A month and a half ago I started writing my dissertation. Data collection started about 10 months ago, and I was still deep in data analysis in January when I realized I was going to have to start writing as I was analyzing data in order to get this thing done in time. A draft will go to my chair on Friday! (Whoa.) What is a dissertation anyway? Google tells me it's "a long essay on a particular subject, especially one written as a requirement for the Doctor of Philosophy degree." A long essay is an understatement. My dissertation as it stands is 250 pages long with 30 tables of data and 20 figures representing theoretical models and emergent relationships in my data on the practices of science blogging.
How do you write a 250 page document chock-full of data tables, figures and references without your brain turning to mush? Good question. Here are my tips for writing a dissertation without going nuts, pulling your hair out, or throwing your computer over a balcony in a fit of rage.
Lay it out
I HATE revising and formatting manuscript drafts. Before I ever started writing my dissertation, I located my graduate school's formatting guidelines and started a new Word document complete with title page, acknowledgement section, table of contents section, chapter sections, reference and appendix sections. I inserted Word document page-breaks between each new section and plugged in the proper formatting and font. This way, I won't have to go back to format my document when I get ready to submit it to the graduate school. I also feel better typing into a Word document that already looks like what a dissertation document or academic paper (APA style for me) should look like - it just makes me feel less "cluttered" inside and gets me in technical writing mode.
Also, lay out an outline of what your dissertation should look like. Is each chapter going to be a separate "study" complete with methods and results, or are your chapters going to be the classic introduction, literature review, methods, results and discussion/conclusion sections? Figure this out early, so you don't have to spend time at the end moving everything around and filling in the gaps. An unorganized layout can lead to great stress down the road when you have 150 pages to juggle around.
Now just start writing
Still analyzing data? Just start writing your literature review or methods section. As soon as you have a solid idea in your head, whether it's an idea about your theoretical approach or about how the results are falling into place within your theoretical framework, write it down! And your introduction and methods sections should be easy to write even if you don't have a solid idea of how your dissertation results flow yet. At some point, you just have to start writing. It won't be perfect, and you may have to rearrange your sections later, but that's OK.
Get into a routine
Seriously, this sounds trite and silly, but it's far from it. Think about how long you have before a draft is due and what you need to do to get there. Try to figure out how much you need to accomplish on a daily basis to meet the deadline. When I started writing a little over a month ago, I was still coding the 50 in-depth interviews I had conducted with science bloggers for a qualitative research component of my dissertation. To code every interview at least twice over, I figured out at the time I needed to get through at least two interviews a day to finish coding with a couple weeks to spare for writing up the results. So no matter what happened, no matter if I had to stay up until 3am on a busy day to get those two interviews done, that's what I did day after day. I also set a goal at the time of ~5 pages a day of dissertation writing. I knew that if I could write ~5 pages of my dissertation manuscript, along with coding and making notes on two interviews, every single day from mid-January to early-March, I could get this thing done.
Take some time to THINK
It often became hard to pull myself away from the computer for even a few minutes at a time to grab coffee with my husband during 4-6 hour stents (if not longer) of analyzing data and writing. But I often found that if I forced myself to pull away, take a walk or a ride in the car, I could think about my findings from an "aerial view" so to speak. On one of these coffee runs in particular, I had an insight on how different themes coming out of my interview analysis fit together in a larger over-arching theme which I decided to call The Ecosystem Approach to science blogging. It's a neat pattern in my data where science bloggers often blend the practices of monitoring other science media streams, adding something personal to the topics or stories they choose to blog about, developing their own content niches and avoiding the niches of other bloggers and journalists. But it took stepping away from the computer and mulling over the findings from a broader perspective for the overarching patterns to emerge.
Don't write your dissertation into fits of disparate findings and arguments. Step away from the computer and think a bit before you throw every piece of data and non-data into your dissertation manuscript.
Work with it - Stay up late if need be
If you are anything like me, writing 10-20 pages at a time is magically easier at midnight than it is at noon. So I embraced the times that it felt easier to write instead of fighting the writing into a perfect schedule, staying up later than I usually would to catch up on writing.
Have a physical outlet
Don't give up on physical activities during dissertation writing. In the last month and a half, I moved up to a more advanced ballet class and started going twice a week, in the mornings, to have an activity that got my body moving and took my mind off the dissertation. Not only do the ballet classes get my morning going, but they keep me from feeling like a blob anchored to my desk. On that note, also get a comfortable office chair for home if you are writing your dissertation from there - and a stand-up desk or podium (I got something like this or this).
That's right. Allow yourself some procrastination - just not too much. Seriously - it takes me 30 minutes to an hour to even get "warmed-up" to start dissertation writing every time I sit down for a new few-hour stretch. I have a procrastination routine, in fact. It goes: check e-mail, check Twitter, check SciLogs.com, check Facebook. Repeat. Read any interesting blog posts that come up on Twitter or SciLogs.com, and tweet about them. Check e-mail again - respond to e-mails if need be. Check Twitter again. Realize I sat down to write an hour ago. Start up Ratatat Pandora station, and get to work. [... 1-2 hours later ...] oh new Twitter interactions! Wait, stay on task...
Share your progress. I've really enjoyed tweeting snippets of my interviews with science bloggers (#MySciBlog), blogging about research insights and uploading new data figures to Figshare as I work on my dissertation. I found that these small acts of sharing my research findings and insights with others online not only allowed for interesting and helpful feedback from other science bloggers and science communication scholars, but also often provided the boosts I needed to keep pushing my more technical dissertation writing forward. What better incentive to keep working than others commenting on how interesting your dissertation concepts or findings are!
Use a reference manager!
For the sake of preserving your sanity, use a reference manager. Something like Endnote or another tool that helps you automatically keep track of your references, and change the citation style in 1-click if need be. Also use Word's automated section headings and figure/table captioning system, so that Word can automatically build your Table of Contents, List of Tables and List of Figures. I shudder to think of having to do this all manually. Auto-captioning your tables and figures and using Word's cross-reference inserts (References --> Cross-reference) is a must-do. Otherwise you have to keep track of changing figure and table numbers as you write.
Persevere and have fun!
Just keep at it, never give up, and enjoy the little victories. Did you write 10 pages today, or finish a particularly chewy results section? Celebrate with an episode of your favorite Netflix show! Take a run and then eat some cake! Celebrate the small successes, and then push for the next section.
All you really have to do is outlast your doctoral committee. The good news is that they are much older than you, so you can guess who will eventually expire first. The bad news is that they are more practiced at this game. - How to write a dissertation