Seven questions to ask about how your chapter ends

Final impressions are lasting — make them good ones

How many times in reading academic work have you skimmed materials from the beginnings and ends of chapters or papers — the title, abstract, opening paragraphs and conclusions — to try to work out quickly what its key messages are? For myself, I do it a lot. It’s an essential skill now for any academic (indeed any professional person), part and parcel of a wider skimming approach called ‘eye-balling’ (as opposed to reading carefully). Without it I guess most of us would be swamped by the volume of stuff to read.

So starts and ends matter hugely to the impression that get of your work. If they are poorly done, obscure or vague, it’s all too easy for readers to decide that you don’t know what you are talking about; or have no substantive findings ( or value-added or new argument) to report. Endings are even more important than beginnings in this respect, because this is where even careful and dedicated readers look for the ‘take aways’ (as a management consultant might say) - that is , the points about your work that will stick in my memory a week from now, after I’ve forgotten many of the details.

Yet often endings are quite poorly written. Many PhDers struggle with Conclusions sections, assigning them little space or focus — after all, (the implied thought seems to be) the examiners are paid to read every word anyway. Even academics writing serious books often seem unsure how to write a ‘Conclusions’ section. Some authors just pick out one particular thought or finding to end on, often from the last section covered; or sometimes even finish with a new side-point, one that’s rather tangential to what the chapter argued. Other finishing sections are overly brief, a slightly disguised version of : ‘Thus you can see that everything I said I’d do at the beginning of the chapter, I have now done’. Occasionally this is accompanied by some verbal handwaving at the academic salience of whatever value-added is being claimed.

Academic authors and PhDers are often diffident about repeating themselves, perhaps feeling: ‘I’ve said it once, and that should be enough for anyone. If you’ve read the chapter text carefully, it should be crystal clear already, and I won’t insult your intelligence by reiterating points’. There are several problems here. First, any academic author is a specialist, someone who has lived with a given problem for years, and perhaps invested much of their life in addressing it. What seems obvious to them, and perhaps even to research team colleagues or PhD supervisors who’ve also lived with the problem, may very well not be clear to others. Second, you just cannot assume that readers have ploughed through the whole chapter already. As I noted above, readers have many styles of absorbing information. They may well rationally choose to wade through the detail of your argument only after they’ve read the Conclusions, and not before. In this case, your ‘self-evident’ conclusion may just seem vague, hard to follow, or off-putting.

So, like chapter starts, endings need to be designed and written to leave a good impression. Does the closing of your chapter look clearly argued, purposeful, non-repetitive and well-paced? Or alternatively, do things repeat, end vaguely or seem to peter out inconclusively, ‘not with a bang but a whimper’? Here is a checklist of seven questions to help you assess how you have rounded off the chapter.

1. Is there a substantial and well-signposted finishing-off section, that is at least one page long, up to around 4 or 5 pages? It should start with a single-word heading, ‘Conclusions’ — and this should not be made part of the section numbering (i.e. not labelled as section 2.6). [Clear] versus Either there is only a malnourished conclusion, of one or two weedy paragraphs [Peters out] Or: A very long Conclusion (6 or more pages), perhaps with sub-sections (like ‘Discussion’) or introducing new empirical material or theory exegesis. [Unclear] If this bit is integrated into normal section numbering (e.g. 2.6) this only helps it disappear as a discrete element.

2. Do the Conclusions start by reviewing top-line findings or key argument points from the Chapter’s component sections? It should pull together key ‘bottom line’ arguments , showing how they inter-relate. [Clear] versus The Conclusions section refers only briefly or randomly to main findings in the Chapter — it does not make clear any core findings or arguments.

3. Do the later pages or paragraphs in the Conclusions section summarize and frame findings in new, synoptic ways? Does the section ‘open out’ for a while to make clear the Chapter’s role in the whole thesis or book argument? And does it link back to at one least one of the key book or thesis themes set out in Chapter 1? [Clear] versus The Conclusions just reiterates bits of text already given — no opening out or linkage back to overall themes. [Repeats only]

4. Does the overall chapter ending helps readers to stand back from the details reviewed earlier in the main body of the text, and instead to see a bigger picture? [Purposeful] versus The finishing section only gestures at analysis, or stays locked in minutiae and detail. [Vague, unclear]

5. Does the final paragraph of the Conclusions establish a link-forward to the next Chapter? [Clear] versus The chapter ends without any forward links. [Unclear]

6. Does the Conclusion section as a whole give the impression that the Chapter ends decisively, in a way that achieves closure for the Chapter topic? Does the finish seem organized and purposeful? [Clear] versus The chapter just seems to grind to a halt or run out things to say, or seems to end superficially or prematurely. [Unclear]

7. Do the Conclusions reinforce a feeling of cumulation and progression in the thesis or book argument, making clear that another component of an overall structure has been put in place? [Purposeful] versus Readers are left adrift, to fend for themselves in sifting out key findings, and perhaps are left less sure of where they are in the overall argument than they were before. [Unclear]


To put these ideas in a wider context, readers at PhD or higher level might find it helpful to read parts of my book: Patrick Dunleavy, ‘Authoring a PhD’ (Palgrave, 2003). See also useful material on the LSE’s Impact blog and on Twitter @Write4Research.