Why You Should Care More About Statistics

In this, the first in a series of articles on statistics, I want to set out some of the main reasons why you, as a biologist, should improve your knowledge of statistics.

The general consensus is that biologists are not strong when it comes to statistics. There’s nothing in our brains that stops us from being good at it, but some effort needs to be put in. I believe every biologist should have a handle on basic statistics. Here’s why…

 1. Statistical knowledge will improve your experimental design

When you truly get your head around hypothesis testing, degrees of freedom and pseudoreplication, you will find it far simpler to design efficient experiments. That means you save yourself time; you’ll have neater methods sections in your thesis or articles; you can confidently defend your methods to thesis examiners, reviewers or conference attendees. If you know why you did what you did, it makes many of the downstream steps so much easier.

On the flip side, biologists who have a vanishing knowledge of statistics can waste a lot of effort on irrelevant experiments and insignificant data. Moving out of the darkness and into the light of statistical understanding will truly change your abilities as a researcher.

2. You will be able to analyse your own data

Or perhaps that should be, analyse your own data without tearing your hair out. Analyses should be fun; you’ve spent weeks, months, years planning and executing an experiment. You’ve got RSI from the pipette, or you see spots from staring down a microscope. You may have had a hint that something was going on, but now you get to prove it. Good results = good mojo. You should get to enjoy that moment where you see the first significant p-value.

3. You’ll get your next job more easily

In job hunting, you want to set yourself apart from the pack. If you nodded and agreed when I said that statistics doesn’t come naturally to biologists, then you probably have a colleague who supports you in your statistical analyses. You love that person, buy them coffee, you’d probably give them a foot massage if they’d help you with a nested ANOVA. Upskill now and YOU can be the one getting the massages. Then when it comes time for the next job interview, you can confidently tell your prospective employers that you’re the one people go to for statistical advice. See it as the valuable skill it is, and learn it in the way you learn new software and laboratory techniques.

4. You will have the power to critique papers with greater authority

This is huge. All of us, when we were “growing up” as scientists, were taught the importance of peer review. Scientific integrity is built upon healthy scepticism; the burden of proof lying on the side of the scientist. As undergraduates we were lectured on how to read the methods and results sections of a journal article.

If you are one of those biologists who detest statistics, how thoroughly do you actually read methods and results sections? Would you notice if someone had pseudoreplicated their experimental design? Or if they had incorrectly paired samples for a t test? Do you even know what I’m talking about?

We all know how to spot conclusive leaps in articles, we’ve learnt to spin our own work for grant proposals and conference abstracts, so we also know how to see when others are doing the same. As biologists, we also need to be able to critique a researcher’s experimental design, analysis and the conclusions they draw from that analysis.

I hope you are now convinced that you need to start learning more about statistics. The good news is that I am here to help. Look out for my next article in which you will begin the journey. Welcome aboard!

In the meantime, please feel free to share your thoughts, fears and views on statistics in the comments section.