Patrick Dunleavy is Professor of Political Science at the LSE and is Chair of the LSE Public Policy Group. He is well known for his book Authoring a PhD: How to plan, draft, write and finish a doctoral dissertation or thesis (Palgrave Macmillan, 2003).
You’ve just published a research article – why should you bother writing a blog post about it? Patrick Dunleavy argues that if you’ve devoted months to writing the paper, dealing with comments, doing rewrites and hacking through the publishing process, why would you not spend the extra couple of hours crafting an accessible blogpost? Here he breaks down in eleven easy steps how to generate a short-form version of your research article.
One of the oddest things that people in academic life regularly say to me is: ‘I’m not paid to write blogposts, only research articles. If my department or the grant-funder wants to start paying me for doing posts, then that would be a different matter’. Or alternatively, the argument goes: ‘I just don’t have the time to do blogging’. Or finally, the clinching rebuttal is: ‘Your blogpost just won’t get cited, and in today’s research environment, only citations count’.
Apparently then a lot of folk suffer from some serious misconceptions about what writing a blogpost entails:
- They think it takes days, weeks, or even months to produce that difficult bit of text — it doesn’t, it takes two or three hours at most.
- They believe that time devoted to a blogpost is time away from your main research — it’s not. Your post is done after you’ve finished and published your journal article — it is just a more readable and hopefully more popular version of that article, with key messages summarized in about 1,000 words.
- Perhaps they also think that publishing a blogpost takes the time and hassle involved in submitting to journals, trekking through box after box of obscure electronic publishing bureaucracy, and then waiting weeks or months before seeing a proof, and months more for publication. But publishing a post is not like that at all. You get your 1,000 words finished in Word or equivalent. Include a table or a chart or two, being scrupulous to present them well. Then send it to a multi-author blog with a big readership in your field and a couple of days later your text is online at the blog.
If you thought all blogs are solo blogs, you could find the previous paragraph puzzling. You might well think: ‘Don’t I have to set up my own blog and then learn WordPress to get blogging? And unless I can assemble an audience all on my own, won’t my solo blog languish unread with the tens of thousands of others online?’ Of course, you can do a solo blog (or publish stuff on your individual website, an even less fruitful use of your time). But with modern blogging packages like Medium.com (where this was originally published) you need to learn nothing at all before starting. Their ‘how to use’ instructions are just: ‘Bang out some text!’ But increasingly, blogging is a collective form of short-form digital publishing. In the multi-author LSE blogs or other outlets mentioned above, it is rather like writing for any short-form magazine. (Indeed, we may soon leave the ‘blogging’ description behind for something that better captures modern realities — perhaps ‘short-form digital publishing’).
Finally, of course a blogpost may well not be cited itself (although now reputable multi-author blogs increasingly are) — but if not, this is because its job is different. Academically a blogpost boosts citations for the core article itself. It advertises your journal article in ways that can get it far more widely read than just pushing the article out into the ether to sink or swim on its own. A post reaches other researchers in your discipline (those who are not digital hermits). And because it’s accessibly written, it travels well, goes overseas, gets re-tweeted and re-liked. It takes the ‘memes’ key to your research into a limited viral spread. It also gets read by academics outside your immediate sub-field and discipline, potentially pulling new audiences to your work.
A post on the right kind of blog, one with a big ready-made audience, also often communicates your key messages to a far wider groups beyond academia itself. Thus it helps to create external impact for your work amongst practitioners in government or business or the professions as well. An LSE colleague’s article in an academic journal has been read and downloaded less than 100 times. But his post about it came out on the LSE Impact blog (and got re-published on LSE’s British politics blog), and was retweeted multiple times. In six months the post was read 42,000 times — quite a difference, even if this is an outlier case.
In the digital world, academic authors now get multiple chances to create their own accessible messages. Without having to accommodate to the limited expertise or limiting agendas of journalists and broadcasters, researchers can explain directly and clearly. The logic here goes like this. First get the eyeballs on your text (i.e don’t [just] solo blog). Then make your case as briefly, clearly and accessibly as you can. If the work is good, it could be retweeted, read and favourited by the right people in your research community. And from that cites will flow. If you can really do communication in an accessible way, your writing may also circulate widely in other disciplines and in the external world outside universities, enhancing your reputation there.
