that doesn’t offer anything new.
- Decide if there is enough recent material to include (or too much). At this point, you may need to go back to the drawing board to either expand on or refine the scope of your article. It’s also helpful to read a few reviews (mini-reviews vs longer reviews) to get a feel for how much material a review can cover.
- Identify and write down the main aim/purpose of your article. What’s the key message you want to get across? Why is this important and timely? Why would people want to read your article?
Note that lots of reviews are commissioned, i.e. the author is invited to write by a journal/editor. So, if you know you want to write a review on a particular topic and have a pretty clear idea of what your review will cover, a good place to start is by contacting a journal to see if they’d consider it. This also then means that you’ll (hopefully) be working alongside an editor from the outset to develop and refine the scope of your article. You’ll also have your target audience, article format and word limit in mind while you’re writing so can tailor the review accordingly.
Before you begin writing
Plan, plan and plan some more! Having worked with authors on review-type articles for years now, I can’t stress this enough.
- Think about the sections/sub-sections you might use. What material would you cover in each of these? What’s the message of each section? How can you link the sections?
- Think about the key concepts/words/specialist terms that you need to introduce and define. Where, when and how should you introduce these? (e.g. in Intro, in a figure, in a text box). What needs to be introduced first? What’s the best order in which to discuss these?
- Think about the display items (figures, text boxes, tables) that might be helpful. How/when should they be used? What material would they contain?
When you start writing
Once you have a plan, you can start writing. I’d suggest that you start with the Title, Abstract and Introduction – these are the first parts that the reader sees of the article so they need careful thought. By starting off with these, you’ll also have the scope/purpose of the article clear in your own mind. You can then work on the main text of the article (the ‘meaty’ bit) and the Conclusions with this scope/purpose in mind, although you’ll need to return to the Title, Abstract and Introduction for a tidy up once you’ve written the main text.
Things to think about:
- Title, Abstract and Introduction: These should be short and self-contained, and should complement each other. Each one in turn should provide more detail, aiming to draw the reader in. Remember: lots of readers will only read the title and abstract (e.g. when they search for articles in Pubmed) so these basically act as a ‘hook’ to grab their attention. They also need to be ‘discoverable’ on the Web, i.e. database friendly and containing the relevant keywords.
- Choosing a title: Choose something that is short, clear and self-explanatory; try to avoid puns/idioms and colloquial phrases or references. Try to convey the key message but also provide context.
- Abstract: The abstract should then aim to highlight the most important parts of the article. The answers to the following 5 questions provide a good starting point: What is the main topic you’re going to focus on? What do we know so far? What is new/why is this now an interesting time for this field? What are the broad implications of these newer findings? What does your review aim to do?
- Introduction: The Introduction should then expand on the Abstract and set the scene. Provide context by first introducing the topic: why is this topic interesting/significant, what do we know about it so far, how has the field progressed, what has the new progress shown? Ideally, the Introduction should end with a clear description of the article’s scope, aims and structure, i.e. a walk-through of the main topics that will be discussed and the order in which these will be covered. This just lets the reader know what they can expect from the article. If possible, introduce or re-iterate the main ‘message’ of the article.
- Conclusions: Emphasize the key message or theme of the article and, if needed, reiterate the data that support this message. Highlight the broader significance of this conclusion. Finally, if possible, bring your voice to the article: What do you think are the most compelling questions raised by these studies? What approach(es) could be taken to address these open questions? Are there technical hurdles that need to be overcome? What are the broader implications of this, i.e. why are further studies needed and what benefits might they offer?
- Display items: Use figures to emphasize or illustrate key concepts/processes, or to introduce or summarize. Remember that figures should ideally act as stand-alone items; you should be able to follow them by eye and without referring to the main text, although each figure should have a clear title and a figure legend the walks the reader through the figure. In general, schematics are easier to follow than images reproduced from primary articles. Tables can be useful for summarizing lots of information, for comparing/contrasting things, or for highlighting advantages and disadvantages. Some journals encourage the use of text boxes, which can house additional or background information or material that is peripheral to the main theme of the text.
General things to think about while you’re writing (and to re-visit before you finish off!)
- Try to group your discussion into sections/sub-sections. This just helps to break up long chunks of text (and helps to keep the reader interested). If you already have a plan (e.g. a list of headings/sub-headings) this structuring will be much easier.
