Building up Stats tolerance as a Stats terrified

or How I got to dislike stats a bit less

I’m not sure why, but I’ve never been in a healthy relationship with stats.

How it started-

I was first introduced to the artistically blurred world of statistics in maths class in high school. I thought everything would go smoothly. Afterall, I was very good at maths, and this was just another maths lesson, so I would obviously understand everything easily, right? No, no I very much would not. I just couldn’t get it, to the point that 2 years later, at the baccalaureate, I got 15/20 in maths, because I simply could not do the statistics exercise worth 5 points. It was an MCQ. I was so bad at stats I could not even pick a random answer in an MCQ, even for something as important as the baccalaureate. See, I told you: unhealthy relationship with statistics.

Needless to say, things did not get better at University. Luckily for me, in my Biology undergrad, the stats requirements were gloriously low. Unluckily for me, in Masters I had to confront ANOVAs, which was deeply unsettling. This is pretty much how I started the PhD: the simple though of statistics would give me deep anxiety to the point of feeling sick, and all I could vaguely understand were T tests, Chi Squares, and 2x2 ANOVAs. Yet, somehow, I decided to research a question that would most definitely require something a bit more complicated than that.

If you are finding yourself in the same situation and you have no idea how on Earth you are going to handle this, I’m telling you: you are not alone, and you can get through this! Here is how I managed to come to terms with statistics.

Disclaimer-

You don’t have to like statistics to be a good researcher, you don’t have to know all the statistics out there to be a good researcher, you don’t have to use fancy statistics to be a good researcher, and actually you don’t even have to use statistics to be a good researcher. All that matters is that you understand (but not necessarily enjoy) the methods and tools (including some statistics for quantitative research) you need to do your research, answer your questions, and understand the research of your peers. That’s it.)

1. Review available support and resources

You are absolutely not alone in this situation, which means that there is a lot of resources in place at your University to support you. The challenge might just be to find them all in the first place. For example, I discovered the existence of the Stats Clinic and the 1:1 Stats Consultancy of my college at the end of my 2nd year (#nevertoolate). It is likely you will also be allowed to audit undergraduate stats courses if you get in touch with the course organiser. These are great to start afresh without the pressure of exams! Of course, there are also a lot of resources online, but as a random google search can feel overwhelming, you can turn to free online courses as well. Finally, there will definitely be people around you, in your lab, or friends of friends, who are better than you at stats, and you will able to help you (especially if you bribe them with cake or wine).

2. Don’t be embarrassed to ask for help

Asking these people around you for help can be embarrassing. Personally, I even felt ashamed when I went to my first Stats Clinic. I felt like I should know all these things, and it was a failure on my part not to know stats (thoughts I would have never had about somebody else, of course). Of course, this mindset is useless, if not counter-productive. Never be ashamed of not knowing things and needing help to learn them. Nobody can know everything, and you obviously know a lot of things the person if front of you does not! Now, in the pursuit of great science, you are simply going to combine your skills. This is pretty much what science is about! In reality, the person who is going to help you with your stats will probably be very intrigued about your research and will have a lot of questions. When you ask for stats help you can end up having really cool conversations about your research, because your statistical analysis is actually intertwined with the meaning of your work.

3. Shift your mindset

Near the end of my PhD, I had an epiphany. All of a sudden, I started to see stats not as a hellish, dreadful, inevitable step of research but as a tool to tell the story of my work. One thing that always bothered me with stats is that there were so many kinds of tests that could be used to answer a question. But actually, when you start thinking about your research as one fluent story that moves from one clue to the next, suddenly things start to look clearer. Sure, you could use that test to answer 3 of your questions at once. But in terms of story, does it make sense to answer all these questions at once? Maybe not. This decision step already narrows down the number of potential tests. This shift in mindset, seeing stats as a story-telling tool, helped me feel less anxious about stats. It did not make me like them (I’m afraid this will never happen), but at least now I am able to think about stats without feeling physically sick, and maybe even calm and serene.

I’m not saying it will all be rainbows and unicorns, but you can get there. It’s a learning journey, and every single step counts. You will reach your goal!

Best,

Bérengère