Last month, we offered suggestions on how to prepare for your thesis defence: Decide whether you need more research results, sketch out a plan for those experiments and for writing thesis chapters, and--importantly--get your supervisor's support for that plan. Now it's time to wrap things up in the lab and start writing.
Writing a thesis is easier said than done, of course, and you have plenty of work ahead. But like any big undertaking, writing a thesis is easier if you break it down into smaller steps.
First things first
If you haven't already made a countdown plan as described in last month's column, start with that first. Then, before you start writing, make sure you and your supervisor agree on the table of contents. This might seem obvious, but we have seen too many students start working on chapters only to have those chapters tossed out later.
Cut the problem down to size
Once you've decided on a table of contents, it's time to expand it into a detailed outline. Your outline will be several pages long and consist of chapter headings, subheadings, figure and table titles, some key words, and essential comments. Your outline will keep you on track and provide you with a framework for the text. It also forces you to break up the writing into manageable pieces.
Determine the format
Your department or university may have a standard format for your thesis. If so, there's probably a standard template you should use. If not, save yourself frustration and time by copying the format from a thesis that appeals to you. Make sure the format or template is easy to use. Once you've sent your thesis to your committee for review, you may consider upgrading your layout. For now, factor the format into your plan, but don't make it your primary concern.
While we're on the subject of format, be sure to use the proper citation format for your list of references. This list can run into the hundreds, so use the approved format for citing literature from the very beginning--both in the text and for the list of the references at the end. Use a good citation-manager program and enter all the information for every article referenced--including titles. You won't want to have to go back and redo this if you've done it wrong!
Transform published articles into thesis chapters
Before you delve into the chapters you have to write from scratch, start by transforming your published articles and submitted manuscripts into thesis chapters. It's not just a matter of stapling your papers together and sticking them into your thesis, however. You'll need to break the publications into pieces and weave them into a cohesive narrative, making sure the various parts fit together nicely without redundancies or gaps in logic. When doing this, keep the following in mind:
Drastically cut back or rewrite the introduction section of each article. There is no need to repeat what you will have already explained in the general introduction and literature survey of your thesis. Don't just delete those introductions, however; parts of your manuscript intros will be useful for your thesis introduction, so paste any relevant text into the intro section of your thesis outline for later editing.
Cut the Materials and Methods section as necessary to avoid repetition with other chapters. Again, you'll probably want to paste some of the Materials and Methods text into the relevant sections in your thesis.
Include text that may have been cut from the final version of the article due to space restrictions.
Update your literature citations (see above).
If someone else wrote one of your publications (i.e., you did the experiments but a more senior person wrote the manuscript), we suggest you rewrite the bulk of the text in your own words. Even if experiments were done in collaboration, a thesis has only one author--you--and the words in it should be yours.
After you've transformed your published articles into chapters, you will have to write new material for the remaining chapters. When you first start writing, it helps to begin with an easy section. This will give you confidence and get you into the writing habit. Because the methodology chapter is relatively straightforward, you might want to start with that one. If you've already written several methodology sections for your peer-reviewed articles, it won't take much time to prepare a first draft for your thesis.
Because a thesis has fewer space restrictions, you should take the opportunity to describe the details of your work that did not make it into published articles. In a thesis, it is better to err on the side of being too detailed than to risk leaving out crucial information. Be generous to the next generation of researchers; a detailed description of your progress and failures will save them a lot of time.
Writing up that last set of experiments
Now that you have worked your way through the initial chapters and have written most of your thesis, it is time to tackle writing up your final project. You probably haven't written an article on this research yet, so you'll need to decide whether to write the article first and then transform it into a chapter or do it the other way around.
If there is stiff competition in your field, your supervisor will probably insist that you write the article first. Otherwise, we suggest that you write the chapter first, as this approach will allow you to describe your work in detail. While the thesis is out for review with your dissertation committee, you can select the appropriate parts from the chapter and transform it into an article to submit to a peer-reviewed journal.
10 Tips for a Stress-Free Thesis
The introduction: The final hurdle
Although it comes first, the introduction will probably be the last chapter you write. The introduction is where you need to place your work in a broader context, explaining why the research is relevant to the scientific community and (assuming it is) to society.
Start thinking about your introduction long before you start writing your thesis. During your final year--or even earlier--create a file in which you collect ideas and article clippings that could be useful for the introduction. A file of good ideas will be a big help in writing a comprehensive and elegant introduction when the pressure is on.
The summary is the one section of your thesis that is sure to be read widely. In a few pages you will have to describe the main findings of your thesis research, so it is best to write this part after you have finished all the other chapters. Do not try to describe all your results in the summary--you're simply summarizing the bulk of your work. Be sure to designate in the summary which chapters contain particular findings.
Safeguard your work
We shouldn't have to remind you to back up your work, but we will anyway. Keep a copy of your thesis on an external hard drive, memory stick, or some other storage device. Back up daily and keep the copy (or copies) in a safe place. For extra security, keep a copy of your work-in-progress off-site on a remote server (in the event of fire or theft). The simplest way to do this is to open a Web-based e-mail account and regularly e-mail your work to yourself. There are also companies that offer online document-storage services.
Going for gold: Writing an error-free thesis
Because a thesis is usually written under severe time constraints, it is difficult to produce one without some typos and other minor errors. Spell checkers help, but they can't catch errors in those hard-to-spell technical terms. In addition, errors of grammar and syntax are not always highlighted, and minor scientific errors can be easily overlooked. Your goal, of course, is to have as few errors as possible.
We suggest you do two things to help make this a reality. First, put the manuscript aside for a short while after you've written the first draft. Once you've gained some distance from the material, read it over again with a sharp eye--not for content, but as a proofreader looking for typographical errors. Second, give a copy of your thesis to one or two trusted peers to read. Devise a creative way to reward them for every error they find (free cups of coffee or beer, or pizza, for example). This will give them an incentive to go through your thesis with a fine-toothed comb. If you can afford it, you may even consider hiring a professional copy editor to do this for you.
Most importantly, while writing your thesis, be sure to take care of yourself. Eat well, exercise, and get plenty of sleep so you're at your best when you sit down every day to write. This is the home stretch of your Ph.D., and you want to make sure you cross the finish line energized and ready for the next step.
Patricia Gosling and Bart Noordam are the authors of Mastering Your Ph.D.: Survival and Success in the Doctoral Years and Beyond (Springer, 2006). Gosling is a senior medical writer at Novartis Vaccines and Diagnostics in Germany and freelance science writer. Noordam is a professor of physics at the University of Amsterdam, the Netherlands, and director of a regional audit organization. He has also worked for McKinsey and Co.
Images. Top: Paul Worthington. Middle: courtesy, Springer.
Comments, suggestions? Please send your feedback to our editor.
Patricia Gosling is a coauthor of Mastering Your Ph.D.: Survival and Success in the Doctoral Years and Beyond (Springer, 2006). Gosling is a senior medical writer at CMPMedica in Malaysia and a freelance science writer.
Bart Noordam is a coauthor of Mastering Your Ph.D.: Survival and Success in the Doctoral Years and Beyond (Springer, 2006). He is a professor of physics at the University of Amsterdam, the Netherlands, and director of a regional audit organization. He has also worked for McKinsey & Co.
This is the second guest post from Prof Peter Downton from the School of Architecture and Design at RMIT University, reflecting on the changes that have occurred during his long research career. In this post Peter talks about practices of writing prior to computers and the effects new technology was introduced.
Although intellectually you may be aware of the fact, I can personally attest to the fact that computers are not necessary if you want to write.
I was one of the first kids on the block with a Mac when they were launched in 1984. Prior to the arrival of desktop computers I had managed to hand write several hundred thousand words: some research reports for local and international agencies; a research masters degree (two actually, for, to my supervisor’s horror, I threw the pile of pages of the first one in the bin and started again – by then I had worked with some of the people whose research I covered and my first effort seemed naive); my parts of a couple of co-authored books; some conference papers; a PhD; and a co-authored film script.
Nearly all of this material was typed by someone else for payment. Reflecting on changes in my writing processes, and those I have seen from supervising higher degree candidates since 1974 and conducting a research seminar since 1988, might reveal some things I hope are useful.