- Each section should begin with a small introduction.
- Each sub-section (and/or even each paragraph) should then have a clear message/point to it, e.g. What question did particular sets/types of studies set out to address? What did these show (and here you can go into the detail)? What could be concluded from these?
- It’s also helpful to add in a few lines to wrap up each section and ease transition into the next section.
- Make sure that all statements are adequately supported by a citation. Cite the source/primary article whenever possible (but note that it is okay to cite Reviews for established concepts or to refer to a large body of evidence).
- Think about the word count and how much can be covered/how much detail you can go in to; you may find that it’s easier to write lots first then trim at a later stage.
- Avoid regurgitating the conclusions drawn in the papers you cite without giving them some thought.
- Don’t shy away from discussing findings that contradict each other. It’s better to highlight what can/cannot be reconciled and the possible cause of any discrepancies. Also use this as an opportunity to draw out the questions that remain and discuss how these questions could be addressed.
- Similarly, remain balanced – make sure you discuss the findings from the field as a whole (and not just the data from a few select labs).
- Make it clear when you are stating results versus providing speculation or alternative interpretations.
- Provide critique if you can…but keep it polite and constructive.
- Remember your audience: the article needs to accessible to expert and non-expert readers alike.
- Introduce/define/explain specialist terms, cell types, tissues, phrases on first mention.
- Consider using display items to house any material that a non-expert reader might find useful.
- Don’t assume the reader knows what you’re thinking and how things link together; you might feel like you’re sometimes stating the obvious but it’s better to do this than to leave readers feeling lost.
- Stick to using clear and simple sentences…but try to vary the pace of your writing, e.g. by using a mixture of long and short sentences.
- A general rule is to write as you would speak, using active rather than passive tense/sentence construction.
- Be thrifty with your words:
completelyeliminate any that aren’t needed.
- Avoid vague sentences. For example, say ‘Factor A causes an increase/decrease in Factor B’, rather than ‘Factor A modulates Factor B’.
Importantly, be patient and don’t get frustrated! A good writing style needs to be developed over time and comes with practice. Of all the things highlighted above (structure, content, accessibility and style), I’d say that style is the hardest to really nail. Getting a good and consistent writing style is also challenging if you have multiple authors working on the same article. In this case, I’d recommend that you nominate one author to do a final comb-through to iron out any inconsistencies, although hopefully you’ll have an editor who’ll also assist with this! On this note, I should point out that the amount of input you receive from an editor will vary from journal to journal, e.g. some journals have dedicated editors who spend a significant amount of time, working alongside the authors, to edit and improve a review.
Finally, some tips from fellow editors!
We have a bunch of experienced editors here at the Company of Biologists so I asked them all for their key pieces of advice. Here are just some of the things they suggested:
- Plan, plan, plan – make sure you have a good idea of the overall structure before you think about details
- Get feedback. Before you submit your review, send it to someone whose opinion you trust and ask them for their honest thoughts. Don’t be discouraged if they give lots of feedback – this is exactly what you want!
- A review shouldn’t just be a list of facts, e.g. X showed this, Y showed this, Z showed this. A narrative thread or argument that connects is much more engaging.
- Take time to pull back and look at the overall structure. Does it make sense? Can you see how the ideas join together and flow from beginning to end?
- Remember that readers aren’t psychic. Explain why you’ve chosen the scope you have, why you’ve chosen to discuss particular examples, why you’re moving on to the next topic. Also make sure you clearly link up relevant observations and state conclusions rather than expecting the reader to make connections.
- Don’t assume that the reader can link two statements that you might be able to link in your mind; you have to explain the link.
- Think about the graphics at an early stage – figures can often feel like a bit of an afterthought but good figures can really help to get the message across far more concisely than text.
- Break the article up into sections so that people can easily find the particular piece of information they might be looking for; recognize that not everyone is going to read from start to finish.
- Remember that your readers will know far less about the topic than you do. So before you dive into the new and exciting findings in the field, make sure you’ve given a clear overview of the system you’re writing about. Imagine that you’re writing for a new PhD student who’s never worked in this particular field.
One final point: there’s no ‘winning formula’. This is just my advice based on the articles I’ve handled and the authors I’ve dealt with, so you may find that some of it doesn’t work for you or that someone else’s advice differs. Ultimately, you should aim to develop a writing approach, technique and style that works for you.