Friday, August 8, 2014

A Quick Paper

For the TRB Conference next year, I wanted to submit something 'quick and dirty' for a presentation to get some feedback for my current research. The problem with research is that any literature review is most probably incomplete, for there is almost always going to be some research that somebody, somewhere forgot to or could not read. That's where conferences come in really handy - circulate your research around, take feedback and see what you missed. If you're lucky, you might just get some useful feedback that can help you move forward.

Now, when I said 'quick and dirty,' I did not mean a badly written paper. If I knew a paper is badly written, I would never submit it. Although this was meant for presentation only, it was a great learning experience for me because, as I now see, the first draft was really pathetic. My adviser was quite nice not to say it so bluntly, but I now see that I made some critical mistakes in the first draft. After getting it read by two of my colleagues, including one senior PhD student, I managed to figure out where I was going wrong:

  1. Make it as simple as possible, but no simpler. There's no need to write every single thing in a paper, but don't make it so concise that the reviewer misses the point. Instead, get to your point directly and explain it well, and leave it to the intelligence of the reviewer to figure out the little things around it. 
  2. A picture should be worth a thousand words. It's actually a blessing that TRB counts every figure as 250 words, because you can put in a massive amount of information into each figure. Therefore, it is important to design figures well. In fact, as my colleague pointed out, a figure should be able to stand on its own, so that the reviewer understands everything about it without having to refer back to the text. The caption is the lone text that is an integral part of a figure. 
  3. Make it obvious. If you're pointing out a feature on a figure or table, make sure you it is already highlighted there so that when a reviewer sees the figure/table, they instantly find what you're talking about. If you expect the reviewer to find it for you, you're making a mistake. 
  4. Look back. Use the past tense in your discussion. You used this methodology, you concluded that. This can be quite confusing because a lot of non-US authors tend to use the present tense, and indeed some well-known facts should be ("The thermal conductivity of a material is temperature-dependent"), but most of the paper would actually be in the past tense. 
  5. Look before you leap. Before stating anything, consider the implications to future study. Don't make sweeping generalizations ("A higher thermal mass is an effective solution...") but rather, stick to the findings of your work ("The results indicated that, under the assumed conditions, a higher thermal mass could be a potentially effective solution..."). Of course, that does make it quite a mouthful, but better safe than sorry! 
I actually made each and every one of these mistakes in my first draft and the process of correcting them led to me compiling the above list. It was painful to realize just how much I have to learn when it comes to writing papers, but overall it was an enriching experience. If a PhD is supposed to teach you how to do research, I'm well on track! 

No comments: