Luckily, the first day of 2021 is a Friday, so my last action of 2020 is to write this blogpost – at least if you don’t account for the dinner and the wine. Last week, I did not publish a post, because I think that, albeit these should come every week, there should also be exemptions. After all, 2020 was a very exhausting year, and I feel the blog deserved a short break.
However, now that 2021 is starting, it’s time to ramp up my studies. This year will be (more or less) concluded by my so-called “20-percent test,” which consists of me presenting a detailed research plan for the three consecutive years. The overall goal of my PhD is to publish at least three peer-reviewed papers in full (co-)authorship, and my initial research proposal, while interesting, can surely use some more fleshing out over the year to arrive at a plan that includes those three papers.
After two months of working at LiU, I have a much better picture of what my PhD will consist of, as you surely have realised by now. The next aim, which I’m not sure I already mentioned here, is to get an overview over all or at least most of the methods used by social scientists in order to arrive at conclusions. This means that I will spend the next two months sifting through all my collected papers, and extract their methods, their underlying theories, their assumptions, their datasets and what kind of causation they attempt to unearth (if they do, that is).
As you can imagine, that’s certainly not the most exciting work, but it’s a worthwhile endeavour, and after I have all of these methods collected, some first impressions will form of what and, more interestingly, how statisticians use mathematical models.
After having finished the first papers during Christmas, I already got the impression that social scientific statistics seem to have a similar relationship to maths as engineering. Mathematicians think of great models to describe relationships between data, and the main goal of statisticians is to find computational shortcuts and apply rules of thumb where we don’t need those extra position after the period. This is something engineers do a lot themselves. When you look at algorithms implemented in various programming languages, you can see that engineering feats are related to clever computational shortcuts that arrive at the same or at least similar results. Something similar – at least that’s my impression – happens when statisticians use mathematical models to analyse data.
As you can see: There’s a lot of interesting stuff to excavate, and I’m sure I can give you some interesting fun facts in two months time, before I start working on the new research proposal itself. But now, rest assured that 2021 will really start only tomorrow, when the hangover is gone. You know the drill ;-)