So after months of lab work, late nights, and statistics you’ve got some data and have p-hacked your way to a statistically significant correlation. Now you just have to write up your findings and force the manuscript through a peer review system based more in tradition than sanity in order to publish before you perish. Here are some tips and advice to help you smuggle your ineptitude past Reviewer 2.
ABSTRACT
The abstract will usually be the last thing you write. Get on with writing the rest of the paper, and hope that along the way you stumble on an idea for what you might actually want to talk about. If you reach the end of the paper and still haven’t got a clue, just summarize what you’ve written.
Make sure that your abstract contains as many fashionable jargon terms as possible so that your paper will come up in another researcher’s Google™ search. Add your key results so that other researchers who do not want to pay for the article can cite you anyway. The title and keywords of your paper should also be chosen to maximize Google™ hits regardless of the actual search terms used. An example abstract is given below.
Abstract — Using next-gen agile machine learning algorithms for sustainable data mining of cloud-based big-data sets, we have identified a really quite interesting trend in community health outcomes. We report on the adaptation of this innovative data-driven analysis to transformational technologies supporting future research into the socio-economic impact of quantum computer-simulated CRISPR gene editing for individualized cancer treatment. The analysis shows that the cross-platform framework improves end-to-end outcomes by a factor of 3 (N=2).
Keywords: less, talk, more, coffee.
Depending on your target journal, the abstract will be between 100 and 300 words long, and will take you approximately a full day to write.
I – INTRODUCTION
Well done. You’ve typed the section heading and are nearly ready to begin writing. You want to minimize the number of distractions to your lack-of-workflow. So now’s a good time to check your phone, email, Facebook, Twitter, Instagram, Reddit, the news, your phone again. Once no more new information is forthcoming from these burdens, there’s nothing left for it, time to start writing.
The introduction to the paper needs to place your work in context with prior and contemporary research. This means you have to cite at least one publication from any researcher who might conceivably be asked to review your paper. Failure to cite your anonymous reviewer is rarely fatal, and will normally result in “the author should reference the important contributions of My Ego et al” being listed as a compulsory revision. However, citing your reviewers early is the best way to butter them up before subjecting them to the caffeine-fueled incoherence to come. The easiest way to do this is to feature them prominently in your first paragraph: “Major recent contributions to the field include; Published Last Week (2019), Huge Knob (2017), Sum Chump (2013), Biggest Name In The Field (2011), Dr Horrible (2008), and Should Have Retired Years Ago (1995).” This also serves to pad out your reference list.
After the boot-licking, the rest of the introduction will consist of a preemptive apology for your incompetence that will shortly become apparent in the succeeding sections. This is the key skill that you will need to develop to get your PhD, the aptitude to take your B.S. to a whole new level.
Your understanding of what you have done will be a hodgepodge of your own experience, your supervisor’s knowledge, ancient wisdom passed down through generations of postdocs, and vague recollections of having read something in a paper somewhere. Inevitably, this will mean that for the statements you actually want to support with references, you will be unable to find a citable source. In this situation, phrases such as; “In our experience”, “In a series of cases”, “It is generally believed that”, and “It is well established that” can be substituted for proper references.
About this time, fatigue will start to set in. To help you keep the words flowing, set goals for yourself. Don’t check Facebook until you’ve finished a paragraph, and if that doesn’t work for you, turn it into a game. See how many Star Wars references and Dr Suess quotes you can hide in the manuscript before your supervisor notices.
II – LANGUAGE
Proper academic language is essential to the composition of a professional-sounding paper. Think back to the all of the good writing practices you were taught about in school, and forget them. Academic journals are no place for clear communication and good grammar. Jargon makes you sound professional and a variety of useful jargon-generators are available online. (For examples of incomprehensible ramblings masquerading as brilliance, I can recommend any of the works of Patrick White.)
Avoid using the active voice. Some journals will permit you to use “we”, but many have disallowed this in an effort to reduce intelligibility. The passive voice both makes your writing more boring to read and more difficult to understand. If you’re not sure whether you are using the active or passive voice, keep this in mind; if you can add the words “by zombies” to the end of the sentence and it still makes grammatical sense, the passive voice is being used.
To further reduce readability, remember that prepositions are the best words to end sentences with. Never use the word “very”, it sounds unprofessional. Use “hella” instead. In science, you need to be more or less specific.
III – FIGURES
Good figures can make complex experimental setups and results much easier for the reader to understand, so make sure to pack as much information into as few figures as possible. Label everything in the figure, especially if it’s unnecessary. If there are so many labels that you have had to use an unreadably small font size to fit everything in, then you’re about right. The legibility of the labels can be reduced further by having them cross over grid lines.
It is important to compose your figures so that they work both in colour and grayscale. Many journals still choose to sacrifice trees and, since your supervisor doesn’t want to pay the colour printing charges, your image will be printed in grayscale. But since most modern humans download the papers they are going to read, you want your image to be attractively colourful. However, older professors (and thus, likely reviewers) usually print out the article, on the department printer, which doesn’t do colour, and so your image needs to work in grayscale.
After a couple of hours navigating the colour palate you’ll find that, if you have more than two colours in your image, there is no combination that works in grayscale that doesn’t look gaudy in colour. So take your most recent iteration and stick it in.
Different journals accept different image formats and you will usually find your target journal uses the format you did not save your image in. High-ranking journals tend to want the latest and trendiest image format that all the graphic design hipsters are using and, being high-ranking, have the power to demand that you submit in that format. Well established (i.e. old) journals with moderate impact factors will tend to use older formats because their author base have resisted any push to upgrade to new formats since the 70s. Younger journals desperately trying to get their impact factor above 0.4 will let you submit just about anything (including a line drawing done in MS Paint, may God rest its soul).
IV – DISCUSSION
The discussion is where you try to tease some sort of meaning out of your results. To this end, statistical analysis is your friend. With the right choice of analysis method, your data can be made to support just about any conclusion you want. Even then, you only need your result to be within an order of magnitude of being “right”. This also makes the work quick to write up, since all you have to do is name the analysis method you chose, and then note the number that popped out of it. This will give even the most experienced reader the impression that some sort of science has happened.
The discussion is also where you apologize again for wasting the reader’s time with what you have vomited onto the last half-dozen pages. This is usually done by finding as many ways as possible to excuse the flaws in your methods, data, and analysis.
V – CONCLUSION
You’re on the home stretch now. Go back and read your introduction to remind yourself what the heck it was you were doing again. By this point, the reader will need reminding too, so start with a sentence that says what this experiment/study/fever-dream was about. The rest of the conclusion will be a few short sentences summarizing your apologies and excuses. The final sentences are your chance to admit that you don’t understand what just happened, and no significant progress has been made, but it must be done in a professional manner. A nice sentence like “Additional research is required to develop a more complete understanding of this phenomenon” serves to bring this train wreck sliding to a halt.
ACKNOWLEDGEMENT
There’s two ways this could go. Either everyone who was ever vaguely aware this research was going on will want to be listed an author, or they will want to distance themselves from the deranged ramblings you have produced. In the latter case, thank them in the acknowledgement.
REFERENCES
References will take at least a day to finish because no two publishers use exactly the same format. The minimum required number of references to give the illusion of comprehensive knowledge will vary from field-to-field, but will always be three times longer than necessary.
David Gozzard 