A quick guide for those who are perhaps new to writing academic papers.
When I first started writing academic papers (as a PhD student) it would be fair to say they were not of the best quality; they are also still not perfect now! While I had a reasonable acceptance rate over time it became clear that a few comparatively simple tips can go a long way. I don’t claim to be the worlds best paper author but below you can find a few simple tips I have learned from experience; there are no doubt many others as well. I am also a reviewer for many conferences, workshops and journals so sit on the other side of the fence quite often. It should be noted that a lot depends on the type of paper you are writing, I predominantly write papers which are the result of observing human behaviour relative to some computing system. Occasionally I also write more theoretical papers. Although many of my tips may be related to these two types of paper, some of the more general ones may be useful to the wider academic writing audience.
- Write the structure of your paper out first. Accompanying this should also be the main points (results) or themes you wish to address.
- It’s tempting to try and sound lofty and to revert to complex phrases just to prove you are clever. Don’t go down this path! As you will find you will often need to describe complex terms and use equally complex wording, therefore where possible please avoid adding complex expressions unless totally required.
- Remember the reviewers (and readers) of your paper may not know every detail of the area you are talking about. You should assume some common background knowledge but you must explain all the key points that people need to know in order to understand your story.
- Generally a paper should tell one story; perhaps with a few sub-points. If you are trying to talk about too many topics at once there is a good chance no one will really understand them.
- A paper is essentially an argument which defends your findings. Remember the more points or stories you add the more points you are defending and the more evidence you have to provide. Following on from this is the point below.
- You must provide evidence to back up any points you make whether these are statements drawn from citations (so cite if you state something in the background) or drawn from your data.
- A good background section always helps to orientate the reader but do not be tempted to add in aspects which are not fully relevant to the work you will present.
- A paper is a presentation of facts that either backup your findings (or theory) or may indeed show the limitations of your work. If you are aware of limitations then it is a good idea to mention and if possible suggest ways to address them.
- Avoid writing a paper that sounds like a sales pitch, instead write in an accessible but neutral way and do not exaggerate or be tempted patronise the reader. If something is clear in the data then it is clear so do not write something like “obviously this indicates” or “as the reader will know X said this”.
- If something isn’t clear in your head it most likely will not be in anyone else’s. Try getting away from the issue for a while and/or strip the idea down to it’s component parts and see how you would then explain these parts (of a whole) to others.
- The argument should provide a clear motivation for the work plus the details of how you went about it plus your findings. Basically you need a clear start, middle and end. Just like a novel.
- A good abstract, introduction and conclusion are essential. The abstract and intro draw people in. Furthermore, people will often read the conclusions more than other parts of the paper.
- Most conferences and journals have a page limit, do not be tempted to exceed this! Also if your paper looks like it will be less than the maximum number of pages then do not be tempted to pad out the work to fit the limit. Reviewers always see through this.
- I am the worst proof reader on earth and have most likely made some mistakes in this article. You must always read your work many times, thoroughly and from start to finish! Preferably do this after having not looked at the paper for a few hours or days, otherwise you will read what you want to see. In addition to basic typos, other things to look for are: consistency of terminology, abbreviations and spelling (are you using US and British English?), formatting and using the correct citation format.
- Use pictures or diagrams instead of (or alongside) textual descriptions where possible, in particular for overviews of processes.
- Where possible use charts to represent data, however always spend time looking at the best chart/graph to represent your data. The wrong chart/graph can frequently cause confusion. Sometimes asking an colleague to look at your data and asking them to suggest a format is a good start.
- It is a good idea to get someone who is familiar with the domain but not the particular piece of work to read through the paper. This can act as a way to get it proof read but more importantly it is a very good way to see if your story is clear.
Some Specific Points
- If you are a non-native speaker of the publications language please, please, please get it proof read by a native speaker prior to submission. This is also a nightmare for reviewers who can only decide on the basis of what is written whether a paper is any good. If they cannot understand what you have written there is a much higher risk of rejection even if we try as far as possible to take into account the fact that the written language may not be the first language of the author.
- The apostrophe is an annoying part of the English language, by far the easiest to understand usage guide can be found here at the slightly fanatical site of: The Apostrophe Protection Society.
- The Economist magazine publishes a useful writing style guide which I have often loaned to others when they were in need.
- Avoid long sentences, this can be tricky if you are coming from certain other languages.
I cannot emphasise this enough, your paper is a story and it is your job to tell it in the best way!
I hope this is of some use and I will no doubt add more tips as they come to mind. Also if you have any tips of your own please feel free to add a comment.