Sorry for the low posting, I've been a bit preoccupied with work, marking exams, creating lesson plans, playing with a new data set, preparing for next year's Green Soc and the like.
A word about the exams: I marked 60 and my fellow PhD student/officemate Iain marked 60. Without consulting on any specific answers, our final mean marks were within 3% of each other. Now that's statistical reliability, baby!
A word about the new data set: Getting new data to prepare for analysis is a little like getting a Christmas present with about 50 layers of packaging, all of varying difficulty to remove. Imagine receiving a package that takes 12 hours to unwrap, knowing the contents will determine the path of your next 3 years of work (and could possibly help cement the PhD or potential career), and if you try and unwrap it too quickly you could break it or worse alter its meaning without even realizing it, thus potentially leading months or years of work down a deadend. That's what it is like cleaning and integrating data in preparation for analysis.
Apologies if you have already seen this video floating around the intarnets, but if you haven't it will brighten your day. It is 5 seconds long, and you should watch it (volume up) whenever you read or hear something mildly surprising. It will never get old.