My handwriting was done on lined paper. I wrote on every second line to allow inserts and corrections, but if I really screwed up the same fate awaited the paper. It was portable: I wrote in cafes, on couches, and in (parked) cars. My entirely part time PhD was squeezed into the corners of an over-busy life, and frequently done at night or in lunchtimes. Combined with handwriting, this meant that the final draft was composed over twenty-two months. My thinking evolved over this period and the result is patchy. At least I could work in various places.
A similar level of portability was not recovered with computers until battery technology allowed lighter laptops. When reference material was required, the hand writer was more desk-boundthan current computer users. Useful material such as outlines, notes to self and diagrammed arguments were kept in arch files along with the emerging text. This was bulky to cart about compared to the one-kilo laptop I currently use.
Writing by hand on paper promotes starting at the beginning. Perhaps this is the beginning of a section or chapter, rather than a whole document, but unless a great deal of cutting, re-ordering and reassembly is undertaken, writing in the same order as it is to be read makes sense. Computers irrevocably altered this for me. Now, I simply start. Often I conclude I commenced in the middle and a beginning is assembled prior to the initial start.
Always the text is provisional at both the level of words and with respect to the structure. Therefore I fiddle. Sometimes improvement results; sometimes it is simply change. The discipline of paper limited the fiddle factor. One person I supervised, only a few years ago, had the entrenched idea that a writer must start at the beginning and work through to the conclusion. He was unable to write at all because he had not decided where to begin. In an act of remarkable desperation he began and produced thirty-five thousand words in eleven weeks as a part time candidate. After mild polishing, this was final draft material.
I have supervised people who bang out drafts at remarkable speed, write a new second take and even a third draft prior to polishing it as a final text. I have had candidates who write in small units, possibly paragraphs, and slowly manipulate them into place in the PhD. Other candidates use whatever new mind mapping, idea assembling, or writing aid they have most recently discovered.
I know those who adore these tools. I see a lot of time go into mastering new ones. I am not necessarily convinced by the outcomes.What matters is the quality of the final product. What works for the individual seems personality-driven, so I am not prepared to support any one approach. Fooling with the writing toy, like tidying up, seems often to be a displacement activity – one more way of avoiding writing.
Post computers I have written books, chapters, and papers. My current mode of working is at odds with the paper-based writing. It is largely toy-free, reliant only on MS Word, although I had some flirtations with voice recognition a decade ago. (The inability to play music or work in cafes I regard as major drawbacks – however I have new voice recognition software on order.) Usually I write a paragraph or two that have popped into the brain. Often I produce an outline.
Nearly all these first thoughts are appallingly naïve and have to be wrestled into something presentable. Everything is in the one document: attempts at structure, outlines of arguments, passages of text, reminders and thoughts in red. Slowly it coalesces. The sensible bits constantly become longer; my first efforts at a whole always need further fleshing out. Re-reading alerts me to what I forgot or failed to convey well. While I feel writing a loose draft and editing it down is a waste of energy, I realise I may well type just as many words working in the opposite direction.
For much of my computer-based writing life I printed what I had produced and scribbled upon that; commenting, re-structuring and editing. This could be done in cafes or on trams. More recently, I take a small laptop to my local café. I also write at a desktop with two 24-inch screens. Pumping up the text size helps proofing. Other documents in electronic form can be piled up for display on the other screen. It parallels having books and papers scattered on my desk.
Many of the guides to writing theses seem to assume that writing is a hated difficulty. This may be true for those who resort to such guides. I have supervised many who enjoy, maybe even love, writing, or who at least (as I do) find the process of word production fairly compelling and hard to avoid.
Editors have long liked me as I have delivered the requisite words, with a low level of error, on or before the agreed date. In my view, this has had nothing at all to do with the mode or means I have employed to write. Mostly it derives from starting early enough. I learned early what a very long process writing can be even for those, like me, who are not alarmed by doing it. I see many people put themselves under pressure by leaving writing until later. I advocate fairly constant writing. A final text can then draw on this material. Be warned, however: it should be commenced much earlier than you intend to.
How to write 1000 words a day (and not go bat shit crazy)
Another way to write 1000 words a day?