So having explained all this to blogging sceptics, the question I ask is — ‘You’ve put eighteen months or two years of your life into doing the research in your article. You’ve devoted months more to writing the paper and sending it to journals, dealing with comments, doing rewrites and hacking through the publishing process. Why would you not spend the extra couple of hours needed now to pull out from your journal article the key bits needed for a good blogpost?’
And yet we know a huge fraction of research is still being churned out only in obscure outlets read by very few people. It’s made available only as hard-boiled, jargon-prone and inaccessible text. It’s illustrated with mounds of ‘dead on arrival’ data that no one will ever look at again. In this guise alone it must stand or fall. Little wonder that (inaccurate) folklore has it that 90% of journal articles go uncited, even by the original author. Clearly thousands of intelligent academics and researchers are still not taking the extra few hours needed to do a blog version of their work, or tweet its publication to their networks and contacts.
Getting to a blogpost — it’s easy if you try
So I hope it will help to offer a brief guide here on generating a short-form digital version of your research article.
1. Let’s assume your journal article is 8,000 words long. Your task is to get a decent version of it in 1,000 words (or somewhat less), as quickly and painlessly as possible.
2. Begin by cutting out the whole of the methodology section — it matters a lot to you, but most readers won’t care. If your methods are innovative, people will probably need to read the the original article to make detailed sense of them. So briefly sum up your methods innovation in intuitive terms within the post (and even then more towards the end)— and then link to the article, using an open access version whenever you can. On the other hand, if your methods are bog-standard, any expert reader will know them already, and any lay readers will be happy to accept the standard approach.
3. Next get rid of the long literature review at the beginning — in the blog context, no one cares about academic credentializing or point-scoring. Also cut out most of any closing discussion of how your results agree with or diverge from other people’s work. A line or two somewhere near the start, and then 2 lines of closing thoughts or pointers at the end of the post, normally suffices.
4. Next, write a narrative heading that gives your essential message substantively. Try to tell readers very clearly and simply what you found out. You need something meaningful, but in less than 140 characters — that way the blog title can also be the tweet. Don’t try just one heading — experiment with six to 10 different variants to find one that really works. If your academic diffidence, or the complexity of the content, stops you achieving a full narrative heading, at the least make sure that there are clear narrative cues in the blog title.
5. Also, try to always include in your blog post a ‘trailer’ paragraph that spells out in no more than 3 or 4 lines, why the post is interesting and gives another take on what the key message is (without repeating the title wording). Your task here is to evoke interest, and give readers a good narrative steer that attracts them into the post, and assures them that they will understand it. The format is well developed in the major LSE blogs, in the Conversation and it’s used in a very brief way at the top of Economist articles. (Not all multi-author blogs are yet geared up for this approach though, especially in America where there is a long tradition of lengthy, journalistic articles [often over 4,000 words], presented in great wodges of unheadlined text, organized only by paragraphing).
6. Now comes the part where you take the key findings and arguments out of your journal article so as to form the main (or body) text of the blog post. At root, what did your research efforts discover or conclude? What do you make of your key findings or conclusions? If you are instead making an argument, integrating ideas, or developing a theme, you still need to have a crystal clear and substantive summation of the central message. Most academic writing actually shies away from these questions, taking refuge instead in formal wording, vacuous discussions, ‘heuristic insights’, banal statements, or multiple hedged qualifications. Make your blogpost very different.
It is vital here to front-load the material, putting it into a quite different sequence from the conventional article (which is end-loaded):
- Start off in a high impact way, ideally trying to begin with something motivating for readers — either a startling fact, a paradox resolved, a key summary statistic, or a great quote. For blog readers something topical linking to a recent development is often a good start. Alternatively promising readers a change in our knowledge, or other new things, is a great motivator. Once readers are hooked in, it’s OK to have a smallamount of context here (3–4 lines) that draws out the salience of the issue. It is worth writing your high impact start carefully and trying to keep it punchy.
- Next explain early on in your body text the core of your finding or argument from the journal version. Try to rigorously avoid the normal academic tendency to do a ‘dance of the seven veils’ in which layers of low-relevance or low-interest material are progressively peeled aside to reveal a tiny kernel of new findings at the end of a long screed of text. Instead, move straight to what worked in your research or experiment or archive search, etc. and tell readers clearly what you found or concluded. In a blogpost the best bits arrive early on, not just at the end. Cut out any text from your article covering intermediate stages, or earlier models, or avenues taken that did not lead to results.
