We all have our demons, and one of mine is stats. Stat’s is not my strong point and I was criticised for the lack of statitical analysis included in my probation report at the end of my first year. I actually think this is quite a good thing, I wasn’t criticised for any of the fundamental ideas or methods of my project, the actual work I’m doing is good, I just need to add the stats. While I don’t enjoy stats I think as far as criticisms go this one is fairly fixable. So, with self improvement in mind, when the latest list of available training courses was issued by the college of Medicine, Biological Sciences and Psychology I signed up for a basic statistics course.
On Monday I went off to a three hour session in the Attenborough Tower (which is far too tall for my liking) and came back with a very useful powerpoint and a much better understanding of when to apply the different statistical tests. The downside is that I didn’t realise that this was a 2-part course. Session 2 was perfectly do-able without session 1, but I was a bit confused when the first exercise we did was exercise 8! I wasn’t the only person who only ended up at this session (we assumed the same course was running more than once due to high demand) but the very helpful tutor was happy to email us the powerpoint presentation from the first session.
Overall this was a really useful session, I’m now much more confident I can choose and apply the right statistical tests to prove the significance of my results. It’s important to remember that throughout your PhD you will receive a lot of criticism, it’s simply a fact of working in academia, and instead of taking it as a personal attack you should use it as a guide to improving your skills. The colleges offer a lot of training courses and while you don’t need to attend all of them there will definitely be some you might find useful. Early on I did a lot of training courses, some of which weren’t that relevant, I’m now in a position where I know my strengths and weaknesses and can choose training accordingly.