Last semester, I defended my Master's Thesis. While writing it, I tallied the daily hours I put toward the project in a Google Document. I'll get to the data analysis momentarily, but first, a little background is in order.
I began serious work on the MA Thesis in Spring 2013 (though I had toyed around with ideas in the latter half of 2012). Initially, I approached the project in the way that seems most common: I tried to work on the thesis for a short amount of time nearly every day. This approach to getting things done has been discussed here before, although my familiarity with that strategy originates with Stephen King's endorsement of that strategy in his On Writing. This strategy did not work well for me; I found it very hard to make meaningful progress in short intervals and was frequently discouraged by my inability to meet daily goals. I tried setting goals based on word count (1000 words per day), page count (3 pages per day), and even just time (2 hours per day), but these changes didn't make a detectable change to my output or my ability to meet the specified goals. While I made some progress on the thesis that spring, it was neither the quantity nor quality that I had hoped for: much of it got scrapped over the summer.
When I resumed work on the thesis in August, I adopted a new strategy: a weekly hour count with no daily requirements and no specific page- or word-related goals. Though I adjusted my weekly target on a few occassions (e.g., when the thesis was under review by my committee), I generally aimed for 12 hours per week. I saw a drastic improvement in results compared to the previous semester. My thesis turned out to be about 35,000 words (including footnotes and references); almost 30,000 of them were written in an 11-week stretch. (It may also be worth noting that I was taking two graduate seminars during this timeframe: writing a thesis in my PhD program is optional, and I did not want to slow down my time to degree.) While there are probably several factors that explain why the writing process went better in the Fall than the Spring, these details suggest that one of them was the change in how I approached the writing process.
But the hour tally reveals more than just this general observation. You can view the tally at this link. Here are some of the details that stood out to me:
- I routinely fell short of my weekly target and fell 22.5 hours short of my total target. I took a few weeks off from thesis work during the semester, but there were 13 weeks with definitive targets. I only met those targets 6 times. What's interesting is that I still got the project done in the desired timeframe. Perhaps 12 hours per week was an unnecessarily ambitious target.
- Of the 114 days that I charted, there were 65 days where I did NO work on the thesis. That's right: a 0 shows up on the grid 65 times. When you crunch the numbers, that means that on average, I only worked on the thesis 3 days a week during this stretch. I suspect this is the most important piece of data because it suggests that I do not work best by trying to chip away at a project on a daily basis. Rather, it seems I needed large blocks of time (evidently of 3-7 hours) during which I could fully engage with the thesis.
- I did very little thesis work on Mondays and Tuesdays but worked a ton on weekends. I only worked on the thesis for 15.75 hours (12.5% of total hours) on Mondays and Tuesdays, but I worked on it for 62.75 hours (50% of total hours) on weekends. My suspicion is that this is partly explained by point #2, since large blocks of uninterrupted time are more likely to crop up on weekends. But here's the interesting part: my schedule on Mondays and Tuesdays was actually much lighter than it was on the other three weekdays. I held office hours during the middle of the week, Thursday featured a 3-hour night seminar, and Fridays are the usual day that the department hosts guest speakers or holds other events. So why were my Mondays and Tuesdays so unproductive? There are a number of possible explanations, but my suspicion is that these days functioned as recovery days. After working on the thesis for 7-10 hours on the weekend (which happened several times), perhaps I just needed a couple days away from it.
I was deeply surprised at these results, since I've been told so often that working daily in small increments is the recipe for writing success. But it's hard to ignore the data.
After I defended the thesis, I shared these observations with two of my committee members. One of them expressed a need to follow a similar approach when writing his own original work; he found that the daily, bit-by-bit approach worked well for him only when the work he was doing did not involve a lot of his own original input (e.g., when he was editing an anthology). Another committee member said that he had to chip away at his writing on a daily basis -- no matter what the particular project was -- and that his attempts at trying other approaches in the past had all failed.
What's the moral of this story? It's hard to infer a ton from one person's examination of a single project, but I suspect there may not be any general one-size-fits-all approach to writing philosophy. Perhaps the only way for us to learn what writing strategy works best for us is to experiment and monitor our progress. I'll certainly be recording my progress again when I get to the dissertation, and I encourage others to give it a try. What you learn about your work habits might surprise you.