- Once that’s done, you can unpack the message a little — perhaps highlighting no more than three specific aspects — ideally the aspects with the widest interest or appeal to readers, or the greatest claim to advance our knowledge.
7. Exhibits. Wherever possible include at least one table or chart, maybe two or three — but try to avoid ever having more than four exhibits.
- Explain tables or charts properly, label them very clearly, and simplify them if they are too complex. Include a short explanatory note under each chart or table that explains what is being shown and helps readers to understand it. Make sure column or row headings in tables, and both the axis labels in charts, are crystal clear.
- Look carefully at any chart or table that you’ll include, and ask yourself ifall its components are really needed — e.g. do you need all the columns of the table, or could you cut out some intermediate ones and just show the final results column(s)? Similarly try to have simple charts, where every bar or line shown is needed because it actively helps build readers’ understanding — otherwise it gets cut out.
- The acid test for any exhibit is: What do readers really need to know?
- Any modern blog comes in full color — so always colorize your dull black and white journal artwork. You also deliver the final charts or artwork to the blog yourself, in a straightforward picture format (that is, saved as . png or .jpeg files). So what you send is what you’ll get. No difficult negotiations will be needed with outsourced journal designers in low-cost, faraway places, who don’t understand your text.
8. Level, style and tone. Don’t assume that readers know what you mean without explanation.
- If you must use specialist vocabulary (‘jargon’) — and in academic work, sometimes you must — keep it to a minimum, and explain all terms likely to be unfamiliar when you first use them.
- Be especially careful with acronyms and initials and formulae. Don’t explain once, use 20 or 50 times. Explain once, then use the full label (or refresh the explanation) every 5 or 6 times the acronym or formula is subsequently deployed. Always re-label or re-explain if you stop using an acronym or formula for 200 words or a page, but now are going to restart.
- Write shorter paragraphs than in a journal — say 150 words. But don’t write bitty text where every sentence is its own paragraph — that style may work for press releases, but ordinary readers will quickly find it disorganizing. Proper paragraphs are units of thought — they give your text a subtle sub-structure that makes it far more understandable, when done well.
- In blogposts all references are unobtrusive hyperlinks — the URL sits behind a relevant highlighted term or short phrase. Of course, digital links go just to the top of the source cited. So if you need readers to find a particular section within it, weave into your own text a quote of five or six words from the source that are distinctive to the passage involved. That way readers can go to the source and use Control+F to find the relevant passage exactly.
9. Try to end the blogpost in a decisive and interesting fashion, one that sums up and encapsulates your argument in a new and neat way, perhaps opening out to next steps or future developments. Again try for a very well-written finish, that leaves a good lasting impression with readers.
10. Below the post, give the title of your long article and a clear link to it, ideally a hyperlink to an open access, full text version.
11. Lastly, include a few (4–5) lines of ‘bio’ about yourself. Ideally this should give your organizational position, link to your Twitter, Facebook or email accounts, and perhaps briefly mention recent books (hyperlink the titles) or other key works.
Writing a blogpost is a great digital networking opportunity, and these elements all help maximize readers’ ability to find out more about you and your work. They are also greatly appreciated by readers. It must always be clear that you have authored the post, and how to digitally reach the full text of your article. Unfortunately some other multi-author or group blogs will still drastically prune or omit these elements, but at least you’ll have tried.
I thank Chris Gilson and Stuart Brown of LSE for helpful ideas incorporated here. To follow up these ideas in more detail see my book: Patrick Dunleavy, ‘Authoring a PhD’ (Palgrave, 2003) or the Kindle edition, where Chapter 5 covers ‘Writing clearly’ and Chapter 6 ‘Developing as a Writer’.
Note: This article gives the views of the authors, and not the position of the Impact of Social Science blog, nor of the London School of Economics. Please review our Comments Policy if you have any concerns on posting a comment below.
Are you currently involved with regional research, policy, and development, and want to elaborate your ideas in a different medium? The Regional Studies Association is now accepting articles for their online blog. For more information, contact the Blog Editor at RSABlog@regionalstudies.org.